w“ E 7 F 7 Dr"! mm mamas V 7 LIBRARY : 5’ The Borrego Mountaih Earthquake of April 9, 1968 GEOLOGICAL SURVEY PROFESSIONAL PAPER 787 DOCU MENTS DEPARTMENT FEB 14 1973 LIBRARY UN I VE RR “ML- !" OLCALEW Tnfiiéfim BORREGO MOUNTAIN EARTHQUAKE, APRIL 9, 1968 Oblique northward View of the Salton Trough in southern California and northwestern Mexico. Names of prominent indicate segments of faults that moved at the time of the 1968 Borrego Mountain earthquake. The distance 2.“on DRR\E\Q QQ‘LQ R b® Q “\V ER BKVX k as” 1' //f” physiographic elements within and bordering the trough are shown in the adjoining line sketch. Lines within shaded areas between meridians shown is 93.5 km along the 33d parallel. Apollo 9 photo AS9—23—3558, courtesy of NASA. The Borrego Mountain Earthquake of April 9, 1968 GEOLOGICAL SURVEY PROFESSIONAL PAPER 787 Contributions from: California Institute of Technology Lamont~Doherty Geological Observatory of Columbia University Seismological Field Survey, National Oceanic and Atmospheric Administration UNITED STATES GOVERNMENT PRINTING OFFICE, WASHINGTON : 1972 UNITED STATES DEPARTMENT OF THE INTERIOR ROGERS C. B. MORTON, Secretary GEOLOGICAL SURVEY V. E. McKelvey, Director Library of Congress catalog-card No. 72-60020] For sale by the Superintendent of Documents, US. Government Printing Office Washington, DC. 20402 - Price $6.25 Stock Number 2401-00218 CONTENTS Page Introduction, by Robert V. Sharp, U.S. Geological Survey ................................................................................................................. 1 Tectonic setting of the Salton Trough, by Robert V. Sharp, U.S. Geological Survey .................................................................. 3 Foreshock, main shock, and larger aftershocks of the Borrego Mountain earthquake, by Clarence R. Allen and John M. Nordquist, Seismological Laboratory, California Institute of Technology .............................................................................. 16 Source parameters of the Borrego Mouuntain earthquake, by Max Wyss, Lamont-Doherty Geological Observatory of Co-' lumbia University, and Thomas C. Hanks, Seismological Laboratory, California Institute of Technology ..................... 24 Aftershocks of the Borrego Mountain earthquake from April 12 to June 12, 1968, by Robert M. Hamilton, U.S. Geologi- cal Survey .............................................................................................................................................................................................. 31 Surface rupture along the Coyote Creek fault, by Malcolm M. Clark, U.S. Geological Survey ............................................. 55 Displacements on the Imperial, Superstition Hills, and San Andreas faults triggered by the Borrego Mountain earthquake, by Clarence R. Allen, Seismological Laboratory, California Institute of Technology, Max Wyss, Lamont-Doherty Geological Observatory of Columbia University, James N. Brune, Institute of Geophysics and Planetary Physics, University of California, San Diego, and Arthur Grantz and Robert E. Wallace, U.S. Geological Survey ................. 87 Continued slip on the Coyote Creek fault after the Borrego Mountain earthquake, by R. O. Burford, U.S. Geological Survey ..................................................................................................................................................................................................... 105 Holocene activity of the Coyote Creek fault as recorded in sediments of Lake Cahuilla, by Malcolm M. Clark, Arthur Grantz, and Meyer Rubin, U.S. Geological Survey ....................................................................................................................... 112 Geologic evidence of previous faulting near the 1968 rupture on the Coyote Creek fault, by Robert V. Sharp and Mal- colm M. Clark, U.S. Geological Survey ........................................................................................................................................... 131 Intensity distribution and field effects, strong-motion seismograph records, and response spectra, by Seismological Field Survey, National Oceanic and Atmospheric Administration ......................................................................................................... 141 Engineering geology, by R. 0. Castle and T. L. Youd, U.S. Geological Survey .................................................................. 158 Intensity of shaking estimated from displaced stones, by Malcolm M. Clark, U.S. Geological Survey... 175 Water-resources effects, by A. O. Waananen and W. R. Moyle, Jr., U.S. Geological Survey ..................................................... 183 Collapse fissures along the Coyote Creek fault, by Malcolm M. Clark, U.S. Geological Survey .................................................... 190 ILLUSTRATIONS [Plates are in pocket] FRONTISPIECE. Oblique northward View of Salton Trough in southern California and northwestern Mexico, showing physio- graphic elements. PLATE 1. Map showing surface ruptures created at the time of and after the Borrego Mountain earthquake of April 9, 1968 (G.m.t.). 2. Map showing breaks along the Superstition Hills and San Andreas faults formed by displacement trig- gered by the Borrego Mountain earthquake of April 9, 1968. 3. Generalized geologic map of the area near the segment of the Coyote Creek fault that ruptured during the Borrego Mountain earthquake. 4. Map showing generalized geology and distribution of effects of the Borrego Mountain earthquake between Coronado and Glamis, Calif. 5. Map showing distribution of effects of the Borrego Mountain earthquake within the epicentral area. Page FIGURE 1. Generalized geologic map of the Salton Trough, southern California ......................................................... 4 2. Composite stratigraphic columns along the flanks of the northern part of the Salton Trough .................. 6 3. Oblique aerial view of southern Borrego Badlands and West Butte, Borrego Mountain ............. 9 4- Map ShOWing seismicity of the northern Salton Trough region ....................................................................... 12 5. Map showing epicenter of the 1968 Borrego Mountain earthquake in relation to major faults and epi- centers of other earthquakes of magnitude 6.0 and greater in southern California since 1912 ........... 16 6. Index map showing location of Caltech seismograph stations that were operating at the time of the Borrego Mountain earthquake .......................................................................................................................... 1‘7 VII VIII FIGURE “39° 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24—28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. CONTENTS . Map showing epicenters of shocks of the Borrego Mountain series of magnitude 3.0 and greater that occurred between April 9, 1968, and April 29, 1969 ............................................................................... Sketch Showing focal-mechanism solution for the Borrego Mountain main shock _____________________________________ . Sketch showing epicenters of figure 7 separated into four consecutive time periods containing roughly equal number of shocks ....................................................................................................................................... Sketch showing attenuation-corrected P-wave spectra of the Borrego Mountain earthquake ...................... Sketch showing attenuation-corrected S-wave spectra of the Borrego Mountain earthquake from WWSSN recordings ........................................................................................................................................... Map of the Borrego Mountain region showing location of seismograph stations, epicenter of main shock, aftershock epicenters, and active faults ............................................................................................ Detailed map of aftershock epicenters, ground breakage, epicenter of main shock, and lines of vertical sections ................................................................................................................................................................... Sketch showing variation of epicenter pattern with time ....................................... Vertical sections through aftershock zone .................................................................. Vertical sections through aftershock zone ............................................................................... Graph showing depth distribution of aftershocks with well-determined hypocenters .................................... Sketch showing equal-area projections of the first-motion pattern for nine aftershocks ........................... Map showing nodal-plane solutions, plotted at the epicenter, for 72 aftershocks that yielded well-de- termined solutions ............................................................................................................................................... Graph showing average P-magnitude residual plotted against average X-magnitude residual for each seismograph station ............................................................................................................................................ Graph showing average P-magnitude and average X-magnitude residuals plotted against station-time correction ............................................................................................................................................................... Graph showing rate of earthquake occurrence plotted against time after the main shock ..................... Vertical aerial photograph of the region of surface faulting associated with the Borrego Mountain earthquake ............................................................................................................................................................ Photographs showing: 24. Central break at location 19.7 showing en echelon fractures as much as 20 m long and 2 m apart within a zone about 2 m wide ............................................................................................. 25. Band of complex fractures about 1 m wide across hardened silty sand crust in the wash at the eastern base of Borrego Mountain ............................................................................................ 26. En echelon, parallel, and complex fractures in a zone 6—10 m wide on the north break at lo- cation 6. 4 ............................................................................................................................................... 27. Time sequence on central break at location 22.5 .................................................................................... 28. En echelon fractures at the north end of Benson Lake ....................................................................... Photograph and sketch showing detail of symmetrical S-shaped overthrusts in the clay-silt crust of Benson Lake ......................................................................................................................................................... Photograph showing tire tracks at location 22.5 displaced about 180 mm across the single fracture at this location .................................................................................................................................................... Photograph showing dune buggy crossing the north break in Benson Lake 6 days after the earthquake Graph showing total right-lateral displacement along the north, central, and south breaks measured at different times _____________________________________________________________________________________________________________________________________________________ Photograph showing segment of an extensional fracture showing dip slip ................................................. Photograph showing fractures with left—lateral displacement at location 17.5 northeast of the cen- tral break ............................................................................................................................................................. Map showing the location of recently active traces of the Coyote Creek fault in the vicinity of the surface rupture of 1968 .................................................................................................................................... Photograph showing view southeast along the central break, showing an old scarp 0.3—0.4 m high in alluvium at the northeast base of Ocotillo Badlands ................................................................................... Photograph showing central break at Old Kane Spring Road ........................................................................ Photograph showing view westward over south break in November 1956 .................................................... Map showing areas of postearthquake displacement along the surface rupture after April 14, 1968 Graph showing rainfall, estimates of dates of surface runoff on the central and south breaks, and dates of field investigations, January 1968 through December 1971 .................................................................. Map showing mean annual precipitation in the region of the 1968 break, 1931—60 ..................................... Photographs showing time sequence of collapse and eventual filling along a fracture of the cen- tral break at location 23. 8 ................................................................................................................................ Map showing fractures that apparently formed near the central break between June 21, 1968, and March 20,1969 ..................................................................................................................................................... Photograph showing tire tracks at location 19.4 offset 240 mm horizontally ................................................. Photograph showing new breaks with 30 mm of right-lateral offset .............................................................. Page 17 20 22 26 27 32 45 46 47 48 48 49 50 51 51 52 56 59 60 60 61 62 62 63 63 64 65 65 67 68 70 73 74 75 76 77 78 78 FIGURE CONTENTS 46. Photograph showing detail of the fracture shown in figure 45 ................................................................... 47. Map showing horizontal displacement ................................................................................................................... 48. Graph showing displacement from April 9, 1968, to March 1971 near location 19.4 ................................. 49. Map showing special features of the south break ................................................................................................. 50. Photograph showing sharp-edged fracture that formed after the channel that it crosses carried water in September 1969 ............................................................................................................................................... 51. Maps showing location of aftershocks and other earthquakes of M23 and areas of creep during three periods after April 8, 1968 ................................................................................................................................ 52. Index map showing relation of the Coyote Creek fault to the three distant faults, the San Andreas, Superstition Hills, and Imperial, upon which triggered movements occurred ....................................... 53. Photograph showing fresh cracks in unconsolidated material at north edge of Highway 80 ................... 54. Map of Imperial fault trace ...................................................................................................................................... 55. Graph showing measurements of fault—crossing geodetic networks between May 1, 1967, and May 1, 1970, showing displacements at time of Borrego Mountain earthquake ................................................. 56—69. Photographs showing: 56. Fresh en echelon cracks crossing Ross Road .......................................................................................... 57. En echelon cracks on Meloland Road first observed following the Imperial earthquake of 1966 but slightly widened during the Borrego Mountain earthquake... 58. Fresh cracks across Heber Road ............................................................................................................... 59. Right-lateral displacement of about 2 cm on the Superstition Hills fault in sec. 23, R. 12 E., T. 14 S ................................................................................................................................................... 60. En echelon cracks indicative of right-lateral displacement along Superstition Hills fault in sec. 9, R. 12 E., T. 14 S ............................................................................................................................... 61. New en echelon cracks of right-lateral habit following the line of desert plants that marks the Superstition Hills fault in NW%. sec. 26, T. 13 S., R. 11 E ......................................................... 62. Small collapse pits, alined vegetation, and variation in abundance of vegetation along the Superstition Hills fault in SW14 sec. 9, T. 14 S., R. 12 E ........................................................... 63. En echelon cracks at base of scarplet along San Andreas fault in sec. 28, R. 9 E., T. 6 S ....... 64. New fractures in old fracture zone along San Andreas fault in sec. 28, R. 9 E., T. 6 S ................ 65. En echelon cracks along the San Andreas fault near Salt Creek, in center of sec. 28, T. 8 S., R. 11 E ................................................................................................................................................... 66. En echelon fractures along the San Andreas fault approximately 2 km south of Salt Creek, in SW14 sec. 34, T. 8 S., R. 11 E ................................................................................................. 67. En echelon fractures, sec. 28, T. 6 S., R. 9 E ........................................... 68. Linear valley eroded along San Andreas fault, Mecca Hills... ............ 69. View southeast along the San Andreas fault ......................................................................................... 70. Map of the main surface breaks associated with the Borrego Mountain earthquake of April 1968, show- ing the locations of three alinement arrays established for afterslip investigation ............................. _ 71. Plan View of typical alinement-array configuration ................................................................................. 72. Plan view of alinement changes at the Borrego Mountain array .............................................................. 73. Plan view of incremental alinement changes at the Borrego Mountain array ............................................. 74. Terrain profile along the Borrego Mountain array, and profiles of relative elevation changes between sets of level readings ............................................................................................................................................. 75. Plan view of alinement changes at Lower Borrego Valley array .................................................................... 76. Graph showing cumulative right-lateral slip at the Lower Borrego Valley alinement array plotted with respect to time elapsed since main shock ....................................................................................................... 77. Graph showing cumulative right-lateral afterslip at the Lower Borrego Valley alinement array plotted with respect to logarithmic time scale ............................................................................................................ 78. Terrain profile along the Lower Borrego Valley alinement array, and profiles of relative elevation changes between sets of level readings ............................................................................................................ 79. Plan View of alinement changes at Fish Creek Mountains array from December 10, 1969, to January 25, 1971 ................................................................................................................................................................. 80. Photographs showing central and branching breaks of the 1968 earthquake at Old Kane Spring Road, showing shallow graben between the breaks and the location of trenches 1, 2, and 4 and profiles 1 and 2 .................................................................................................................................................................. 81. Index map showing the three principal breaks of the 1968 rupture of the Coyote Creek fault ............... 82. Structural profiles across the prominent scarp along the central break of 1968 ......................................... 83. Sketch showing south wall of trench 1 showing progressively greater offset of older strata and bend- ing of strata (drag) across the branching break of the 1968 event ......................................................... 84. Graphs showing relation of vertical offset to 01‘1 age along a branching break of the 1968 earthquake, determined from late Holocene sediments expOSed in trench 1 ................................................................. IX Page 79 80 81 82 82 83 88 88 90 91 92 92 93 93 94 94 95 96 96 97 97 97 99 100 106 107 108 108 109 109 109 110 110 114 116 116 118 120 X FIGURE 85. 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. 106. 107. 108—114 115. 116—119 120 121—130 131 132 CONTENTS Graph showing deposition rates of sediments southwest of the branching break of 1968 in trenches 1 and 4 ................................................................................................................................................................ Map showing historic earthquakes of magnitude 6 and greater along the San J acinto fault zone ......... Graph showing recurrence intervals at a point in the San J acinto fault zone for different values of rupture length ..................................................................................................................................................... Sketch showing pre-1968 fractures and fissure collapse along the Coyote Creek fault as exposed in a natural cut at location 30.7 ............................................................................................................................... Sketch showing south wall of trench 4, showing a surface fracture associated with the 1968 earth- quake that developed where there had never been a break before ......................................................... Regional map of the Coyote Creek fault, showing its relation to faults outside the 1968 rupture zone Simplified geologic map of the area at the north end of 1968 Coyote Creek fault ruptures ..................... Geologic map of the area near Squaw Peak ................................................................................... Geologic map of the Ocotillo Badlands ...................................................................................................... Three-dimensional view of possible configuration for Coyote Creek fault beneath Ocotillo Badlands ..... Isoseismal map of the Borrego Mountain earthquake .......................................................................................... Map showing location of strong—motion seismographs that were triggered by the Borrego Mountain earthquake .............................................................................................................................................................. Map showing location of strong-motion seismographs in the Los Angeles area that were triggered by the Borrego Mountain earthquake ............................................................................................................ El Centro accelerograph record from the Borrego Mountain earthquake .......................................... San Onofre and San Diego accelerograph records from the Borrego Mountain earthquake ...................... Graph showing maximum acceleration plotted against epicentral distance for the Borrego Mountain earthquake ............................................................................................................................................................. Accelerograph record from the Central Engineering building at the Jet Propulsion Laboratory in Pasadena, ninth floor ......................................................................................................................................... Map showing location of the ground-level displacement meters actuated by the Borrego Mountain earthquake ...................................................................................................................................................... i ....... Graph showing maximum displacement plotted against epicentral distance for the Borrego Mountain earthquake ............................................................................................................................................................. El Centro seismoscope records superimposed on a map of the city ............................ Seismoscope records from San Fernando, San Gabriel, and Puddingstone dams ........................................ Response spectra from El Centro ............................................................................................................................ Response spectra from San Diego and San Onofre .............................................................................................. . Photographs showing: 108. Rockfalls and soil falls ............................................................................................................................. 109. Soil falls along Fish Creek Wash ........................................................................................................... 110. Incipient slump north of Ocotillo Badlands .......................................................................................... 111. Terrain in central Ocotillo Badlands .......................... 112. Slump scarps generated in the epicentral area .......................................... 113. Headwall scarp of slump or block glide at Highway 80 crossing of New River... 114. Slump in artificial fill ............................................................................................................................... Sketch map of seismic—compaction ruptures in epicentral area ...................................................................... . Photographs showing: 116. Seismic-compaction ruptures in epicentral area ................................................................................. 117. Scarp along edge of compaction-produced graben ............................................................................... 118. Sand boils along the New River ............................................................................................................. 119. Shattered crusts overlying poorly indurated materials in epicentral area ................................... . Map of shattered brittle crust east of Ocotillo Wells .......................................................................................... . Photographs showing: 121. Dislodged or displaced cobbles east of Ocotillo Wells ...................................................................... 122. Damaged chimney in Ocotillo Wells ....................................................................................................... 123. Partially collapsed brick wall in Westmorland .................................................................................... 124. Collapsed fieldstone cellar east of Ocotillo Wells. 125. Cracked concrete-block wall in Salton City .......................................................................................... 126. Damage to plaster and stucco walls of a two-story motel in Salton City ------------------------- 127. Cracked and spalled concrete bridge pier east of Ocotillo Wells .............................. 128. Cracked concrete railroad bridge abutment near Glamis ...... 129. Light truck that sank into sand southeast of Ocotillo Wells ........... 130. Projected and displaced light objects southeast of Ocotillo Wells .................................................... - Map showing area of moved surface gravels _________________________________________________________________________________________________________ . Photograph showing gravel on top 0f ridge A ....................................................................................................... 123 129 132 134 144 145 147 148 149 150 151 152 153 155 155 156 162 163 163 166 166 167 167 168 168 169 169 169 169 170 171 171 172 172 176 176 FIGURE TABLE CONTENTS 133. Stereo view of a 0.6—m-high tetrahedron tipped over by shaking ..................................................................... 134. Photograph showing boulder propelled from its nearly horizontal seat and rotated 180° ........................... 135. Sketch showing forces acting between a block and the inclined surface on which it rests, as a result of friction, adhesion, and shaking of the surface ...................................................................................... 136. Graph showing relation between coefficient of friction and minimum horizontal acceleration necessary to slide a block that rests on a horizontal plane ........................................................................................... 137. Graph showing relation between coeflicient of friction and the minimum horizontal acceleration neces- sary to slide a block that rests on an inclined plane .................................................................................... 138. Graph showing relation between height and excess acceleration between blocks and the surface on which they rest for different values of cohesive strength .......................................................................... 139. Map of selected observation wells, gaging stations, and compaction-recorder sites that showed effects of the Borrego Mountain earthquake ............................................................................................................ 140. Graph showing fluctuations recorded at observation wells in the Salton Sea basin during and after the Borrego Mountain earthquake .......................................................................................................................... 141. Graph showing fluctuations recorded at observation wells in central and southern California during and after the Borrego Mountain earthquake ............................................................................................... 142. Graph showing fluctuations recorded at gaging stations in response to the Borrego Mountain earth- quake ....................................................................................................................................................................... 143—150. Photographs showing: 143. Collapse fissure on the south break that developed on a tectonic fracture after the rains of September 1969 ................................................................................................................................... 144. Aerial View southwestward of collapse fissures and channels leading into them on south break at location 31.4, January 1970 ......................................................................................................... 145. Detail of channel and collapse fissure shown at a in figure 144. 146. Detail of the collapse fissure at b, figure 144 .................................................................................. 147. Collapse across a silty mound ................................................................................................................. 148. Collapse fissures on the south break ...................................................................... 149. Collapse fissures on the south break, January 1970 ......................................... 150. Detail at 0, figure 144, October 1969 .................................... 151. Sketch showing hypothetical development of a collapse fissure ...................................................................... 152. Map showing relict (preearthquake) collapse fissures near the south break of the 1968 rupture along the Coyote Creek fault ....................................................................................................................................... 153-156. Photographs showing: 153. Aerial view southward over part of the most prominent preearthquake fissure .......................... 154. South end of the relict collapse fissure shown at a in figure 153 ................................................... 155. A taut root across the relict collapse fissure at b in figure 153 ...................................................... 156. Concentric collapse fissures at location 18 ........................................................................................... TABLES 1. Caltech trailer-mounted stations in operation during the Borrego Mountain aftershock period ........... . Earthquakes of magnitude 3.0 and greater in the area from 32°45’ to 33°30’ N. and from 115° to 116°30’ W. from April 9, 1968, through April 28, 1969 ............................................................. '. ........... NJ 3. Borrego Mountain earthquake data for stations of the Pasadena network. 4. P- and S-wave spectral data ...................................................................................... 5. Source parameters .............................................................. 6. Local observations of radiated seismic energies ........................ 7. Seismograph station data ................................................................. 8. Borrego Mountain aftershocks ................................................... 9. Width of the main rupture along the Coyote Creek fault ...... 10. Summary of creep at locations indicated in figure 39 .................................................................. 11. Earthquakes of M>3 in the area from 32° 45’ to 33° 15’ N. and from 115° 45’ to 116° 15’ W. from April 29, 1969, through 1970 ............................................................................................................................ . Caltech fault-crossing geodetic networks in the Imperial Valley area ........................................................... . Vertical deformation along faults bounding the graben between the central break and branching break shown in figure 80 ..................................................................................................................................... . C” ages of mollusk shells in Holocene sediments near the 1968 break ......................................................... . Recurrence intervals for 1968-size events along the Coyote Creek fault, based on the vertical com- ponent of late Holocene displacements along the central and branching breaks of 1968 ................. r—IH 05M H... on: XI Page 177 178 179 179 180 181 184 186 186 189 191 192 193 194 195 196 197 198 199 202 203 204 204 205 117 119 121 XII TABLE 16. 17. 18. 19. 20. 21. 22. . General characteristics of strong-motion seismographs ....................................................................... 24. 25. 26. 27. 28. 29. 30. 31. CONTENTS Earthquakes of magnitude 6.0 and greater along the San Jacinto fault zone since 1899 ......................... Rate of late Holocene strike-slip deformation on the Coyote Creek fault based on 200-year recur- rence interval for 1968-size events ................................................................................................................ Rate of late Holocene vertical deformation at selected points along the Coyote Creek fault based on ZOO—year recurrence interval for 1968-size events _______________________________________________________________________________________ Rate of late Holocene strike-slip deformation on the Coyote Creek fault based on vertical deformation rates and ratio of horizontal to vertical displacement ............................................................................. Strong-motion seismograph records from the Borrego Mountain earthquake ............................................. Strong—motion seismographs not triggered by the Borrego Mountain earthquake but within potential operable range ....................................................................................................................................................... Locations of strong-motion seismographs triggered by the Borrego Mountain earthquake. Maximum accelerations recorded at stations less than 120 miles from the epicenter .................................. General characteristics of displacement meters ................................................................................................... Maximum amplitudes recorded by ground-level displacement meters .................... . Relative maximum displacement spectra from the El Centro seismoscope records ..................................... Maximum seismoscope amplitudes from selected dams in southern California ............................................. Adhesion of cobbles to the surface of ridge B ......................................................................................................... Fluctuations of water level in observation wells, April 8, 1968 ....................................................................... Fluctuations and net changes in water level recorded at gaging stations in California, April 8, 1968 ........................................................................................................................................................................... Page 123 124 125 125 146 146 148 148 149 151 151 152 154 178 185 188 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 INTRODUCTION By ROBERT V. SHARP U. S. GEOLOGICAL SURVEY Seismicity of the Imperial Valley region, perhaps the most exceptional of California’s areas of active tectonism, has been dominated in this century by earthquakes of intermediate magnitude originating along two members of the San Andreas system: the San Jacinto fault zone and the Imperial fault. The most recent shock in the series of nine or more mag- nitude 6 and larger earthquakes along these faults since 1915 occurred at 2:29 G.m.t.,1 April 9, 1968, near Borrego Mountain in northwestern Imperial Valley. The magnitude 6.4 shock, California’s largest in the more than 18 years between the Arvin-Teha- chapi earthquakes of 1952 and the San Fernando earthquake of 1971, was felt over most of southern California and adjacent parts of Nevada, Arizona, and northwestern Mexico. It caused little damage, mainly because it centered in a sparsely settled des- ert region. Horizontal (right-lateral) surface dis- placement accompanied the earthquake along several strands within the complex San J acinto fault zone; the Coyote Creek fault branch nearest the epicenter moved a maximum of 38 cm. Smaller surficial offsets occurred along the Superstition Hills fault and the Imperial fault, also Within the San J acinto zone but at considerable distance from the epicentral area, as well as along a segment of the San Andreas fault about 50 km northeast of the epicenter. (See frontis- piece.) Although the Borrego Mountain earthquake cannot be regarded as a great seismic event either from the standpoint of magnitude or its effects on the works of man, the recognition of many signifi- cant geologic and seismologic consequences of this shock underscores the necessity of intensive docu- mentation of all major earthquakes. 1Greenwich mean time. Corresponding Pacific standard time was 6 :29 p.m., April 8, 1968. Several organizations have cooperated on this de- tailed summary of investigations on the Borrego Mountain earthquake. Data on surface rupturing, including postearthquake creep, were obtained jointly by personnel from the US. Geological Sur- vey’s National Center for Earthquake Research and the California Institute of Technology. Seismo- graphic studies of the main shock were conducted by the Seismological Laboratory at California Institute of Technology and by the National Oceanic and At- mospheric Administration (NOAA) , and aftershock activity was recorded through joint efforts of seis- mologists of US. Geological Survey and California Institute of Technology. Engineering effects, dis- turbance of water resources, and the evidence of previous fault movement were documented by the US. Geological Survey. With the exception of inves- tigations of continuing postearthquake surface creep and the history of Holocene displacement, most of the data reported here are essentially complete. In- asmuch as some of the conclusions on creep may be modified by future events, that part of this investi- gation should be regarded as a summary of progress as of January 1972. A preliminary report on the Borrego Mountain earthquake has been published in the Bulletin of the Seismological Society of America2 and reprinted in Mineral Information Service of the California Division of Mines and Geology.3 The papers assembled here report several impor- tant observations and conclusions, some of them unique, that help to clarify mechanisms of crustal strain release and its relationship to earthquakes. ”Allen, C. B... Grantz. A., Brune, J. N., Clark. M. M., Sharp, R. V.. Theo- dore, T. G., Wolfe, E. W., and Wyss, M., 1968, The Borrego Mountain. Cali- fornia, earthquake of 9 April 1968; A preliminary report: Seismol. Soc. America Bull., v. 58, p. 1183—1186. 31968, v. 21, no. 7, p. 103—106. 2 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 Perhaps the most significant feature, heretofore unrecorded for any other earthquake in California, involves shaking-induced horizontal displacements along faults that lie relatively far from the earth— quake hypocenter and the main body of aftershocks. Another unusual aspect of this seismic event was the uneven distribution of postearthquake displace- ment by creep along the principal Coyote Creek fault breaks; the fault segment that initially shifted the most at the time of the earthquake has subsequently crept very little, whereas on another sector where the initial displacement was relatively small, move- ment has continued and apparently has not yet stopped (1972). The observed aftershock distribu- tion is likewise complex and widespread and is only partly associated with the positions of known steeply dipping surface faults, apparently reflecting com- plexity of the San Jacinto fault zone and perhaps other geologic structures at depth. Furthermore, the complexity of the aftershock distribution contrasts distinctly with that observed after the 1966 Park- field-Cholame earthquakes, the only other shocks of intermediate magnitude associated with predomi- nantly strike-slip surface movement that have oc- curred recently in the San Andreas system; the aftershock pattern for those events was confined within a narrow band along the zone of surface breakage. The epicenter of the main Borrego Moun— tain shock lies near the middle of the distribution of aftershocks, apparently reflecting bilateral fault- ing (fracturing propagating in two directions away from the point of initial rupture) that has not been observed in other recent California earthquakes. Another contrast to the 1966 Parkfield-Cholame earthquakes is the relatively great width of the zone of rupturing developed along the lines of surficial displacement along the Coyote Creek fault. The ground surface, moreover, broke almost exactly at those positions where prior movements are demon- strable by geologic and geomorphic evidence; a simi- lar relationship of prehistoric repetitive offsets has been established in a number of studies elsewhere within the San Andreas system. Although displace- ment was principally horizontal along the Coyote Creek fault break, in several locations vertical offset also was evident but was variable in sense along the length of the fault; however, in every case the sense matched that of previous Quaternary movement shown by geomorphic features. That the history of vertical movement appeared to be this systematic has suggested the possibility of estimating the fre- quency of recurrent displacement through the study of local vertical offsets of datable flat-lying Quater— nary beds transected by the Coyote Creek fault. De- tailed followup studies on the record of Holocene vertical displacement were accomplished by section- ing the fault with several trenches; the data on re- cent fault movements serve as a basis for projections of future activity. Another study attempted was the field estimation of accelerations developed in the epicentral area during the earthquake through ob- servations of dislodged stones mantling slopes and ridges. Aside from these considerations of general significance revealed by the study of the Borrego Mountain earthquake, the exact positions of active fault traces in the region of breakage also constitute new data of appreciable local importance. Special mention should be made of the paper on source parameters of the Borrego Mountain earth- quake, which presents a new technique for estimat- ing the length of surface rupture and the amount of displacement based solely on teleseismic data. The capability of remote determination of these impor- tant measures now permits comparison of past, as well as future, tectonic ruptures that could not be investigated in the field, such as those that lie be— neath ocean basins. Although the San Andreas fault system is prob- ably the most thoroughly studied zone of active faulting in the world, there have been relatively few opportunities in recent times to study major earth- quakes originating along it. The reports on the 1971 San Fernando earthquake (US. Geological Survey Professional Paper 733, 1971), the 1966 Parkfield- Cholame earthquakes (US Geological Survey Pro- fessional Paper 579, 1967), and the 1952 Arvin— Tehachapi earthquakes (California Division of Mines and Geology, Bulletin 171, 1955) are the only others pertaining to the San Andreas system that are comparable in scope and detail. Because of the scientific interest now focused on the geologic and seismologic effects of large earthquakes, similar in- vestigations are being made in tectonically active areas throughout the world. Probably the greatest value in the comparison of such studies lies in the recognition of the similarities and differences in regional tectonic mechanisms now prevalent in the mobile part of the earth’s crust and their implica- tions with regard to earthquake occurrence and crustal strain release. The papers presented here mark a significant step toward better understanding of these mechanisms. TECTONIC SETTING OF THE SALTON TROUGH By ROBERT V. SHARP U. S. GEOLOGICAL SURVEY ABSTRACT The Salton Trough, an extension of the Gulf of California physiographic province, includes the Coachella and Imperial Valley areas of southern California and the Colorado River delta region of northwestern Mexico. Active right-lateral faults of the San Andreas system have dominated in the for- mation and subsequent development of the trough-gulf de- pression. Marine and nonmarine sediments have accumulated within the trough since it formed in Pliocene or possibly Mio- cene time. Tectonic activity is continuing at present at a high rate, as indicated by high levels of seismicity, geodetic measure- ment of crustal strain, geomorphic evidence of recent fault movements, and historic movements on faults in association with earthquakes, the latest having been the shocks in 1968 at Borrego Mountain and in 1971 at Superstition Hills. INTRODUCTION The Gulf of California physiographic province and its northern onland extension, the Salton Trough, together define a remarkably linear and narrow de- pression over 1,400 km long. The Salton Trough encompasses the low-lying areas of the Colorado River delta region in Mexico and the Imperial and Coachella Valley regions situated between the Penin- sular Ranges of coastal southern California and the mountains northeast of the Salton Sea. (See frontis- piece for a regional geomorphic view of the trough.) The land surface in the north-central section of the Salton Trough is below sea level, having been cut off from the Gulf of California by the aggrading deltaic cone of the Colorado River. In Holocene time this basinal segment of the Salton Trough has been peri- odically inundated by bodies of water, the latest having formed the Salton Sea in 1905. (See Menden- hall, 1909, for a brief summary of its formation.) The present geographic limits of the Salton Trough correspond approximately to the boundaries of a late Cenozoic marine and nonmarine deposi- tional basin that has been downwarped, downfaulted, and laterally translated between the bordering ranges along faults of the San Andreas system. (See fig. 1.) The basin extends southward the length of Baja California and is filled with as much as 6 km of sediments in central Imperial Valley (Tarbet, 1951; Biehler and others, 1964, fig. 6) ; most of the sediments may have accumulated in Quaternary time alone. The basin is underlain and bounded by Mesozoic and older crystalline rocks. On the west side of the trough, they consist predominantly of mid-Cretaceous granitic rocks of the southern Cali- fornia batholith, as well as prebatholithic metasedi- mentary and metavolcanic rocks. East of the trough, diverse types of Precambrian crystalline rocks, as well as Mesozoic granitic rocks and other rocks of unknown age, are exposed. In the axial part of the trough, sedimentation under intermittently marine and nonmarine conditions may have been essentially continuous since Pliocene time but perhaps began even earlier. TECTONIC CONTROL OF THE CONFIGURATION OF THE TROUGH Active fault zones of the San Andreas system parallel the northeast margin of the Salton Trough and obliquely transect its southwest flank. Intermit- tent right-lateral movement on some members of the system has continued to change the shape of the basin, and vertical components of offset on many of these faults have outlined the dominant physio- graphic elements within the trough or bordering on it. Despite its overall northwest-trending linearity, the Salton Trough is not simply a narrow, grabenlike structure. South of the Santa Rosa Mountains, for example, where it is relatively broad, the trough is actually a complexly folded and faulted crustal down— warp. Structural complexity within the southwest flank of the trough is reflected physiographically by the projection of several structurally high basement blocks from which the Cenozoic cover has been stripped. In California these insular mountain masses include Borrego Mountain, Fish Creek Moun— tains, Superstition Mountains, and the Coyote Moun- 8 4 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 tains (fig. 1), as well as Sierra de los Cucapas south of the International Boundary. Part of the irregu- larity of the west margin of the trough also is due to right-lateral offset on strike—slip fault zones that enter obliquely from the Peninsular Ranges to the west. For example, right-lateral translation amount- ing to 24 km on breaks within the San J acinto fault zone (Sharp, 1967) has produced the jutting projec- tions of the trough margin represented by Clark Valley and northern Borrego Valley. 34° ORIGIN OF THE BASIN AND THE STRATIGRAPHIC SUCCESSION The origin of the Salton Trough, in particular the age of initial sedimentation, is difficult to pin down with the available evidence. The age of the trough is uncertain chiefly because the oldest strata known are exposed only in small areas along the perimeter. Moreover, in the axial part of the trough (Imperial Valley), basement lies at depths as great as 6 km (Biehler and others, 1964, p. 132), and the deepest x1 , \l ‘ "' liz'e'ERMAnlfilnbwfiz ‘« F4017 W “€377 ‘ ZONE" ‘" .Coachella __ .-. ’ /Z. Wermal l 20 EXPLANATION o—-o u._ 5' ,, G N UI Quaternary (Holocene) alluvium, dune sand, and playa deposits Cenozoic intrusive rocks or their volcanic equivalents Ce nozoic'stratified rocks and interbedded volcanic rocks ~l / A, \‘/_ ‘ \ \/ 1 s -_I Pre-Cenozoic crystalline rocks Contact W—— 210 MILES l 30 KILOMETER AREA OF FIGURE 1; Faults active in Cenozoic time: solid where exposed; dashed or queried where inferred; dotted where concealed. Saw toothed where historic movement has occurred ,1 FIGURE 1.—Generalized geologic map of THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 5 well (4.1 km) does not penetrate all the sedimentary fill (Muffler and White, 1969, p. 170). On the basis of stratigraphic and paleontologic evidence, all that can be said of the segment of the trough in Cali- fornia is that a structural depression was defined there by Pliocene time. However, pre-Pliocene basin fill may underlie the oldest exposed fossil-bearing beds or may be concealed in the axial part of the trough. By contrast, paleontologic evidence from the San Felipe area, on the Gulf of California, estab- lishes the age of the trough at that latitude as Mio- cene or older (Hertlein, 1968, p. 405). Furthermore, the age of the gulf is postulated as about 4 million years (Pliocene) from rates of sea-floor spreading at the junction with the East Pacific Risenear the tip of Baja California, and a protogulf may have formed in the late Miocene (Larson and others, 1968; Moore and Buflington, 1968). Early Cenozoic sedimentary deposits are unknown within the Salton Trough, although they occur very near both flanks. The early and middle Eocene ma- rine Maniobra Formation of Crowell and Susuki SEA ma/vro OCOTILLO ‘ “" BADLANIS 33° Cali’pairio IMPERIAL VALLEY ::: Holtville @ Colexico _:F +\ the Salton Trough, southern California. 476-246 0 - 72 - 2 6 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 (1959) and overlying nonmarine beds of probable Oligocene age (Crowell, 1962, p. 28) crop out in the Orocopia Mountains, northeast of the trough, but they lie east of the San Andreas fault and, according to the now widely accepted view, have been trans- ported as much as a few hundred kilometers by right-lateral slip. Coarse nonmarine gravels prob- ably correlative with conglomerate of the Eocene Poway Group (Kennedy and Moore, 1971) lie very near the southwest margin of the trough (Sharp, 1968) and may locally project under the upper Ceno- zoic deposits within the trough. The Poway Group, however, is widespread across the Peninsular Ranges west of the trough (see, for example, Woodford and others, 1968), and it is unlikely that these gravels represent Eocene basin fill. Regional Cenozoic stratigraphy of the Salton Trough north of the Mexican border has been briefly summarized by Tarbet and Holman (1944), Tarbet (1951), and Dibblee (1954). Composites of their stratigraphic columns, together with modifications and additions to stratigraphic subdivisions and age designations that have resulted from subsequent work, are shown in figure 2. In Mexico, stratigraphic sequences have been outlined by Real (1948) and Anderson (1950) in their pioneer studies on Baja California and the Gulf of California. Although stratigraphic descriptions for this southernmost .0 L Cenozoic deposits in the principal areas of exposure 0 : 8 Q, o- v . o. Lu Southwestern Slde Northwestern Slde Northern end Holo— cene Alluvium, dune sand, and playa deposits __7_E . W E Brawley Fm, OCOUHO CQI- Ocotillo Conglomerate Cabezon Fanglomerate <( E WM L1,: E Borrego Fm. Borrego Deformed gravels of 0 g g Canebrake Fm Whitewater River 0 '5 Ce"3 ' - 7M 0' Canebrake PWV?WV' Palm Spring Fm.b Conglomerate Painted Hill Formation _ _ 7 _ ._ _ 7 _ _ 7 7 2 . . g Imperial Fm. ; Imperial Fm. >- = . 7 Imperial Fm. fl 7 Q. . a: . Alverson Canyon _ Hathaway Fm. < __ _ 7_ _ 7 7 E _ Split Mountain Formation (marineic ' Lu 2 °c’ Fish Creek Gyp.c f p— : q, d ef Mecca Split Mountain a, u . - - . , Coachella Fanglomerate 8 2 Split Mountain Fm. (terrestrial) Fm Formation 2 a. Anza Formation d FIGURE 2. — Composite stratigraphic columns along the flanks of the northern part of the Salton Trough. Wavy lines indi- cate unconformable contacts, smooth lines indicate conformable contacts, queried lines indicate contacts not exposed. Section on southwestern side after Tarbet and Holman (1944) and Dibblee (1954), with modifications from (a) Downs (1957), (b) Downs and Woodard (1962) and Downs and White (1968), (c) Durham and Allison (1962), (d) Woodard (1962; 1963), and (e) Weismeyer 1968). Section on northeastern side after Dibblee (1954), with modifications from (f) Proctor (1968). Section at northern end after Allen (1957). THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 7 sector of the trough are not yet comprehensive, a number of recent studies in various parts of the region have compared or correlated sedimentary units with the Cenozoic sections exposed north of the border. (See Durham and Allison, 1960; Mer- riam, 1965; Hertlein, 1968; Slyker, 1969.) Cenozoic sediments in the Salton Trough range from boulder gravel (or its consolidated conglomer- ate—fanglomerate equivalents) through predominant— ly sandy flood-plain and deltaic deposits to lacustrine or marine silts and clays. Continental conditions of sedimentation may have prevailed in some marginal parts of the trough throughout its history, although in much of the basin deposition has been alternately marine and nonmarine. Coarse clastic strata, mostly around the periphery of the basin, have been given the names Anza, Split Mountain, and Mecca Forma- tions, Coachella Fanglomerate, Canebrake Conglom- erate, Painted Hill Formation, deformed gravels of Whitewater River, Ocotillo Conglomerate, and Cabe- zon Fanglomerate (fig. 2). All these coarse deposits are continental in origin, and many have yielded vertebrate fossils. Strata consisting chiefly of sand but locally interbedded with gravel or silt constitute the upper, marine part of the Split Mountain For- mation, the marine Imperial Formation, the terres- trial Hathaway Formation, and the mostly terrestrial but locally paralic Palm Spring Formation. The Borrego and Brawley Formations consist mostly of silt and clay beds of continental origin; however, the Borrego also contains some sediments deposited in brackish water. Other lithologies represented in the Cenozoic stratigraphy include bedded gypsum deposits (Ver Planck, 1952) termed the Fish Creek}, Gypsum (Dibblee, 1954) and mafic volcanic rocks that make up the Alverson Canyon Formation and isolated flows in the Coachella Fanglomerate and Painted Hill Formation (Allen, 1957). The long-standing controversy regarding the age of the abundantly fossiliferous Imperial Formation bears on the problem of antiquity of the Salton Trough because the underlying strata of the trough generally lack fossil material. Although estimates of the age of the Imperial Formation have ranged from Cretaceous to Pliocene according to different investigators, the most recent interpretations of the fauna] evidence by Durham and Allison (1960, p. 63) favor a Pliocene age. Moreover, despite the tropical affinities of the Imperial fauna that hinder comparison with the cooler water marine faunas found along the Pacific Coast of California, the local occurrence of one temporally restricted echinoid (Encope tennis) in both the Imperial and the late Pliocene San Diego Formations suggests similarity in age of the two units (Allison, 1964, p. 19, 21). Only two definite occurrences of fossils from pre- Imperial strata have been reported from the Salton Trough—one not diagnostic as to age (Tarbet and Holman, 1944; Durham and Allison, 1962; Allison, 1964, p. 21) and the other indicating pre-Imperial deposition possibly only back to late Pliocene time (Allen, 1957, p. 330; Allen, written com‘mun., 1969). Pre-Imperial sediments of Miocene age, perhaps similar to the diatomite described near the head of the Gulf of California (Hertlein, 1968, p. 405), may extend northward into the Salton Trough but have not yet been found on the surface or below. All known fossiliferous deposits above the Im- perial Formation in the Salton Trough are late Pliocene or younger. Although vertebrate fossils have been reported from a number of widely scat- tered locations on both sides of the northern part of the trough, extensively studied faunas indicating Blancan to Irvingtonian age have been recovered only from the Canebrake Conglomerate and the Palm Spring Formation west of the Fish Creek Mountains (Downs, 1957; Downs and Woodard, 1962; Woodard, 1962; Downs and White, 1965). An impoverished marine fauna also recovered in the Palm Spring Formation indicates a brackish- water environment (Downs and Woodard, 1962) that may have prevailed intermittently up through the deposition of fossil-bearing strata of the Borrego Formation (Durham, 1950, p. 23). A large number of source areas around the Salton Trough probably were contributing elastic material at any given time. Coarse elastic units along the margins of the basin reflect local provenances both in their distribution and the lithology of the clasts, which can be correlated with different source ter- ranes (Dibblee, 1954; Allen, 1957; Hays, 1957; Proctor, 1968). Studies of provenance for fine detritus in the Salton Trough have suggested a more complex picture. Merriam and Bandy (1965) and Muffler and Doe (1968) showed that sandy Cenozoic strata in the southern and southeastern parts of the trough resemble material presently being trans- ported down the Colorado River channel more than they do modern sediment derived from ranges pe- ripheral to the trough. Nevertheless, sedimentary structures in some of the same formations elsewhere along the west margin of the Cenozoic basin indicate local provenance from the Peninsular Ranges. CENOZOIC IGNEOUS ROCKS Cenozoic intrusive rocks, or their volcanic equiv- alents, occur at several localities within the Salton Trough. In the largest area of exposure, in the 8 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 northwestern Chocolate Mountains, hypabyssal in- trusive rocks are largely silicic in composition (Morton and Jennings, in Jennings, 1967). Rhyolitic volcanic domes at the southeast corner of the Salton Sea protrude through Holocene sediments on the floor of Imperial Valley (Kelley and Soske, 1936; Robinson and Elders, 1970, 1971; Elders and Robin- son, 1970). These domes and the dacitic volcano of Cerro Prieto in Baja California (see frontispiece for location) constitute the only known Holocene volcanism closely associated with the San Andreas fault system north of the Gulf of California. The domes consist of rhyolite, obsidian, and pumice that have been radiometrically dated at about 16,000 years (Muffler and White, 1969, p. 162). A body of Cenozoic extrusive rocks, named the Dos Palmas Rhyolite by Dibblee (1954, p. 24), crops out in a very small area near the San Andreas fault east of Mecca. Other Cenozoic rocks, termed the Truck- haven Rhyolite by Dibblee (1954, p. 24), that are exposed near the west shore of the Salton Sea have been reinterpreted as silicified sediments by Weis- meyer (1968, p. 30—35). Although extensive basaltic and andesitic flows in the eastern Peninsular Ranges near Jacumba may continue eastward into the Salton Trough, such a relationship has not been established. These rocks are middle Miocene in age, according to a radio- metric age by Hawkins (1970, p. 3326). On the basis of stratigraphic position, Dibblee (1954, fig. 1) assigned a Miocene age to mafic volcanic rocks of the Alverson Canyon Formation of Tarbet and Holman (1944) sporadically exposed within the adjoining southwest flank of the trough. Because of upward revision in the age of the Imperial Formation, how- ever, the Alverson Canyon Formation may be younger than the volcanic rocks near J acumba. CENOZOIC DEFORMATION IN THE SALTON TROUGH The Cenozoic deposits of the Salton Trough have been considerably folded and faulted, and both pro- cesses appear to be continuing at the present time. Faulting is dominated by three active right-lateral strike-slip zones, part of the very broad and complex San Andreas fault system in southern California. Major faults within the trough, as well as those farther south under the Gulf of California, define an en echelon pattern in which the separate strands trend more westerly than the axis of the depression. (See Rusnak and others, 1964; Biehler and others, 1964.) Much has been written regarding the amount of cumulative displacement on the various fault zones within the San Andreas system in California, but major disagreements remain. Current ideas relating to the history of movement within the fault system as a whole are discussed by Crowell (1962) and in Dickinson and Grantz (1968). Folding of the Cenozoic deposits is evident throughout the marginal areas of the Salton Trough. The intensity of deformation by folding is especially pronounced along certain belts bordering on or bounded by the active faults traversing the trough. The pattern of modern sedimentary accumulation and the incipient development of anticlinal hills also reflect continuing diastrophism in the trough, partic— ularly on the southwest side. Many features of the stratigraphy of the Cenozoic deposits attest to regional crustal instability through- out the history of the basin. Although there was considerable initial relief on the basin floor, pro— gressive diastrophism and erosion have allowed the youngest to the oldest sediments to rest directly on basement. Regional unconformities exist, but local- ized intraformational unconformities are far more common. Lack of lateral continuity, particularly in the coarser detritus and in lacustrine beds, is evident all along the margins of the trough. By contrast, sediments in the axial parts of the trough are almost uniformly fine grained; sediments penetrated by deep Wells in the southern Imperial Valley cannot be directly correlated with the Cenozoic formations exposed along the edges of the trough (Dibblee, 1954, p. 25; Muffler and Doe, 1968, p. 388). Wide- spread instability in the flanking regions of the trough is evident in the marked erosional truncation of thick sections of sediments that accumulated .probably since Pleistocene time, such as those ex— posed in the southern Borrego Badlands (fig. 3). The detritus has been redeposited elsewhere in the axial part of the basin, perhaps beneath the Salton Sea. Features such as these suggest that the axial part of the trough may well have been continuously sub- siding, whereas parts of the flanks have alternately received sediment and undergone erosion. CENOZOIC FOLDING Fault movement has outlined the Salton Trough, but major crustal downwarping not attributable to faulting has also taken place, particularly south of the middle of the Salton Sea where the trough widens markedly. Downwarping in the axial part of the trough is demonstrated by the distribution of various stratigraphic units and confirmed by data from wells and geophysical work. (See Biehler and others, 1964, fig. 7.) Cenozoic sediments throughout the trough display a multitude of intermediate- and small-scale folds, THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 9 FIGURE 3. — Oblique aerial view of southern Borrego Badlands (central and right foreground) and West Butte, Borrego Moun- tain (left foreground). Poorly consolidated late Cenozoic strata in the badlands dip moderately to the north (right) and are beveled by a relatively flat erosional surface. A thickness of about 3.5 km of sediments probably has been eroded from the beds at the left side of the view since Pleistocene time. as well as homoclinal dips over large areas. Generally the older strata are more deformed than the younger. Terrace gravels and alluvium, for example. show only local signs of folding. However, some of the most pronounced folding, including overturning and isoclinal folding, is not restricted to older strata but occurs in linear belts adjacent to fault zones, such as in the Mecca Hills east of the town of Mecca and in the Ocotillo Badlands. (See Sharp and Clark, this volume.) Other zones of strongly folded strata resembling diapirs with clay cores appear to de- generate along strike into faults. Examples of the latter occur in the Borrego Badlands. That the land surface is being folded at present is demonstrated by small incipient hills that reflect anticlinal growth near at least one active fault zone and by closed playa basins scattered through- out the southwest side of the Salton Trough. (See Sharp and Clark, this volume.) Furthermore, warped shorelines of ancient Lake Cahuilla, which covered most of the trough earlier in Holocene time, indicate relatively recent large-scale diastrophic changes (Stanley, 1963, 1966). CENOZOIC FAULTING The principal fault zones in the Salton Trough consist of the San Andreas zone, near the north- east margin, a group of boundary faults that are concealed along the southwest edge of Coachella Valley, the widely branching San J acinto fault zone which transects the southwest flank of the trough, and the Elsinore fault zone along the southwest edge of the trough. (See fig. 1.) Except for the breaks on the southwest side of Coachella Valley, these faults all display the surficial features char- acteristic of the San Andreas system throughout California: linearity, northwest-southeast trend, physiographic evidence of recent activity, and clear- cut surficial evidence of right-lateral strike-slip off- set. Episodes of historic right-lateral movement on several of the faults, together with the rate of crustal strain accumulation and the seismicity, all indicate that the Salton Trough region is one of the most active areas along the San Andreas system. Larson, Menard, and Smith (1968) as well as Moore and Buflington (1968) discussed the signif- icance of the San Andreas fault system in terms of the transformation of ocean-floor spreading move- ments near the tip of Baja California to strike-slip movement within the Gulf of California and the Salton Trough. According to the transform-faulting model, coastal California and Baja California‘are drifting northwestward relative to the region east of the Salton Trough and the Gulf of California; movement is achieved by a combination of right- lateral slip along northwest-trending faults and horizontal spreading motions on offset segments of the East Pacific Rise situated between the fault 10 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 strands within the trough and gulf. The distribution of recent small earthquakes in the Colorado River delta region (Lomnitz and others, 1970) is consistent with the transform model. Future geologic, seismo— logic and geodetic investigations may provide evi- dence that will indicate whether strike-slip faulting occurs in conjunction with spreading in the trough. SAN ANDREAS FAULT ZONE The San Andreas fault zone comprises two princi- pal fractures in northern Coachella Valley. An east- ern branch (Mission Creek fault of Allen, 1957, and Proctor, 1968) extends along a nearly straight line .to the San Bernardino Mountains, and a western branch (Banning fault) curves to a nearly east-west trend at the north end of the trough. Near Indio, the zone includes a number of subsidiary breaks that lie on the northeast side of the main fault. Southeast of Indio, the San Andreas zone is relatively straight and continuous to the northeast shore of the Salton Sea. Throughout much of Coachella Valley, the fault zone is defined in alluvium by scarps and by ground- water and (or) vegetational effects (Hope, 1969) and in the areas of exposed Cenozoic strata by crushed sediments and gouge and by geologic dis- continuities. Using gravity data, Biehler, Kovach, and Allen (1964) estimated that pre—Cenozoic base— ment rocks have been vertically downdropped at least 3.2 km on the southwest side of the San Andreas fault in Coachella Valley. Surficial expression of the San Andreas fault zone dies out southeast- of the Salton Sea. Although sporadic surficial expression of breakage nearly on line with the fault-trace projection from the north- west is evident on the northern part of Imperial Valley (Babcock, 1971, fig. 2), the large displace- ment proposed for the San Andreas fault (for example, see Crowell, 1962) suggests that it may continue much farther southeastward in Imperial Valley (Biehler and others, 1964, p. 137). Gravity data (Kovach and others, 1962, p. 2867—2869) indi- cate a major concealed structural break along the northeast side of Imperial Valley that lies well east of a linear projection of the exposed fault trace northwest of the Salton Sea. However, on the basis of electrical resistivity profiles observed near Braw- ley, Tsvi Meidav (unpub. rept.) suggested that the San Andreas fault bends to a more southerly trend, bringing the break to a position closer to the axis of the trough. Moreover, the possibility of crustal spreading movement beneath the southeastern Salton Sea area and its transformation into horizontal slip on the San Andreas fault to the northwest has been recently suggested (Meidav, 1970; Lomnitz and others, 1970). Historic movement on the San Andreas fault with- in the Salton Trough has not been recorded in con- junction with any earthquake prior to the Borrego Mountain event of 1968. Major seismic events at other times in the trough, however, may have pro- duced ground breakage that escaped detection. CONCEALED FAULTS ON THE SOUTHWEST SIDE OF COACHELLA VALLEY A long generally southeast-trending fault zone is suspected to exist along the southwest side of Coach- ella Valley (queried dotted line in fig. 1). Several lines of evidence point to its existence: (1) The San J acinto and Santa Rosa ranges appear to have been uplifted relative to the floor of the valley, and the slopes along the east flank of the highlands are steep (see Allen, 1957, p. 347; Proctor, 1968, p. 35) ; (2) there is a pronounced gravity gradient along this line (Biehler and others, 1964, p. 138) ; and (3) structural features in the crystalline rocks and over- lying gravel beds in these ranges suggest that down- folding along the east margin of the ranges cannot account for the observed gravity gradient. The lack of scarps or other surficial features along the valley floor and the fact that many deeply incised canyons along the edge of the ranges have nearly flat alluvial floors indicate that the zone has been inactive for a relatively long period. Consequently, the only clues to the position of these faults are the gravity data; the fault zone shown in figure 1 is only a schematic representation of these data. The breaks indicated by the gravity data appear to end near the southeast terminus of the Santa Rosa Mountains. Moreover, farther south in the basin the basement surface apparently slopes gently rather than abruptly eastward from points due south of the Santa Rosa Mountains to the deepest parts of the trough near its axis, as shown by well data (Oake— shott, 1952, p. 12) and basement depths determined by seismic and gravity methods (Biehler and others, 1964, p. 133). SAN JACINTO FAULT ZONE The San J acinto fault zone enters the Salton Trough from the northwest at about the latitude of the Salton Sea and cuts diagonally into the basin. (See fig. 1.) Throughout its length in southern California the zone is characterized by southeast- ward-divergent branching of multiple strands (Dib- blee, 1954; Sharp, 1967). Where it enters the trough, the zone is about 10 km wide and consists of three recognizable strands. The northernmost, the Clark fault, extends through Clark Valley and along the south tip of the Santa Rosa Mountains (Dibblee, 1954; Sharp, 1972). The middle break lies along the THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 11 west edge of Clark Valley and extends southward from the Clark fault into the Borrego Badlands, where it apparently dies out. The third strand, the southwest one, is the Coyote Creek fault, along which the principal breakage associated with the 1968 Borrego Mountain earthquake occurred. The Coyote Creek fault follows the northeast edge of Borrego Valley and, with en echelon overlaps and branching, extends along the margins of Borrego Mountain and the Ocotillo Badlands southeastward to a point near the easternmost Fish Creek Moun- tains. Mapping in the area of the 1968 breakage (see Sharp and Clark, this volume) has demonstrated highly complex patterns of surficial fractures along the Coyote Creek fault. Other faults lying farther southeast in western Imperial Valley and in the Colorado River delta region in Mexico have also been regarded as part of the San Jacinto fault zone (Beal, 1915; Dibblee, 1954; Biehler and others, 1964; Merriam, 1965; Sharp, 1967). The Superstition Mountain fault and the Superstition Hills fault fall-well within the southeastward projection of the zone encompassing all breaks near Borrego and Clark Valleys, as does the fault extending southeastward from Cerro Prieto in Baja California. (See frontispiece and fig. 4.) Moreover, the Superstition Mountain fault may join the Coyote Creek fault as a branch in the subsur- face, although this cannot be proved yet. Alterna- tively, the principal fracture of the Coyote Creek fault could turn to a more southern trend and con- nect with recently observed cracks of possible tec- tonic origin near US. Highway 80 about 7 km east of Plaster City. (See Sharp and Clark, this volume.) The Imperial fault, as defined by the line of break- age during the major 1940 Imperial Valley earth- quakes, may also belong to the San Jacinto zone. This fault lies nearly on the projection of the Clark fault, and the two could join at depth (Sharp, 1967, p. 726). However, it is also possible that movement on the Imperial fault is partly transferred north- eastward to the San Andreas fault by crustal spreading near the southern Salton Sea (Lomnitz and others, 1970). Displacement across the San Jacinto fault zone in the Peninsular Ranges northwest of the Salton Trough has been principally right lateral and has amounted to about 24 km since movement began (Sharp, 1967). About 19 km of this has been on breaks east of Coyote Mountain, and about 5 km has been along the Coyote Creek fault. Vertical components of displacement on various breaks along the Coyote Creek fault have produced a number of topographically high welts, some of which show separations probably in excess of 1 km. The crystalline rocks and sediments exposed at Borrego Mountain were elevated in such a manner. In addi- tion, sediments exposed in the Ocotillo Badlands have been squeezed up between en echelon breaks along the Coyote Creek fault. (See Sharp and Clark, this volume.) Except for slippage on the San Andreas fault, all the lines of movement associated with the 1968 Borrego Mountain earthquake coincide with fault strands lying within the broad envelope of the San J acinto zone. Ground displacement had been detected several times on some of the same fault strands prior to 1968. The first, and by far the largest, was right-lateral offset of as much as 4.3 meters that occurred on the Imperial fault during a series of major earthquakes on May 18, 1940, the largest of which was of magnitude 6.4 (Buwalda and Richter, 1941; Ulrich, 1941; Richter, 1958, p. 488; Trifunac and Brune, 1970). A small amount of right-lateral offset that was detected on the Superstition Hills fault in February 1951 probably originated in asso- ciation with a magnitude 5.6 earthquake in the Sup- erstition Hills on January 23, 1951 (Allen and others, 1965, p. 768). Very slight right-lateral move- ment recurred on part of the Imperial fault at the time of a magnitude 3.6 shock on March 4, 1966 (Brune and Allen, 1967b) and again after the magnitude 5.3 Superstition Hills earthquake of September 30, 1971 (C. R. Allen, oral commun., 1971) ; the 1966 event was the smallest earthquake ever known to have been associated with surface displacement. ELSINORE FAULT ZONE The Elsinore fault zone, a distinct yet surficially discontinuous zone of fractures, extends southeast- ward from the Peninsular Ranges into the Salton Trough near the Tierra Blanca Mountains (fig. 1). In the trough the zone is best defined and most continuous where it bounds the southwest sides of the Coyote Mountains and the Sierra de los Cucapas (frontispiece). As with breaks of the San Jacinto zone that are concealed by alluvium on the floor of Imperial Valley, continuity between many individual strands cannot be demonstrated. The total strike-slip offset on the Elsinore fault probably is small compared with displacement esti- mated on the San Jacinto and San Andreas zones (Sharp, 1968; Baird and others, 1970). Historic movement has not been documented, and, indeed, within the Salton Trough only the sector south of the International Boundary has been characterized by appreciable seismicity in recent times. 12 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 SEISMICITY IN THE SALTON TROUGH A remarkably high rate of tectonic activity in the Salton Trough is signified by the record of seismic activity (fig. 4) and the accumulation of horizontal crustal shear strain, as well as by geomorphic evi- dence. In terms of earthquake frequency, both for minor to moderate—sized events and for smaller shocks down to the microearthquake range, the Sal— ton Trough is evidently the most active area lying astride the San Andreas system (Biehler and others, 1964; Allen and others, 1965; Brune and Allen, 1967a). However, very great earthquakes have not . occurred in the Salton Trough, at least within the period of the historic record (Allen and others, 1965, p. 783). MINOR T0 MODERATE-SIZED EARTHQUAKES Although the regional level of seismic activity in the Salton Trough is at present the highest known in California, there is great diversity in apparent rates of seismicity associated with the different fault zones within its boundaries. The San J acinto fault zone both within the trough and in its sector farther northwest is apparently the only break within the San Andreas system in southern California that is clearly delineated by post-1934 earthquake epicenters (Allen and others, 1965, p. 772) ; moreover, of the 10 historic earthquakes of magnitude greater than 6 located within the area of figure 4, eight occurred on or near breaks within the San J acinto zone. In contrast, the San Andreas fault zone has not in the period of the historical record been associated with a large number of earthquakes, at least along the zone north of the Salton Sea where physiographic expression of recent activity along the break is evi- dent. Southward from the Salton Sea, a diffuse belt of seismicity converges with epicenters associated with the San Jacinto zone. In a similar way, the Elsinore fault zone north of the International Bound- ary is rather loosely associated with a belt of rela- tively small earthquakes, although in Mexico and northwest of the trough larger earthquakes have centered within this zone (Richter, 1958, p. 533; Biehler and others, 1964; Allen and others, 1965). A detailed study of seismicity in the Colorado River delta region (Lomnitz and others, 1970) showed that minor earthquakes are markedly concentrated along lines of recognized faulting, a characteristic not re- vealed by the earlier seismic records. MICROEARTHQUAKE ACTIVITY Brune and Allen (1967a) conducted a regional survey of the San Andreas fault system in southern California to measure microearthquake frequency and distribution for comparison with the levels of activity of larger earthquakes. In the Salton Trough, they found that there is Virtually no microearth- quake activity on the San Andreas fault zone be- tween Desert Hot Springs and the Mexican border, an observation that agrees with the low level of seismic activity shown by the record of larger earth- quakes since 1934 (Brune and Allen, 1967a, p. 291). In terms of microearthquake activity, the San Ja- cinto fault zone appears to be the most active zone 1,. W Wu?" * ' W 4 e- Rog] . ./ - O o ’/ _' , ' o ,/ »- . . o / . 4 :' - ' if Q / ‘_ - _ 9“ , .. . I. fi‘ , ,, _ ' M . . u . / : - , A - 6.2 (Mar. l9, l954) 6.0 (May IS, 1940),! M, \. /“ >21 . .~ . \ I . . mp5? / ' Q ,- 17' - I 9-: . . /—\ — ’ 6. (Mo I9, I940) ° 25 “'LES Kg“ h lJAC/Nm ./. ’ flu” . zO/VE/ . 6 .1 F ULT y o SSKILOMETERS V 7 ' ~ _ , ‘ - A\_’ (:99; 0 (June 23, ISISZ/ 6 I ~ ‘ e. - V , (May I9 I940) - (Apr. 3, I968) , .- 6.5 (Oct 2|, I942) $63 ' EXPLANATION , , ‘ \ ' A? ° . . 4.5.4.9 ‘ _ ‘5 ‘ l / ' "-- 0 5.0-5.4 ‘ I. ,{u ‘ \ A \ / \ , 05.5-33.9 _ ‘1 is ,I 4. ’ A} .6.0'6.9 . We .- m‘ z\ W ‘ . .. , m, 6LsI _ , 0%, Earthquake epIcenter. = . a n / , showing magniIude g“, , g ‘1 "f 5 9g m \a 1 ’ ' " . FIGURE 4.—Seismicity of the northern Salton Trough region. Seismic events less than magnitude 6 for period 1934—61 adapted from California Department of Water Resources (1964), with modifications from Trifunac and Bruce (1970). Seismic events of magnitude greater than 6 for period 1912—68 adapted from Allen, St. Amand, Richter, and Nordquist (1965) and Allen and Nordquist (this volume). THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 13 in southern California, again in strong contrast with the San Andreas zone (Brune and Allen, 1967a, p. 291; Arabasz and others, 1970). RECENT CRUSTAL STRAIN Geodetic data obtained from repeated surveys by the US. Coast and Geodetic Survey in the northern part of the Salton Trough over the past third of a century enable an estimate of the rate at which right-lateral crustal strain has been accumulating across the San Andreas system (Whitten, 1956; 1960; Miller and others, 1970). According to recent syntheses of geodetic data (Miller and others, 1970, figs. 8, C8), buildup of right-lateral shear-strain component as measured across the entire Imperial Valley averaged about 2—3 cm/yr between 1934 and 1967. Over shorter intervals of time (1934—41, 1941— 54, and 1954—67) in this period, however, strain rates varied remarkably. Crustal strain near the Imperial fault in the central part of Imperial Valley has been consistently right lateral, averaging about 2 cm/yr since the large 1940 earthquakes, whereas the data suggest that strain in the left-lateral sense has prevailed in other parts of the Imperial Valley region at least at certain times. For example, an area in eastern Imperial Valley along the southeast- ward projection of the San Andreas fault apparently shows strain accumulation in the left-lateral sense parallel to the trough axis after 1941 (Miller and others, 1970, fig. 6). Left-lateral strain buildup of about 1 cm/yr across the San Andreas fault zone in the northernmost part of the Salton Trough has also been reported by Hofmann (1968, fig. 1) for the periods 1959—65 and 1965-67. The apparent temporal variability of strain accumulation suggests that time bases appreciably longer than a third of a century will be required to obtain a meaningful estimate of the average rate in the Salton Trough region. How- ever, it is interesting to note that the average rate of strain accumulation measured over the longest term is comparable to recently measured rates of crustal strain across the San Andreas fault zone in central California, ranging from 0 to about 4 cm/yr at different locations (Hofmann, 1968). REFERENCES CITED Allen, C. R.. 1957, San Andreas fault zone in San Gorgonio Pass, southern California: Geol. Soc. America Bull., v. 68, p. 315—350. Allen, C. R., St. Amand, Pierre, Richter, C. F., and Nordquist, J. M., 1965, Relationship between seismicity and geologic structure in the southern California region: Seismol. Soc. America Bull., v. 55, p. 753—797. Allison, E. C., 1964, Geology of areas bordering Gulf of Cali— fornia, in van Andel, T. H., and Shor, G. G., Jr., eds., Marine geology of the Gulf of California: Am. Assoc. Petroleum Geologists Mem. 3, p. 3—29. Anderson, C. A., 1950, Geology of islands and neighboring land areas, pt. 1 of The 1940 E. W. Scripps cruise to the Gulf of California: Geol. Soc. America Mem. 43, 53 p. Arabasz, W. J., Brune, J. N., and Engen, G. R., 1970, Locations of small earthquakes near the trifurcation of the San Ja- cinto fault southeast of Anza, California: Seismol. Soc. America Bull., v. 60, p. 617—627. Babcock, E. A., 1971, Detection of active faulting using oblique infrared aerial photography in the Imperial Valley, Cali- fornia: Geol. Soc. America Bull., v. 82, p. 3189—3196. Baird, A. K., Welday, E. E., and Baird, K. W., 1970, Chemical variations in batholithic rocks of southern California [abs.] : Geol. Soc. America Abs. with Programs, v. 2, no. 2, p. 69. Beal, C. H., 1915, The earthquake in Imperial Valley, Califor- nia, June 22, 1915: Seismol. Soc. America Bull., v. 5, p. 130—149. 1948, Reconnaissance of the geology and oil possibili- ties of Baja California, Mexico: Geol. Soc. America Mem. 31, 138 p. Biehler, S., Kovach, R. L., and Allen, C. R., 1964, Geophysical framework of northern end of Gulf of California struc- tural province, in van Andel, T. H., and Shor, G. G., Jr., eds., Marine geology of the Gulf of California: Am. Assoc. Petroleum Geologists Mem. 3, p. 126-143. Brune, J. N., and Allen, C. R., 1967a, A micro—earthquake sur— vey of the San Andreas fault system in southern Cali- fornia: Seismol. Soc. America Bull., v. 57, p. 277—296. 1967b, A low-stress-drop, low-magnitude earthquake with surface faulting: The Imperial, California, earth- quake of March 4, 1966: Seismol. Soc. America Bull., v. 57, p. 501—514. Buwalda, J. P., and Richter, C. F., 1941, Imperial Valley earth- quake of May 18, 1940 [abs.]: Geol. Soc. America Bull., v. 52, p. 1944—1945. California Department of Water Resources, 1964, Crustal strain and fault movement investigation: California Dept. Water Resources Bull. 116—2, 96 p. Crowell, J. C., 1962, Displacement along the San Andreas fault, California: Geol. Soc. America Spec. Paper 71, 61 p. Crowell, J. C., and Susuki, T., 1959, Eocene stratigraphy and paleontology, Orocopia Mountains, southeastern Califor- nia: Geol. Soc. America Bull., v. 70, p. 581—592. Dibblee, T. W., Jr., 1954, Geology of the Imperial Valley region, in Jahns, R. H., ed., Geology of southern California; chap. 2, Geology of the natural provinces: California Div. Mines Bull. 170, p. 21—28. Dickinson, W. R., and Grantz, Arthur, eds., 1968, Proceedings of conference on geologic problems of San Andreas fault system: Stanford Univ. Pub. Sci., v. 11, 377 p. Downs, Theodore, 1957, Late Cenozoic vertebrates from the Imperial Valley region, California [abs.] : Geol. Soc. America Bull., v. 68, p. 1822. Downs, Theodore, and White, J. A., 1965, Late Cenozoic verte- brates of the Anza-Borrego Desert area, southern Cali- fornia [abs.]: Geol. Soc. America, Program for AAAS Joint Mtg., Berkeley, p. 110-111. 1968, A vertebrate faunal succession in superposed sediments from late Pliocene to middle Pleistocene in Cali— fornia: Internat. Geol. Cong., 23d, Prague 1968, Tertiary- Quaternary boundary, v. 10, p. 41—47. 14 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 Downs, Theodore, and Woodard, G. D., 1962, Middle Pleisto- cene extension of the Gulf of California into the Imperial Valley [abs.]: Geol. Soc. America Spec. Paper 68, p. 21. Durham, J. W., 1950, Megascopic paleontology and marine stratigraphy, pt. 2 of The 1940 E. W. Scripps cruise to the Gulf of California: Geol. Soc. America Mem. 43, 216 p. Durham, J. W., and Allison, E. C., 1960, The geologic history of Baja California and its marine faunas: Systematic Zoology, v. 9, p. 47—91. 1962, Stratigraphic position of the Fish Creek gyp- sum at Split Mountain Gorge, Imperial County, California [abs.]: Geol. Soc. America Spec. Paper 68, p. 22. Elders, W. A., and Robinson, P. T., 1970, Possible sea-floor spreading in the Imperial Valley of California—a model for magma generation [abs.]: EOS (Am. Geophys. Union Trans.) , v. 51, p. 422. Hawkins, J. W., 1970, Petrology and possible tectonic signifi— cance of Late Cenozoic volcanic rocks, southern California and Baja California: Geol. Soc. America Bull., v. 81, p. 3323—3338. Hays, W. H., 1957, Geology of the central Mecca Hills, Rive-r- side County, California: Yale Univ. Ph. D. thesis. Hertlein, L. G., 1968, Three late Cenozoic molluscan faunules from Baja California, with a note on diatomite from west of San Felipe: California Acad. Sci. Proc., v. 30, p. 401— 405. Hofmann, R. B., 1968, Recent changes in California fault movement, in Dickinson, W. R., and Grantz, Arthur, eds., Proceedings of conference on geologic problems of San Andreas fault system: Stanford Univ. Pub. Geol. Sci., v. 11, p. 89—93. Hope, R. A., 1969, Map showing recently active breaks along the San Andreas and related faults between Cajon Pass and Salton Sea, California: U.S. Geol. Survey open-file map, scale 1:24,000. Jennings, C. W., 1967, GeolOgic map of California, Olaf P. Jen- kins edition, Salton Sea sheet: California Div. Mines and Geology. Kelley, V. C., and Soske, J. L., 1936, Origin of the Salton vol- canic domes, Salton Sea, California: Jour. Geology, v. 44, p. 496-509. Kennedy, M. P., and Moore, G. W., 1971, Stratigraphic rela- tions of upper Cretaceous and Eocene formations, San Diego coastal area, California: Am. Assoc. Petroleum Geologists Bull., v. 55, p. 709—722. Kovach, R. L., Allen, C. R., and Press, Frank, 1962, Geophysi- cal investigations in the Colorado Delta region: Jour. Geophys. Research, v. 67, p. 2845—2871. Larson, R. L., Menard, H. W., and Smith, S. M., 1968, Gulf of California: a result of ocean-floor spreading and trans- form faulting: Science, v. 161, p. 781—784. Lomnitz, Cinna, Mooser, Federico, Allen, C. R., Brune, J. N., and Thatcher, Wayne, 1970, Seismicity and tectonics of the northern Gulf of California region, Mexico, Prelimi— nary results. Geofisica Internacional, v. 10, no. 2, p. 37—48. Meidav, Tsvi, 1970, Possible sea-floor spreading in the Imperial Valley; 1, Structural setting [abs.]: EOS (Am. Geophys. Union Trans.), v. 51, no. 4, p. 421. Mendenhall, W. C., 1909, Groundwater of the Indio region, Calif.: U.S. Geol. Survey Water-Supply Paper 225, 56 p. Merriam, Richard, 1965, San Jacinto fault in northwestern Sonora, Mexico: Geol. Soc. America Bull., v. 76, p. 1051— 1054. Merriam, Richard, and Bandy, O. L., 1965, Source of upper Cenozoic sediments in Colorado delta region: Jour. Sed. Petrology, v. 35, p. 911—916. Miller, R. W., Pope, A. J., Stettner, H. S., and David, J. L., 1970, Crustal movement investigations —triangulation, Imperial Valley, vicinity of El Centro: U.S. Coast and Geodetic Survey Data Rept. 10, 117 p. Moore, D. G., and Buffington, E. C., 1968, Transform faulting and growth of the Gulf of California since the late Plio- cene: Science, v. 161, p. 1238—1241. Muffler, L. J. P., and Doe, B. R., 1968, Composition and mean age of detritus of the Colorado River delta in the Salton trough, southeastern California: Jour. Sed. Petrology, v. 38, p. 384-399. Muffler, L. J. P., and White, D. E., 1969, Active metamorphism of upper Cenozoic sediments in the Salton Sea geothermal field and the Salton trough, southeastern California: Geol. Soc. America Bull., v. 80, p. 157—182. . Oakeshott, G. B., 1952, Exploratory wells drilled outside oil and gas fields in California to December 31, 1950: Cali— fornia Div. Mines Spec. Rep. 23, 77 p. Proctor, R. J., 1968, Geology of the Desert Hot Springs—upper Coachella Valley area, California: California Div. Mines and Geology Spec. Rep. 24, 50 p. Richter, C. F., 1958, Elementary seismology: San Francisco. W. H. Freeman, 768 p. Robinson, P. T., and Elders, W. A., 1970, Possible sea-floor spreading in the Imperial Valley of California: Xenoliths in rhyolite volcanoes as samples of the basement [abs.]: EOS (Am. Geophys. Union Trans.), v. 51, p. 422. 1971, Late Cenozoic volcanism in the Imperial Valley, California [abs.] : Geol. Soc. America Abstracts with Pro- grams, v. 3, no. 2, p. 185. Rusnak, G. A., Fisher, R. L., and Shepard, F. P., 1964, Bathym— etry and faults of Gulf of California, in van Andel, T. H., and Shor, G. G., Jr., eds., Marine geology of Gulf of Cali- fornia: Am. Assoc. Petroleum Geologists Mem. 3, p. 59—75. Sharp, R. V., 1967, San Jacinto fault zone in the Peninsular Ranges of southern California: Geol. Soc. America Bull., v. 78, p. 705—730. 1968, The San Andreas fault system and contrasting pre-San Andreas structures in the Peninsular Ranges of southern California [abs.], in Dickinson, W. R., and Grantz, Arthur, eds., Proceedings of conference on geo— logic problems of San Andreas fault system: Stanford Univ. Pub. Geol. Sci., v. 11, p. 292—293. 1972, Map showing recently active breaks along the San Jacinto fault zone between the San Bernardino area and Borrego Valley, California: U.S. Geol. Survey Misc. Geol. Inv. Map I—675, scale 1:24,000. Slyker, R. G., 1969, Geologic and geophysical reconnaissance of the Valle de San Felipe region, Baja California, Mexico [abs.]: Geol. Soc. America Spec. Paper 121, p. 560. Stanley, G. M., 1963, Prehistoric lakes in Salton Sea basin [abs.]: Geol. Soc. America Spec. Paper 73, p. 249—250. 1966, Deformation of Pleistocene Lake Cahuilla shore- line, Salton Sea basin, California [abs.] : Geol. Soc. Amer- ica Spec. Paper 87, p. 165. Tarbet, L. A., 1951, Imperial Valley, in Possible future petro- leum provinces of North America: Am. Assoc. Petroleum Geologists Bull., v. 35, p. 260-263. THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 15 Tarbet, L. A., and Holman, W. H., 1944, Stratigraphy and micropaleontology of the west side of Imperial Valley, California [abs]: Am. Assoc. Petroleum Geologists Bull., v. 28, p. 1781—1782. Trifunac, M. D., and Brune, J. N., 1970, Complexity of energy release during the Imperial Valley, California, earthquake of 1940: Seismol. Soc. America Bull., v. 60, p. 137—160. Ulrich, F. P., 1941, The Imperial Valley earthquake of 1940: Seismol. Soc. America Bull., v. 31, p. 13—31. Ver Planck, W. E., 1952, Gypsum in California: California Div. Mines Bull. 163, 151 p. Weismeyer, A. L., Jr., 1968, Geology of the northern portions of the Seventeen Palms and Fonts Point quadrangles, Im- perial and San Diego Counties, California: Univ. South— ern California M.A. thesis. Whitten, C. A., 1956, Crustal movements in California and Nevada: Am. Geophys. Union Trans, v. 37, p. 393—398. 1960, Horizontal movement in the earth’s crust: J our. Geophys. Research, v. 65, p. 2839—2844. Woodard, G. D., 1962, Stratigraphic succession of the west Colorado Desert, San Diego and Imperial Counties, South- ern California [abs]: Geol. Soc. America Spec. Paper 68, p. 63—64. 1963, The Cenozoic succession of the west Colorado Desert, San Diego and Imperial Counties, southern Cali- fornia: California Univ., Berkeley, Ph. D. thesis. Woodford, A. 0., Welday, E. E., and Merriam, Richard, 1968, Siliceous tufl’ clasts in the upper Paleocene of southern California: Geol. Soc. America Bull., v. 79, p. 1461—1486. FORESHOCK, MAIN SHOCK, AND LARGER AFTERSHOCKS OF THE BORREGO MOUNTAIN EARTHQUAKE1 By CLARENCE R. ALLEN and JOHN M. NORDQUIST SEISMOLOGICAL LABORATORY, CALIFORNIA INSTITUTE OF TECHNOLOGY ABSTRACT The Borrego Mountain earthquake, magnitude 6.4, occurred at 02:28:59.1 G.m.t. on April 9, 1968 and has been assigned a hypocenter at 33°11.4’ N., 116°07.7’ W., h=11.1 km. The focal-mechanism solution indicates right-lateral slip on a fault striking N. 48“ W. and dipping 83° NE., which is consistent with the field observations of faulting and the regional tectonic framework. A single foreshock of magnitude 3.7 preceded the main shock by one minute, but no other precursory activity has been identified. During the year following the event, 135 after- shocks of magnitude 3.0 and greater have been identified and located, outlining a broad zone of activity centered on but dis- placed 2—3 km northeast of the 33-km-long surface rupture on the Coyote Creek fault. Fracturing at depth during the after— shock period evidently occurred throughout the width of the San J acinto fault zone, but initial surface faulting was local- ized along the Coyote Creek fault at the zone’s southwestern margin. The area of aftershock activity enlarged progressively with time, and the region of the original epicenter became rela- tively inactive late in the aftershock period, leading to a doughnut-shaped epicentral distribution of late aftershocks. Inasmuch as the epicenter of the main shock was roughly mid- way along the zone of aftershock activity, the faulting pre- sumably was bilateral. This kind of faulting is unusual in California. INTRODUCTION The earthquake of magnitude 6.4 that occurred near Borrego Mountain, Calif, was the largest earth— quake to occur in the conterminous United States since 1959. It ranks among the larger shocks recorded in the southern California region since the establish— ment of modern seismographic stations (fig. 5). The earthquake was preceded by a minor foreshock, was followed by a series of aftershocks that lasted for several months, and was accompanied by right-lat- eral surface faulting along a 33-km segment of the Coyote Creek fault (Clark, “Surface Rupture Along the Coyote Creek Fault,” this volume). Probably the most surprising and significant phenomenon associ- ated with the earthquake was the triggering of small 1Contribution No. 1953, California Institute of Technology, Division of Geological and Planetary Sciences, Pasadena, Calif. 16 displacements on a number of distant faults far outside the aftershock area (Allen and others, this volume). The Borrego Mountain earthquakes occurred in a fortunate location with respect to the distribution of seismographic stations that were in operation at the time of the main shock. Seven stations of the Caltech I _ j I 58° ”9° x ”7° “5° °B/:hap \\ \ ' \:\ \\ \\ \\Lone Pm: ‘\\~\ " \. \l:\ “rt/m: ' ‘\ ~\ m'\ k x Li/ \ °Ealver ova/ave m, ,-- \\ °Earslm;\— \ \ \ .Polmdu/e ‘ \ \.\ \ .1 \ °50n Bambldlmz __ “W’fi . ¥ . l9546. 60 FIGURE 5.— Epicenter of the 1968 Borrego Mountain earth— quake in relation to major faults, and epicenters of other earthquakes of magnitude 6.0 and greater in southern Cali- fornia since 1912. (Adapted from Allen and others, 1965.) THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 network surrounded the epicenter within 150 km, including a temporary trailer-mounted station 47 km distant at Obsidian Butte (OBB, fig. 6). Thus we "'1- San Bernardina A RVR Palm Springs 0 0 India AHAY Epictnlu Ann M 3 mm 7 055 K \3 GLA‘ Faun break San Diego ECC BAR El Cznlm ____ ‘ _ I ..... u "5' IOO MILE S 50 KM Emm- o FIGURE 6. — Location of Caltech seismographic stations (solid triangles) that were operating at the time of the Borrego Mountain earthquake. The area of the detailed epicenter map (fig. 7) is also shown. have some confidence in the epicentral locations of most of the larger shocks of the series, although depth control was minimal until the time that the first portable units were installed in the epicentral area. Although the main shock occurred in the Evening (6:29 p.m., April 8, 1968, local time), rapid com- puter location of the epicenter using the Caltech telemetered array allowed immediate planning of a field—recording program. Several field crews were dispatched to the epicentral area before 9:00 p.m., and the first backpack-mounted seismograph was installed within 5 km of the epicenter shortly after midnight. By 3:00 a.m., the first film-recording trailer-mounted unit was in operation at Ocotillo Wells (OCT, fig. 7), and two more trailer units were installed the next day (ELW and FOR, fig. 7) (table 1). These trailer units operated up to and following TABLE 1. — Caltech trailer-mounted stations in operation dur- ing the Borrego Mountain aftershock period 5mm“ Lat (N.) Long (w.) Period of operation (G.m.t) Code Name ELW Ella Wash ................ 33°17.20’ 116°08.95’ 4—10—68 (03:05) to 5—13—68 FCR Fish Creek Range. 33°01.82’ 116°03.50’ 4— 9—68 (19:23) to 6—12—68 OBB Obsidian Butte ........ 33°10.15' 115°38.17’ 1— 8—68 to 9-15—68 12—17-68 to 12—20—68 3—21—69 to 6—14-69 OCT Ocotillo Wells .......... 33°09.60’ 116°09.04' 4— 9—68 (10:12) to 6—11—68 the time that the much more extensive U.S. Geologi- cal Survey array began yielding high-precision hypocentral information on April 12 (Hamilton, this volume). This paper describes the events that took place prior to the installation of the U.S. Geological Survey array and summarizes the larger aftershocks Aqua Cohen's s ‘ 4/ '\ pnngs Too/E‘N EXPLANATION / EARTNOUAKE EPICENTERS 1 © M = 6.4 O M = 4,5— 5.7 0 M=4.0-44 Cracks of 10-69 I M :15 — 3.9 Plaster City 0 M 23.0 — 3. 4 3% FIGURE 7.— Epicenters of shocks of magnitude 3.0 and greater that occurred between April 9, 1968 and April 28, 1969. All shocks listed in table 2 are included, except for those with “D”—qua1ity locations. Of the 126 epicenters shown, 69 were located by Caltech and 57 by the U.S. Geo- logical Survey (Hamilton, this volume). Heavy solid lines are accurately located faults; dotted lines are concealed faults; the zigzag line is the approximate trace of the 1968 surface break along Coyote Creek fault (Clark, “Surface Rupture Along the Coyote Creek Fault,” this volume). that occurred during the entire aftershock period, including those aftershocks that followed the removal of the U.S. Geological Survey equipment in June. ACKNOWLEDGMENTS This study was supported by National Science Foundation Grants GA—1087 and GA—12868 to the California Institute of Technology. The authors ap- preciate the help of J. N. Brune, G. F. Davies, C. F. Richter, and W. T. Thatcher, and the critical com- ments of R. M. Hamilton and R. V. Sharp. SEISMIC HISTORY The San J acinto fault zone, of which the Coyote Creek fault is a member (Sharp, this volume), has been the locus of repeated moderate seismic activity within the entire historical record (Allen and others, 1965). Assuming that the zone extends southeast to the Gulf of California, at least 10 shocks of magni- 18 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 tude 6—7 have occurred along this line since 1912, and the fault zone is better delineated by seismic activity in this magnitude range than is any other individual fault in California. Many of the epicenters have been about equidistant along the fault (fig. 5), and the 1968 epicenter lies roughly midway between the epicenters of the 1942 Lower Borrego Valley earthquake (M:6.5) and the 1954 Santa Rosa Mountains earthquake (M:6.2). HYPOCENTRAL DETERMINATIONS Table 2 is a listing of all shocks of magnitude (ML) 3.0 and greater that occurred from the time of the foreshock at 0227 G.m.t. on April 9, 1968 to the time of the Coyote Mountain earthquake on April 28, 1969, about 1 year later. The list was terminated with the Coyote Mountain earthquake TABLE 2. —Earthquakes of magnitude 3.0 and greater in the area from 32°45 to 33°30 N. and from 115° to 116°30' W. from April 9, 1968 through April 28, 1969 [“Q” indicates quality, as discussed in text. Depths indicated by “R” were restricted to that depth in the computer solution. Source “T” is California Institute of Technology; source “G” is U.S. Geological Survey. Quadrangle names are those of USGS 15-minute quadrangles. All magnitudes are assigned by Pasadena. A-. B-, and C-quality epicenters are shown in the maps of figures 7 and 9] Time A 3 '5 3: z V >, 5 V ._. v M L: 'E 5 8 Quadrangle Y M D H M s ‘5 5 g 8 5‘ ’5 “ ° ‘1 .2 t: a 2 q 2 68 04 09 02 27 36.7 33 10.5 116 07.3 B 3.7 4.8 T Borrego Mountain 68 04 09 02 28 59.1 33 11.4 116 07.7 A 6.4 11.1 T Do. 68 04 09 02 33 09.0 33 10.0 116 07.0 D 4.3 T Do. 68 04 09 02 36 47.0 33 10.0 116 07.0 D 3.7 T Do. 68 04 09 02 39 30.0 33 10.0 116 07.0 D 4.4 T Do. 69 04 09 02 44 48.0 33 10.0 116 07.0 D 3.6 T Do. 68 04 09 02 47 48.5 33 05.3 116 05.7 C 3.7 T Do. 68 04 09 03 01 43.0 33 10.0 116 07.0 D 3.9 T Do. 68 04 09 03 03 53.5 33 06.8 116 02.2 C 5.2 T Do. 68 04 09 03 08 39.9 32 49.6 116 03.3 C 3.5 T Carrizo Mountain 68 04 09 03 22 22.0 33 10.0 116 07.0 D 3.3 T Borrego Mountain 68 04 09 03 48 10.3 33 06.3 116 02.2 C 4.7 T Do 68 04 09 03 58 36.0 33 03.3 115 59.6 C 4.3 T Kane Spring 68 04 09 04 05 05.3 33 12.8 116 11.3 C 3.7 T Borrego Mountain 68 04 09 04 15 47.5 33 06.7 116 01.5 B 3.6 T Do. 68 04 09 04 29 56.9 33 09.4 116 04.7 C 3.2 T Do. 68 04 09 04 46 50.0 33 10.0 116 07.0 D 3.0 ...... T D0. 68 04 09 05 00 54.7 33 03.9 116 01.4 B 3.7 1.3 T Do. 68 04 09 07 20 47.2 32 59.6 116 04.6 C 3.2 5.0R T Carrizo Mountain 68 04 09 07 35 46.3 33 11.5 116 01.9 C 3.5 2.8 T Borrego Mountain 68 04 09 07 36 23.0 33 10.0 116 07.0 D 3.6 ...... T Do. 68 04 09 07 38 21.7 33 09.9 116 07.0 C 3.1 10.0R T Do. 68 04 09 07 40 46.8 33 07.2 116 05.5 C 3.2 10.0R T Do. 68 04 09 08 00 38.5 33 06.4 116 00.4 C 4.0 4.2K T Do. 68 04 09 08 02 26.0 33 10.0 116 07.0 D 3.5 ...... T D0. 68 04 09 08 27 22.6 33 14.2 116 16.2 B 3.6 2.8 T Borrego 68 04 09 08 43 52.1 33 20.8 116 13.0 B 3.4 7.6 T Rabbit Peak 68 04 09 09 26 26.1 33 05.2 116 03.1 B 3.8 3.0 T Borrego Mountain 68 04 09 09 38 33.0 33 14.1 116 16.0 C 4.0 ' 5.2 T Borrego 68 04 09 10 42 08.9 33 10.6 116 07.7 C 3.1 10.0R T Borrego Mountain 68 04 09 11 11 49.6 33 13.1 116 09.1 C 3.1 10.0R T D0. 68 04 09 11 17 54.5 33 06.2 116 03.6 B 4.0 4.8R T Do. 68 04 09 12 20 01.0 33 15.4 116 15.0 B 3.6 3.4 T Rabbit Peak 68 04 09 12 24 28.0 33 10.0 116 07.0 D 3.0 ...... T Borrego Mountain 68 04 09 15 25 17.7 33 08.1 116 03.8 C 3.2 10.0R T D0. 68 04 09 16 20 56.3 33 13.2 116 12.5 C 3.4 10.0R T Do. 68 04 09 17 25 36.7 33 07.4 116 03.5 B 3.5 —0.6 T Do. 68 04 09 18 31 03.8 33 18.9 116 18.3 B 4.7 12.6 T Clark Lake 68 04 09 19 32 46.1 33 02.2 116 02.5 C 3.0 10.0R T Borrego Mountain 68 04 09 21 39 04.1 33 06.9 116 06.9 C 3.3 5.0R T Do. 68 04 10 00 01 11.3 33 06.8 116 04.5 B 3.7 2.2 T D0. 68 04 10 00 17 41.6 33 10.2 116 05.2 C 3.3 10.0R T Do. 68 04 10 04 05 12.6 33 06.6 116 04.2 C 3.4 10.0R T Do. 68 04 10 05 26 28.0 33 01.5 116 00.1 C 3.2 10.0R T Do. 68 04 10 05 33 52.7 33 04.7 116 06.3 C 3.1 10.0R T Do 68 04 10 06 05 12.6 33 13.8 116 16.3 C 3.2 10.0R T Borrego 68 04 10 09 22 57.5 33 08.3 116 05.4 C 3.5 10.0R T Borrego Mountain 68 04 10 13 33 04.9 33 15.5 116 05.0 B 3.9 0.4 T Rabbit Peak 68 04 10 19 11 06.3 33 13.9 116 10.8 B 3.3 4.6 T Borrego Mountain 68 04 10 23 57 58.0 33 02.7 116 00.1 C 3.0 10.0R T Do. TABLE 2. ~Earthquakes of magnitude 3.0 and greater in thc area from.32°/.5' to 33°30’ N. and from 115° to 116°30’ W. from April .9, 1968 through April 28, 1969—Continued . A. a) 3 Time ; E h E :1 * # w a; :2- E :5 3 Quadrangle a; 5 s 2: a 2 Yr Mo Da H M S q a O z D [2 68 04 11 01 12 24.1 33 13.0 116 03.3 B 3.2 5.2R T Borrego Mountain 68 04 11 11 00 24.4 33 11.3 116 10.0 C 3.0 10.0R T Do. 68 04 11 15 56 32.0 33 10.0 116 07.0 D 3.5 ...... T Do. 68 04 11 15 56 57.0 33 10.0 116 07.0 D 3.7 ...... T Do. 68 04 11 17 01 51.0 32 58.1 116 15.1 B 3.5 5.2 T Mount Laguna 68 04 11 22 28 23.7 33 06.2 116 02.9 B 3.4 —0.8R T Borrego Mountain 68 04 12 01 37 01.0 33 10.1 116 04.2 A 3.4 5.0 G D0. 68 04 12 11 29 50.2 33 18.3 116 16.3 C 3.1 10.0R T Clark Lake 68 04 12 13 42 49.5 33 16.9 116 15.8 A 3.3 2.6 G Do. 68 04 12 15 26 45.5 33 01.0 115 58.6 A 3.2 7.9 G Kane Spring 68 04 12 21 02 59.1 33 04.4 116 02.6 A 3.1 6.8 G Borrego Mountain 68 04 13 01 23 49.3 33 16.6 116 14.2 A 3.3 1.3 G Rabbit Peak 68 04 13 08 27 05.9 33 06.0 116 02.9 A 3.2 1.5 G Borrego Mountain 68 04 13 10 05 18.2 33 03.3 116 01.0 A 3.2 2.4 G Do. 68 04 13 18 23 24.5 33 06.4 116 03.6 A 3.3 8.7 G Do. 68 04 14 01 23 04.2 32 57.8 116 15.0 A 3.2 11.2 G Mount Laguna 68 04 14 01 26 18.3 33 08.5 116 05.9 A 3.0 4.9 G Borrego Mountain 68 04 14 12 55 58.7 33 14.2 116 11.4 A 4.3 10.8 G Do. 68 04 14 16 46 30.4 33 09.4 116 06.0 A 3.0 4.9 G Do. 68 04 15 10 07 11.6 33 18.9 116 13.3 A 3.3 3.0 G Rabbit Peak 68 04 15 12 28 07.5 33 17.4 116 15.6 A 3.5 3.0 G Clark Lake 68 04 15 21 56 50.3 33 02.7 115 59.3 A 3.4 7.3 G Kane Spring 68 04 15 22 07 06.1 33 02.9 115 59.0 A 3.3 8.2 G Do. 68 04 16 03 30 30.0 33 02.9 115 59.2 A 4.8 8.3 G Do, 68 04 17 02 43 47.5 33 03.2 115 59.6 A 3.7 7.9 G Do. 68 04 17 03 14 25.8 33 14.0 116 11.3 A 3.4 11.4 G Borrego Mountain 68 04 17 11 16 26.9 33 16.8 116 14.9 B 3.0 5.0 T Rabbit Peak 68 04 20 09 37 37.6 33 22.0 116 14.6 A 3.2 10.1 G o. 68 04 24 09 03 12.5 33 08.0 116 05.3 A 3.4 5.7 G Borrego Mountain 68 04 24 22 09 54.2 33 03.4 116 01.7 A 3.0 1.9 G Do. 68 04 27 09 32 30.0 33 11.9 116 09.4 A 3.0 5.1 G Do. 68 04 27 19 09 10.4 33 15.1 115 58.3 A 3.1 0.3 G Durmid 68 04 29 00 52 51.2 33 06.9 116 04.7 A 3.0 3.3 G Borrego Mountain 68 05 01 23 53 37.2 33 15.1 115 58.0 A 3.0 0.3 G Durmid 68 05 02 00 55 30.0 33 17.4 116 15.6 A 3.5 6.6 G Clark Lake 68 05 02 23 19 26.0 33 09.0 116 04.8 A 3.1 7.6 G Borrego M0untain 68 05 03 12 15 40.3 33 01.8 115 59.8 A 3.1 3.8 G Kane Spring 68 05 06 10 53 36.9 33 02.0 116 00.1 A 3.5 9.2 G Borrego Mountain 63 05 06 17 31 47.6 33 02.4 115 56.9 A 4.0 6.7 G Kane Spring 68 05 07 07 56 37.8 33 16.3 116 06.0 A 3.3 2.9 G Rabbit Peak 68 05 08 16 23 38.1 33 11.5 116 09.8 A 3.3 3.1 G Borrego Mountain 68 05 08 22 30 06.3 33 10.7 116 07.6 A 3.0 6.9 G Do. 68 05 09 10 21 46.2 33 06.1 116 01.9 A 3.6 7.3 G Do. 68 05 10 04 32 25.2 33 02.1 116 02.4 A 3.4 5.3 G Do. 68 05 10 05 28 07.8 33 02.0 116 02.4 A 3.3 4.8 G Borrego Mountain 68 05 11 08 10 04.0 33 02.4 116 00.3 A 4.2 8.8 G Do. 68 05 11 08 46 03.6 33 02.4 116 00.3 A 3.5 8.5 G Do. 68 05 11 10 38 31.7 33 16.7 115 58.1 A 3.0 1.3 G Durmid 68 05 12 14 24 16.3 33 27.7 116 24.4 A 3.3 6.3 G Clark Lake 68 05 13 18 33 52.2 33 18.6 116 15.7 A 3.3 8.5 G o. 68 05 20 04 04 17.9 33 04.3 116 08.2 A 3.2 8.7 G Borrego Mountain 68 05 21 14 34 31.3 33 04.2 116 01.6 A 3.2 1.1 G D0. 68 05 21 23 19 39.4 33 02.1 115 59.6 A 3.0 4.8 G Kane Spring 68 05 22 13 26 55.4 33 18.6 116 13.4 A 4.4 7.5 G Rabbit Peak 68 05 26 06 46 28.1 33 02.4 116 02.4 A 3.2 4.4 G Borrego Mountain 68 05 31 00 29 02.1 33 03.6 115 55.8 A 3.2 3.5 G Kane Spring 68 06 02 00 02 43.9 33 02.6 116 00.2 A 3.3 7.2 G Bor‘l‘ego Mountain 68 06 03 09 10 22.4 33 11.7 116 09.7 A 3.3 7.7 G Do. 68 06 04 03 48 19.7 33 02.4 116 02.3 A 3.2 5.1 G Do. . 68 06 06 09 12 02.0 33 00.6 115 55.7 A 3.2 6.4 G Kane Spring 68 06 06 13 18 04.9 33 00.8 115 55.6 A 3.1 7.6 G Do. 68 06 08 21 37 00.9 33 15.9 116 02.5 A 3.1 1.3 G Rabbit Peak 68 06 09 14 20 15.6 33 16.8 115 59.5 A 3.1 2.9 G Durmid . 68 06 09 14 48 29.6 33 04.6 115 55.5 A 3.0 5.1 G Kane Spring 68 06 11 05 32 17.3 33 20.9 116 20.0 A 3.1 10.1 G Clark Lake 68 06 14 16 38 12.3 33 02.1 116 00.7 B 3.1 5.0R T Borrego Mountain 68 06 26 21 35 11.3 33 17.5 115 59.4 B 3.0 5.0R T Durniid 68 06 26 21 38 23.3 33 15.8 116 00.7 B 3.1 —0.6 T Rabbit Peak . 68 08 03 10 02 18.7 32 57.7 116 12.7 B 3.4 2.3 T Carrizo Mountain 68 08 12 07 07 07.7 32 57.8 116 01.4 B 3.0 4.5 T Do. 68 08 14 09 19 21.8 33 18.5 116 12.5 B 3.7 —0.7 T Rabbit Peak 68 08 22 19 35 18.7 32 58.3 115 50.1 C 3.3 —-2.0 T Plaster City . 68 09 29 04 06 07.4 33 06.6 116 01.6 C 3.3 5.0R T Borrego Mountain 68 10 07 14 15 21.6 33 17.4 116 25.6 C 3.6 10.0R T Clark Lake 68 10 19 03 50 18.6 33 01.4 115 59.4 B 3.0 —1.3 T Kane Spring . 68 10 23 19 54 42.7 33 04.3 116 02.4 B 3.1 —1.1 T Borrego Mountain 68 10 28 11 51 55.7 33 04.4 116 01.5 B 3.2 —-0.8 T Do. 68 10 28 23 53 13.0 32 59.0 116 17.0 B 3.1 10.0R T Mount Laguna . 68 10 31 04 04 44.0 33 03.5 116 02.2 B 3.1 10.0R T Borrego Mountain 68 11 05 11 50 50.6 33 26.0 116 26.8 B 3.0 6.6 T Clark Lake . 68 11 10 07 42 00.1 33 13.1 116 04.3 B 3.0 10.0R T Borrego Mountain 68 11 23 10 33 50.6 33 20.7 116 14.1 C 3.2 10.0R T Rabbit Peak 68 11 28 02 29 58.0 32 59.8 116 15.1 C 3.2 10.0R T Mount Laguna . 68 12 14 22 30 30.3 32 53.6 116 13.8 B 3.4 10.0R T Carrizo Mountain 68 12 17 22 53 51.2 33 02.7 115 51.8 B 4.7 8.0R T Kane Spring 69 01 30 16 08 51.8 33 03.8 116 17.4 C 3.1 ~2.0 T Borrego . 69 03 29 06 57 26.8 32 57.2 116 13.6 C 3.4 —2.0 T Carrizo Mountain 69 04 28 23 20 42.9 33 20.6 116 20.8 B 5.8 20.0R T Clark Lake _ THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 19 (M:5.8) because it was larger than any earlier aftershock; it occurred somewhat farther northwest than most earlier aftershocks (fig. 7), and it had numerous local aftershocks of its own. Table 2 is thought to represent a relatively homogeneous cov- erage of earthquakes of magnitude 3.0 and greater during the 1—year period, except for a period of about 2 hours following the main shock, when some small aftershocks could have been missed because of the congested records. For the purpose of this study, aftershocks were arbitrarily assumed to include all shocks that occurred within the area of figure 7. This area includes the following US. Geological Survey 15-minute quadrangles (from left to right, top to bottom, in fig. 7) : Clark Lake, Rabbit Peak, Durmid, Borrego, Borrego Mountain, Kane Spring, Mt. La- guna, Carrizo Mountain, and Plaster City. The establishment of precise time correction fac- tors for the permanent stations of the Caltech net- work was made possible by the presence of the 20-station USGS (US. Geological Survey) array in the epicentral area (Hamilton, this volume) and by the calibration of this array with three widely sepa- rated explosions (Hamilton, 1970). Fifty-five of the largest USGS-located shocks that occurred during the 2-month period between April 12, 1968 and June 12, 1968 were used to establish station time correc- tions to be applied to the Caltech computer-location program (Nordquist, 1962). Average time correc- tions for P-arrivals of the aftershocks at the nearby stations are given in table 3; only those stations with time corrections shown were used in the routine TABLE 3. — Borrego Mountain earthquake data for stations of the Pasadena network P-wave Time Time Distance Fr 1: Station arrival time, correc- correc- to hypo- molt': main shock tion, tion, center, m 1. n, (G.m.t.) main after- main billet]: Code Name shock shocks shock (5 H ) (s) (s) (km) “1““ Y BAR Barrett ........... +0.3 +0.1 76.8 D CLC China Lake... 320.5 CWC Cottonwood... . . 401.7 ECC El Centro .................. . .0 70.2 C ELW Ella Wash (temporary) .............. — .4 15.8 FCR Fish Creek Range (temporary) .............. — .2 21.9 44.2 315.9 I) 17.6 +1.2 + .7 122.9 C 34.4 + .9 + .2 242.0 D 11.7 — .5 — .5 74.2 D 47.6 348.4 D MWC Mount Wilson... 30.8 — .1 —1.1 212.0 C OBB Obsidian Butte (temporary) ........ 07.5 — .6 — .6 47.3 OCT 0cotillo Wells ’ ...... -— .3 11.9 C 11.1 — .2 — .1 41.8 C Pasadena. 30.0 —1.1 216.8 C RVR Riverside... 22.0 + .9 + .5 146.4 D SCI San Clemen Island ................. 32.1 + .6 + .2 226.7 C SBC Santa Barbara. .. 49.2 ........ C ’ .. 282.6 385.4 C 469.0 D 373.6 D locations of aftershocks. These corrections Were used to determine the Caltech hypocenters listed in table 2, except for the main shock, which is discussed separately below. Even with these corrections, not all shocks could be assigned good hypocenters, mainly because of poor arrivals masked by contemporaneous events or wind noise or because of sporadic station failures. “B” quality epicenters are thought to be accurate within 5 km, “C” quality within 15 km, and “D” quality greater than 15 km, although the quality assignments in table 2 are generally conserv— ative. “A” quality locations, such as those of the main shock and the US Geological Survey epicen- ters, are the subject of special investigation and indi- vidual error assignments. A measure of the accuracy of hypocentral determinations from the Caltech stations alone (“T” source in table 2), when the above corrections are applied, can be determined by comparing the depths assigned to all Caltech “B” hypocenters that were independently located by both networks. The average difference in depth assign- ments is 2.0 km; this amount of difference suggests that most of the “B” locations in table 2 are indeed accurate to within 5 km and many are much more accurate. FORESHOCK ACTIVITY A single foreshock of magnitude 3.7 preceded the main shock by about 1 minute, and its hypocenter can be considered identical with that of the main shock within the limits of location error. Perhaps the most significant feature of the foreshock activity, however, is the complete absence of any other fore- shocks or preceding regional activity. Records of the 47—km-distant station at Obsidian Butte, for ex- ample, show no hint of any other earthquakes in the epicentral region within the previous few hours and days, and the general level of activity at this station had been unusually low during the preceding 4 months. Within the year preceding the main Borrego Mountain shock, only two earthquakes that might be considered within the zone of subsequent after- shock activity along the Coyote Creek fault are listed in the Caltech Local Bulletin. MAIN SHOCK LOCATION The main shock was located by using correction factors that were based on arrival-time residuals of only the 11 largest shocks that were recorded by both the USGS and Caltech networks — all of these shocks exceeded magnitude 3.4. The restriction to larger shocks permits a more direct comparison with the main shock, particularly for the more distant stations. These time corrections, as well as the 20 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 P—wave arrival times used in the solution, are given in table 3; only those stations with time corrections shown in table 3 were used in this solution. The computed hypocenter is 33°11.4’ N., 116°07.7’ W., h:11.1 km. Standard errors in the computer solution of the x, y, and 2 components are 0.65, 0.96, and 2.01 km, respectively, but these are felt to be unrealistic in view of the somewhat arbitrary assignment of correction factors. On the basis of numerous at- tempts to locate the epicenter using a wide variety of assumptions and models, we believe that the com- puted epicenter is correct to within 3 km, but be- cause the nearest station (OBB) is 47 km away, depth control is poor. Nevertheless, the compara- tively large hypocentral depth of the main shock relative to that of most aftershocks is consistent with the observations of Hamilton (this volume), who notes that all A- and B-quality USGS hypocenters of shocks exceeding magnitude 3.4 are at depths below 6 km; their average depth is 8.06 km, com- pared to an average depth of 4.97 for the hypocenters g of all A- and B—quality aftershocks listed in table 8. MAGN ITUDE A local magnitude (ML) of 6.4 has been assigned to the Borrego Mountain main shock by C. F. Rich— ter, who used records of seven widely spaced stations of the Pasadena network. Most of the Wood-Ander— son torsion seismometers of the network were driven off—scale by the main shock; to determine the magni- tude it was necessary to compare the main shock with a well-recorded aftershock by using low-magni- fication seismometers (for example, 100x). The magnitude 5.2 aftershock of 0304, April 9, was re- corded on 12 torsion instruments and on 17 low- magnification instruments that remained on-scale during the main shock; comparing the amplitudes on the various records is consistent with a magnitude assignment of 6.4 for the main shock. Berkeley re- ported a magnitude of 7.1—7.2 for the same shock (Niazi and others, 1969), and the US. Coast and Geodetic Survey PDE2 card reported a magnitude of 6.1 from the magnitude-determination system in use at that time. FOCAL MECHANISM The focal mechanism for the Borrego Mountain main shock is shown in figure 8. The focal mechanism was solved by following the technique of Sykes (1967) ; a focal depth of 10 km and a crustal velocity, of 6.0 km per sec were assumed. The two nodal planes are both relatively well defined and must be near vertical. It is reasonable to assume that the fault is represented by the nodal plane that strikes 2Preliminary determination of epicenter. 3 FIGURE 8.— Focal-mechanism solution for the Borrego Moun- tain main shock. Diagram is an equal-area projection of the lower hemisphere of the radiation field. Solid symbols repre- sent compressions; open symbols, dilatations; square sym- bols, stations of the Pasadena network; circular symbols, distant stations of the Worldwide Standard Seismograph Network (WWSSN) and Canadian network; crosses, wave character indicating stations near nodal plane. 9 and 8 are strike and dip of nodal planes, and arrows indicate sense of shear on the plane chosen as the fault plane. northwest and dips steeply to the northeast, at least 80°, and that has a right-lateral displacement of almost purely strike-slip character. The only incon- sistencies in the solution are from nearby stations of the Pasadena network (table 3; square symbols in fig. 8), where there is a discrepancy between the first motions at Fort Tejon and Riverside compared with those at Berkeley. The more distant long-period arrival at Berkeley, which was clearly compressional, is given precedence in the solution. Although 15 of the 17 nearby stations are clearly consistent with the more distant stations, minor inconsistencies in the nearby stations are not surprising, particularly in View of the very complicated crustal structure in the epicentral region. Stations of the Pasadena net- work that are presumed to have received direct P arrivals are arbitrarily plotted at the edge of the net in figure 8. The fault-plane strike of 13220 (N. 48° W.) corre— sponds within a few degrees to the trend of major aftershock activity (fig. 7) and to the trend of sur- face faulting (Clark, “Surface Rupture Along the Coyote Creek Fault,” this volume). The steep dip, however, seems to preclude the possibility that the THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 21 apparent offset of the line of major aftershocks 2—3 km northeast of the trace of surface fracturing can be caused by dip of the fault surface. Such a hypothe- sis would require a much shallower dip of the fault plane (for example, 70°) than is permitted by the focal-mechanism solution; furthermore, this hy- pothesis is not suggested by most of Hamilton’s vertical cross sections (fig. 16). Late in the aftershock period, a number of events occurred near the southeastern extremity of faulting that may be related to the creep history of surface displacement. During the aftershock period, little or no creep occurred on the northwestern segment of the surface trace (north of Highway 78), but con-‘ siderable creep took place along the southeastern segment (Clark, “Surface Rupture Along the Coyote Creek Fault,” this volume). When one compares the relative aftershock activity in the two areas, it seems reasonable that at least some of the creep in the southeastern segment may have been associated with the much more numerous aftershocks in this area, such as the increased activity that took place here shortly after April 14 (fig. 9C). EPICENTRAL DISTRIBUTION OF AFTERSHOCKS AS A FUNCTION OF TIME Inasmuch as it was not possible to assign accurate epicenters to a number of larger aftershocks that occurred within a few hours of the main event (table 2), arguments as to possible changes of after— shock activity with time are necessarily somewhat limited. Within 2 hours of the main shock, it is. likely that aftershocks had occurred essentially over the entire length of the surface rupture as it was mapped in the days following the event. (Note, how- » ever, that most of the initial epicenters listed in table 3 are of C-quality.) Nevertheless, there is a definite suggestion of gradual enlargement of the zone of aftershock activity in the days and weeks following the earthquake. Figure 9 shows the distri- bution of epicenters during four successive time periods having roughly equal numbers of shocks of magnitude 3.0 and above, and the progressive expan- sion of areal activity is obvious, particularly late in the aftershock period (fig. 9D). For example, most of the numerous events near the southeast end of the fault trace, near 33°02’ N. 116°00’ W., occurred after April 14 (fig. 90). Likewise, the two outlying areas of minor activity situated symmetrically with respect to the fault trace near Salton City and Agua Caliente Springs (fig. 7) became active primarily after April 27 (fig. 9D). Two of the largest aftershocks occurred near the extreme ends of the aftershock zone many months after the main event; these are the Coyote Mountain earthquake (M:5.8) of April 28, 1969, 476-246 0 - 72 - 3 which may not be an aftershock in the usual sense, and the shock of December 17, 1968 (M:4.7) , which lies almost on the trace of the Superstition Hills fault (fig. 7, 9D). It is also significant that during 'the late period, when aftershocks were occurring over a wide area, almost no shocks exceeding magni- tude 3.0 occurred in the area of original surface faulting except at its southeast end. Thus, the late— period aftershock distribution takes on almost a doughnut shape. TECTONIC IMPLICATIONS The hypocentral distribution of the Borrego Moun- tain earthquakes leads to several conclusions of tec- tonic significance: (1) Aftershocks are not limited to a single fault plane; instead they are distributed over a wide zone—in sharp contrast to the after- shock distribution of the Parkfield-Cholame earth- quakes (Eaton, 1967) ; (2) the zone of most concentrated aftershock activity parallels but is somewhat northeast of the line of surface rupture along the Coyote Creek fault; (3) the epicenters of all shocks above magnitude 4.5, including that of the main shock, seem to lie along a single line 2—3 km northeast of the Coyote Creek fault trace, with the exception of the shock of December 17, 1968 that is near the Superstition Hills fault; (4) the zone of rupturing at depth evidently extends farther north- west than does the zone of surface faulting; and (5) ,the epicenter of the main shock lies roughly midway along the zone of aftershock activity and thus seems to reflect bilateral faulting; the fracture propagates both northwest and southeast from the point of initial rupture. As previously discussed, it seems unlikely that the dip of the fault plane is so shallow that the surface break represents the same fault plane as that indi- cated by the hypocenters of the larger aftershocks. Nor, in the light of the careful explosion calibration carried out by Hamilton (1970; this volume), is it likely that there is a major systematic error in the epicentral locations of the aftershocks. The seeming lack of direct correlation between the trace of surface faulting and the aftershock distri- bution may be related to the complex local geology. It is obvious from the geologic map (fig. 7) that the San J acinto fault zone in this region (as opposed to the San J acinto fault itself) is not a simple fracture —not nearly so simple, for example, as the San Andreas fault in the Parkfield-Cholame area. The zone comprises many individual Quaternary breaks: northwest of the 1968 epicenters are the Buck Ridge, San Jacinto, and Coyote Creek faults, and to the ‘southeast are the Superstition, Hills and Superstition Mountain faults, as well as a possible extension of 22 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 .dmzin shock \ Q 4-9-68 4—10—68 through 4-14-68 (first 22 hrs.) 0 O . . o 0 .. 9. ° 2 © 4-15-68 through 5-24—68 5-22-68 through 4-28-69 M=‘3.o-3.4 M=3.4-3.9 M=4.0—4.4 M=> 4.5 O ’10 20 3O 4O 5O L_.L..___.L_.L_.L_l Miles 0 50 '—*—"—L—-‘-—’ Kilometers FIGURE 9,—Epicenters of figure 7, separated into four consecutive time periods (A to D) containing roughly equal numbers of shocks. Boundaries and coordinates of individual maps are the same as those of figure 7. THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 23 the Coyote Creek fault, which is represented by northwest-trending cracks across Highway 80, 4 km east of Plaster City, that were first observed in October of 1969 (fig. 7). Therefore it is not surpris- ing that a major earthquake within the zone is a complex event, with aftershocks distributed over a number of individual breaks. Indeed, the aftershock zone of the 1968 event alines rather well with the center of the entire band of Quaternary faulting along the San J acinto fault zone, whereas the surface faulting lies at the southwest edge of it. However, the question of why the surface faulting is apparent- ly localized only along the margin of the fractured zone remains unanswered. It is interesting that although the Borrego Moun- tain and Parkfield-Cholame earthquakes are of some- what comparable magnitude and displacement, there are significant mechanical differences between them. In addition to the markedly different patterns of aftershock distribution, Max Wyss (oral commun, 1970) has pointed out that the apparent stress at all depths is higher at Borrego Mountain than at Park- field-Cholame; average values for 12 events in the two localities differ by two orders of magnitude. Despite the fact that both earthquakes are low- stress—drop events (Wyss and Brune, 1968; Wyss and Hanks, this volume), the absolute stress at Borrego Mountain was considerably higher, which may be related to the markedly different patterns of aftershock activity. The geological conditions at Parkfield-Cholame, typified by a relatively simple fault break that was perhaps “lubricated” by the presence of abundant serpentine in the fault zone (Allen, 1968), might logically be associated with earthquakes at relatively low levels of absolute stress. In the Borrego Valley area, on the other hand, the complex and discontinu- ous fault pattern might be Visualized as necessarily demanding a relatively high absolute stress for breaking to be initiated, so that the faulting then occurred over a wider zone of pervasive fracturing. The epicenter of the Borrego Mountain main shock lies roughly midway along the zone of aftershock activity (fig. 7) and thus seems to reflect bilateral faulting— with the fracture propagating both northwest and southeast from the point of initial rupture. This relation is in marked contrast to most other recent earthquakes in California, in which the initial epicenter has been at one end of the aftershock zone, thus reflecting unilateral faulting. Among these earlier earthquakes that have been well documented are the 1933 Long Beach earthquake (Richter, 1958), the 1940 Imperial Valley earthquake (Trifunac and Brune, 1970), the 1948 Desert Hot Springs earth- quake (Richter and others, 1958), the 1952 Kern County earthquake (Richter, 1955), and the 1966 Parkfield-Cholame earthquake (McEvilly, 1966; Eaton, 1967). Indeed, the 1906 San Francisco earth- quake is perhaps the only other large earthquake in California for which there is good evidence of bilat- eral faulting (Bolt, 1968). REFERENCES CITED Allen, C. R., 1968, The tectonic environments of seismically active and inactive areas along the San Andreas fault system, in Proceedings of conference on geologic prob- lems of San Andreas fault system, Stanford, Calif., 1967: Stanford Univ. Pub. Geol. Sci., v. 11, p. 70—80. Allen, C. R., St. Amand, Pierre, Richter, C. F., and Nordquist, J. M., 1965, Relationship between seismicity and geologic structure in the southern California region: Seismol. Soc. America Bull., v. 55, p. 753—797. Bolt, B. A., 1968, The focus of the 1906 California earthquake: Seismol. Soc. America Bull., v. 58, p. 457—471. Eaton, J. P., 1967, Instrumental seismic studies, in Brown, R. D., and others, The Parkfield-Cholame, California, earthquakes of June-August 1966: US. Geol. Survey Prof. Paper 579, p. 57—66. Hamilton, R. M., 1970, Time-term analysis of explosion data from the vicinity of the Borrego Mountain, California, earthquake of 9 April 1968: Seismol. Soc. America Bull., v. 60, p. 367—381. McEvilly, T. V., 1966, Preliminary seismic data, June-July 1966, in Parkfield earthquakes of June 27—29, 1966, Mon- terey and San Luis Obispo Counties, California—prelimi- nary report: Seismol. Soc. America Bull., v. 56, p. 967—971. Niazi, M., Somcrville, M., and Dengler, L., 1969, Earthquakes and the registration of earthquakes from January 1, 1968, to June 30, 1968: California Univ., Berkeley, Seismog. Sta. Bull., v. 38, no. 1, p. 1—161. Nordquist, J. M., 1962, A special-purpose program for earth- quake location with an electronic computer: Seismol. Soc. America Bull., v. 52, p. 431—437. Richter, C. F., 1955, Foreshocks and aftershocks, in Earth- quakes in Kern County, California, during 1952: Califor— nia Div. Mines Bull. 171, p. 177—197. 1958, Elementary seismology: San Francisco, W. H. Freeman and Co., 768 p. Richter, C. R, Allen, C. R., and Nordquist, J. M., 1958, The Desert Hot Springs earthquakes and their tectonic en- vironment: Seismol. Soc. America Bull., v. 48, p. 315—337. Sykes, L. R., 1967, Mechanism of earthquakes and nature of faulting on the mid-ocean ridges: Jour. Geophys. Re- search, v. 72, p. 2131—2153. Trifunac, M. D., and Brune, J. N., 1970, Complexity of energy release during the Imperial Valley, California, earthquake of 1940: Seismol. Soc. America Bull., v. 60, p. 137—160. Wyss, M., and Brune, J. N., 1968, Seismic moment, stress, and source dimensions for earthquakes in the California-Ne- vada region: Jour. Geophys. Research, v. 73, p. 4681—4694. SOURCE PARAMETERS OF THE BORREGO MOUNTAIN EARTHQUAKE1 By MAX W YSS LAMONT-DOHERTY GEOLOGICAL OBSERVATORY OF COLUMBIA UNIVERSITY AND THOMAS C. HANKS SEISMOLOGICAL LABORATORY, CALIFORNIA INSTITUTE OF TECHNOLOGY ABSTRACT Spectral analysis of teleseismic body phases at several azimuths was used to determine the-moment, fault length, dis- location, stress-drop, and radiated energy of the Borrego Mountain earthquake. The results agree well with the same parameters obtained from the surface fracture and after- shock distribution and local observations of radiated energy. INTRODUCTION The Borrego Mountain earthquake provided an ideal opportunity to check the reliability of tele- seismic methods to estimate the moment, dimen- sion, and dislocation of a seismic source. Although the event was large enough to be recorded by the WWSSN (Worldwide Standard Seismograph Net— work) at all epicentral distances, it was small enough to be recorded well on the Wood-Anderson instruments (magnification = 100) operating at Pasadena and Riverside. More important, this earthquake produced a well-defined surface rup- ture with a measured right-lateral displacement (Allen and others, 1968; Clark, this volume). De- tailed studies of the aftershock sequence provided further information on the extent of the source region (Hamilton, this volume). In the equivalent double-couple representation of a seismic source (Burridge and ‘Knopoff, 1964), the seismic moment Mo (Maruyama, 1963; Haskell, 1964) was shown by Aki (1966) to be proportional to the product of the fault, area A and the average dislocation E, y . ’ M0 = ,TAZ (1) where p. is the shear modulus in the source region. With this relation, M0 can be evaluated from field measurements of Z and A, the latter calculated 'Contribution 1819. Lamont-Doherty Geological Observatory and Contribution 1930, Division of Geological Sciences, California Institute of Technology, Pasadena, Calif. from rupture length and depth based on field map- ping and aftershock studies. Provided that the dislocation theory models an earthquake source correctly, M0 can be determined from the long- period amplitude spectral density (00) of body or surface waves. The two independently obtained values of M0 will be compared in order to check the validity of the dislocation theory. Previous com- parisions of teleseismically estimated Mo with that obtained from fault area and surface dis- placement suffered from considerable uncer- tainties in the field observations (Aki, 1966; Wyss and Brune, 1968). Another important parameter that can be esti- mated from teleseismic spectral data is the source dimension r. Kasahara (1957) has related the corner/peak frequencyfo of P—wave spectra to the radius of a spherical source model. Berckhemer and Jacob (1968) have estimated the rupture area of deep earthquakes using f0 of P-waves. Brune (1970) proposed a relation between f0 of S-waves and the radius of a circular rupture area. The present results are a part of a larger study, one purpose of which is to check if any one of the proposed relations between f}. and r is correct for earthquakes with known rupture length. ACKNOWLEDGMENTS Helpful discussions with J. N. Brune, C. B. Archambeau, and W. R. Thatcher are gratefully acknoWledged. We also thank L. R. Sykes, R. V. Sharp, B. Isacks, and R. M. Hamilton for critically reading the manuscript. Some of the work was completed while the senior author was at the Insti- tute of Geophysics and Planetary Physics, Univer- sity of California, 'La Jolla. This research was supported by National Science Foundation grants NSF GA—19473, NSF GA—12868, and NSF GA—22709. THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 25 DATA The vertical components of P-waves recorded at eight WWSSN locations (table 4) at distances between 37° and 86° were Fourier-analyzed; the resulting displacement spectra 0(a)) and their azimuthal distribution are shown in figure 10. In order to cover as large a frequency range as possible, the spectra of the long-period records as well as of the short-period records have been de- termined and plotted on the same graph. 0(a)), as presented in figures 10 and 11, has been corrected for the instrument response and for attenuation using the results of Julian and Anderson (1968) for an earth model MM8 (Anderson, Ben—Menahem, and Archambeau, 1965). The units of 0(a)) are given in centimeter-seconds for both P- and S-waves. . The S-wave spectra of four stations are shown in figure 11. Only one station furnished a usable short-period record. The corner frequency fl), how- ever, is clearly defined by the long-period spectra alone. In order to avoid contamination of the S phase by other phases, only stations between 63° and 75° were considered. The window length ranged from 30 to 60 seconds. P- and S-wave spectra were also recorded at local stations (Pasadena, A=220 km; and Riverside, A=145 km). The recordings were obtained from low-magnification Wood-Anderson instruments (gain=100). These data were considered useful only in the range 0.5 st 2.5 Hz. Although these data will not be useful in the moment or source dimension determinations, they provide a lower bound check for the radiated seismic energy. In order to estimate the moment and the total radiated energy, the following corrections of the spectral amplitudes measured at any station have been made. For the combined displacement ampli— fication by the free surface and the crust at the receiver, an average value of 2.5 for P and S has been assumed (Ben-Menahem and others, 1965). For the radiation pattern correction, a vertical right—lateral strike-slip source with strike azimuth of N. 48° W. was used. The decrease in amplitude due to geometrical spreading was accounted for, and the spatial integration of the energy radiated in different directions was performed using Wu’s (1966) results. Because of the presence of the free surface at the source, only half of theintegral given by Wu (1966) for the Whole space was taken for the energy integration. SEISMIC MOMENT At the long-period end, 50 2 T 2 10 sec, the spectral amplitudes are approximately constant and at shorter periods fall off approximately as 2?”. The long-period levels have been approximated as indicated in figures 10 and 11 and given as (20 in table 4. This approximation of the spectrum is in accordance with the results of dis- location theory (Aki, 1967; Brune, 1970), provided that the spectral information in this period range is a meaningful representation of the ultralong- period level. On the other hand, the lack of data for T > 50 sec admits the possibility that the chosen level may only be the maximum of a broad peak as required by Archambeau (1968). The seismic moment M0 was estimated from the cor— rected ()0 using equations for the far-field dynamic displacement by Keilis-Borok (1960) and Ben- Menahem, Smith, and Teng (1965). The moments obtained from teleseismic spectra are in good agreement with the moments esti- mated from field observations. Mo obtained from the different stations are given in table 4. a)”, where w: TABLE 4. —- P- and S— wave spectral data Long-period Corner 82:33" Dr‘st. «21:12.05; Asfaftli‘ghn Phase Component amplitude frequency Dim:ln/s£on MOE?!“ EneErgy (deg) (deg) (deg) (cmsec) (H02) (km) (10.5 dyne-cm) (10" erg) AKU ............... 63.4 27 292 S EW 2.2 ' 10“2 0.069 17.4 5.9 3.8 AKU ..... S NS 1.4 ' 10"2 .05 24.0 5 9 3.8 ESK ...... 74.9 33 307 P Z 5.0 ' 10‘4 .20 10.5 — —— ESK.. S EW 1.6 ' 10"2 .063 19.0 8 1 8.0 ESK.. S NS 2.5 ' 10‘2 .063 19.0 8.1 8.0 NAT. 85.4 99 304 P Z 1.4 ' 10'3 .22 9.5 [11.] —- TRN. 54.8 101 303 P Z 3.5 ' 10‘3 .126 16.7 11. 1.6 LPB ...... 67.4 129 318 P Z 3.3 ' 10‘3 .12 17.5 — — LPB ................. S EW 1.4 ' 10‘2 .052 23.1 7.0 3.3 LPB ...... S NS 2.0 ° 10‘2 .05 24.0 7.0 3.3 ARE ............. 65.2 132 320 P Z 3.2 ' 10’3 .16 13.1 -——— — ARE ............. S EW 1.6 ' 10'2 .063 19.0 5 5 2.5 ARE ................ S NS 2.0 ' 10‘2 .047 25.5 5.5 2.5 KIP .................. 38.8 263 63 P Z 4.0 ' 10‘3 .16 13.0 7.4 1.5 MAT ................ 81.9 308 55 P Z 1.8 ' 10‘3 .21 10.0 —— -— COL ................. 37.1 338 133 P Z 5.0 ' 10‘3 .126 16.7 12.0 1 9 26 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 ESK \: \ \ -\ ;\ -:\ '.\ \’ -4— 25$§£ .H.‘:-\ 1 1' -2 -l 0 -2 '*'\-"‘\::\ ’— \ \ \ — NAT \ \ \-‘. \. -3’- '.\ V .\. . ,.-'_\ ' :\: :33 '. ,A. ::_z\ -4— "YO -2 31 6 ' “‘ __\ \ \ \ \ TRN \ \-. \'_ \ '\ “\f: I'.\'. ‘ -\ . _\ \\ . ~5~ at ' ‘ . 1 APRIL 9, I968 . ‘2 -l O -2 -l 0 FIGURE 10,—Attenuation-corrected P-wave spectra of the Borrego Mountain earthquake. Solid line obtained from long-pe- riod vertical WWSSN instruments, and dots obtained from short-period Vertical WWSSN instruments. Azimuths with respect to strike of fault. Vertical scale is log of displacement spectra, in centimeter- seconds; horizontal scale is log frequency, in Hertz. THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 27 S(E/W) S(N/S) LOG AMPLHUDE SPECTRAL DENSFTY(CM-SEC) 0:2 LOG FREQUENCY(HZ) FIGURE 11. — Attenuation-corrected S-wave spectra of the Bor- rego Mountain earthquake from WWSSN recordings. Unfortunately, some stations were close to nodes in the P-wave radiation pattern. If the cor- rection due to the radiation pattern alone was larger than a factor of 10, the moment was not computed, since small uncertainties in the focal mechanism could introduce large errors. The average of the P- and S-wave moments of table 4 is compared with M0 derived from observations in the source area in table 5. The depth of faulting was assumed to be the depth to which the after- shock activity extended, h=12 km (Hamilton, this volume). M0, based on teleseismic (dynamic) mea- surements, agrees within approximately a factor of 2 with the field (static) observations. The teleseismically determined moments are likely to be affected by reflections from the free surface. The inclusion of pP in the spectral analysis is expected to cause an overestimate of M0. This may explain why the teleseismic moments are larger than the static ones (table 5). The long— period data, however, do not indicate the expected degradation of the radiation at periods long com- pared to the timelag At of the reflected phase (A! 2 2:? S 3 sec; h=source depth S 10 km, a=P-wave velocity=6.0 km/sec). The facts that the Borrego Mountain earthquake ruptured the sur- face and had a depth of rupture comparable to the. length of rupture undoubtedly complicates an image point source representation of reflected phases. Other error sources for individual stations are uncertainties in the fault plane solution and the effect of local upper mantle and crustal struc- ture. These two effects could combine to form an error that in general should not exceed a factor of three in the moment determination for any one station. In the average moment determination from the several stations, some of these errors should be cancelled out, and it is encouraging to see that the RMS (root mean square) errors are small (table 5). TABLE 5. —Source parameters Moment Dislo- Stress- Field observation or Mo nggth Ajea De’pth cation drop spectral estimate (1025 (k_ )r (k "1) (k' 17 A0 dyneAcm) m m m) (cm) (bar) Surface rupture .......... 3.6 33 396 ‘12 30 4 Aftershock zone ......... 4.9(6.1) 45(56) 540(672) 12 30 — P-wave (Brune) iiiiiiiiiii 10.:2 26:6.5 615 — 30 20 S-wave (Brune) ,,,,,,,,,,, 6.6:1 42:6 1460 — 14 3 P-wave (Kasahara) ..... - 8 53 — — — P-wave (Berckhemer and Jacob) ............... — 8.5 58 — — — ‘Assumed from aftershock distribution. The moments obtained from field observations will be affected by errors in the estimates of the rupture area and the dislocation. The zone of in- tense aftershock activity in. the figures shown by Hamilton (this volume) is 45 km long. If the sporadic aftershock activity toward the north is included, a length of 56 km is obtained. We feel, however, that the intense aftershock zone outlines the rupture area, whereas the more northern events were caused by strain accumulations be- yond the fault end. This view is supported by the Coyote Mountain earthquake of April 28, 1969, and 28 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 its aftershock sequence, which were located in this zone of sporadic aftershock activity, 33°20’N., 116°21'W. (R. M. Hamilton, oral commun., 1970; Thatcher and Hamilton, 1971). Even though we feel that 45 km is the value that should be used for the fault length as outlined by the aftershocks, the source parameters based on 56 km are also computed and given in table 5 in parentheses. The length of the surface rupture differs by a factor of 1.3 (1.7) from the extent of the intense (extended) aftershock activity. The depth of the main shock given by Allen and Nordquist (this volume) is 11 km, which differs from the depth of the deepest aftershocks by a factor of 1.1. These factors may combine to an error of less than 2 in the area esti- mates. Insofar as no creep occurred on the northern branch of the fault, three-quarters (Brune and Allen, 1967) of the maximum surface displacement, 38 cm (Clark, “Surface Rupture Along the Coyote Creek Fault,” this volume), was taken to represent the average displacement at depth. It is hard to imagine that the displacement could be in error by more than a factor of 3. Under this assumption, the statically estimated moment might be uncertain to a factor of 5. This is an extreme value, however, and we believe that the static moment is estimated to within a factor of 3 from the correct value. The agreement in table 5 can be considered very good, and it is concluded that the moment of a double-couple source can be determined equally well from teleseismic body phases or from field observations in the source region. FAULT LENGTH The corner frequencyfo was defined by the inter- section of the horizontal line representing the long-period spectral level and the line drawn through the short—period data, if they were avail- able. Otherwise, fi, was estimated from the decay of the long-period data. . Brune (1970) related f0 for S-waves to the source dimension r by r: 2.215 = 1.2 (2) 27Tf0 (S) f0 (S)’ where r is measured in kilometers. Because the flat amplitude level at frequencies smaller thanfo is due to interference of waves with wavelength A longer than the source size, it would seem reasonable to obtain a version of equation 2 valid for P-waves by keeping A0 constant and sub- stituting a, the P-wave velocity, for ,8, the S—wave velocity. To estimate the source dimension from the P-wave spectra, we use the relation r: 2.210; = 2.1 (3) 27Tfo (P) fo(P)' The source radii obtained from 2 and 3 and the appropriate f0 are given in table 4. The values from stations at different azimuths are generally consistent, indicating that such effects as focusing due to rupture propagation, substation crustal reverberations, and 6-second microseisms are not sufficient to disguise the gross behavior of body- wave spectra of the Borrego Mountain earthquake. Averages for the P- and S-wave determinations of L=2r and of the ruptured area A =7rr2 are com- pared to surface rupture and aftershock length in table 5. The agreement between the four length determinations is very close, if Brune’s equations are used, and it is concluded that the source dimen- sions of an earthquake can be obtained with satis- factory accuracy from spectral P— and S-wave analysis using Brune’s (1970) results. It is to be noted, however, that the source dimensions ob- tained from P-wave spectra are consistently smaller than those from S-wave spectra. This fact may indicate that the step from equation 2 to equation 3 is not completely correct and needs clarification. Kasahara’s (1957) model relates the radius of a spherical source to f}, by 0.66 " fo(P) ' (4) If the average value forfi, (P) from table 4 is used in equation 4, we obtain L=2r=8.2 km. In Berck— hemer and Jacob’s (1968) theory, the maximum rupture velocity c has to be assumed; their rela— tion is r c2 1 —E_;§;. (5) Even if a rather high rupture velocity of 3 km/sec is assumed, the average area AU from 5 is only 58 kmz. The diameter of a circular fault with this area is approximately 8.5 km. The values for A0 and r that are estimated by equations 3, 4, and 5 are compared in table 5 to the values observed in the source region. It is evident that the theories of earlier authors (Kasahara, Berckhemer, and Jacob) underestimate the source dimensions, but Brune’s relation gives results which are in excellent agreement with the field observations. The use of body waves for source size determina- tion has great advantages over the surface wave method (Ben-Menahem, 1961). The excitation of surface waves is strongly dependent on the crustal structure, and the source depth interference can be severe (Tsai, 1969). Also, for smaller events it is Ao THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 difficult to extend surface wave spectra to the short periods required without mode contamina- tion, and deep earthquakes often do not excite measurable surface waves. A disadvantage of the body-wave method is that at a range of shallow depths the surface reflections pP and sS cannot be separated from the direct phase, and the inter- ference of these phases with the direct waves can be expected to complicate the analysis. ENERGY ESTIMATES The energy in a body wave spectrum is propor- tional to j: |Q(w)w Izdw. For a theoretical spectrum as given by Haskell (1964), Aki (1967), or Brune (1970), the energy integral can be conveniently represented by 0% 1% E €th f"o' (6) The constant k contains corrections for geomet- rical spreading, free surface and crustal effects at the station, the radiation pattern, integration of the radiation pattern over the focal sphere, rate of decay of the spectrum for w>w0 (here assumed to be (0—2), as well as material constants. The energy radiated from a surface source into the halfspace below was taken to be half the integral given by Wu (1966) for a whole space. The total energy radiated by the Borrego Mountain earthquake as P- and SH-waves is given in table 4; the average teleseismic estimates are E(SH)=O.44'1020 ergs, E(P)=0.16'10'20 ergs. The teleseismic observations indicate that the total radiated energy, including SV, was approximately 0.69'1020 ergs. The radiated seismic energy has also been esti- mated from the seismograms written by the Wood- Anderson torsion seismograph operating at a gain of 100 at Pasadena and Riverside. Table 6 gives estimates of the seismic energy radiated for all components of P— and SH-waves in the frequency range 0.5 S f S 2.5 Hz. The bulk of the seismic energy, however, is not included in the above esti- mate, because the corner frequency occurs outside of the frequency range available for analysis. We can check whether the locally recorded spectra agree approximately with the teleseismic ones by TABLE 6. — Local observations ofradiated seismic energies Station and Phase Energy. 10" ergs‘ Pasadena: S(EW). .. Riverside: {Estimate based on spectral amplitudes corrected for seismic attenuation, with Q=500 in the bandwidth 0.5 g f; 2.5 Hz. 29 extrapolating the slope of the local spectra as a straight line to lower frequencies. At the average corner frequency, this straight line should inter- lsect with the long-period flat level of the spectrum, which at that particular station corresponds to the average moment. The agreement is fairly good. If we assume that the long—period level corre- sponding to the average moment, the corner fre- quency, and the slope of spectral decay to high frequencies are given, we estimate the total radi- ated energy based on the Pasadena record to be approximately 21020 ergs. The energy estimated from the energy-magni- tude relationship (Gutenberg and Richter, 1956) amounts to several times 1021 ergs, depending on the relation used. The spectral estimates dis- cussed above are an order of magnitude smaller. The energies estimated by integration agree with the energy estimated by DeNoyer (1959) for earth- quakes with comparable magnitude. The problem of energy should be studied in detail, but for this task complete spectra, including the corner fre- quency from nearby stations, are needed. An interesting result is that the ratio of the S-wave to the P-wave energy is 3.3 despite the con— siderably larger long-period S-wave amplitudes. Since the energy is a function of the ratio of amplitude over period, the larger long-period S- wave amplitudes are partly offset by a smaller fo (S). This relation is best understood in terms of equation 6, where ()0 is proportional to %, while f0 and k are proportional to v (v=respective propa- gation velocity). Therefore, the ratio of S to P wave energy should be (oz/)8)2 3. Gutenberg and Richter (1956) assumed a ratio of 2; this ratio agrees with our observation. STRESS-DROP The stress-drop Ao=ol—o-2 (01=stress prior to rupture, oz=stress after completion of the event) can be estimated from the moment and the source radius through the relation (Brune, 1970) 7 M0 A0— 16 r3 . (8) From the field observations, the stress-drop was also estimated using Knopoff’s (1958) equation Ao=%%, where u=dislocation and h=depth. The three different estimates are given in table 5. They agree well, and the average stress-drop is Ao=9 bar. This value is much smaller than the stresses that the earth’s crust is able to maintain. 30 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 From the stress-drops of large shallow earth- quakes, Chinnery (1964) estimated that the crust can accumulate 100 bar shear stresses. Brune, Henyey, and Roy (1969) obtained an upper bound of 200 bars for the stresses accumulated on the San Andreas fault, based on heat-flow measure- ments. Evidently, the Borrego Mountain earth- quake is another low stress-drop event like the Parkfield (Aki, 1967; Wyss and Brune, 1968) and the Imperial (Brune and Allen, 1967) earthquakes. This result lends further support to the observa- tion that the stress-drop of small earthquakes is in general smaller than that of large earthquakes (King and Knopoff, 1968; Wyss, 1970). CONCLUSIONS With the assumption of a double-couple source model, the use of teleseismically observed body- wave spectra has been successful in determining the fault dimension and seismic moment (and, equivalently, total offset and stress-drop) of the Borrego Mountain earthquake. The results are in close agreement with independently obtained values resulting from local field observations. The average fault length is 37 km, the average moment is 6.3 ' 1025 dyne cm, and the average stress-drop is 9 bars. The radiated seismic energy estimated from teleseismic observations is 0.69' 1020 erg and from local observations, extrapolated to include the whole frequency range, is 2'1020 erg. REFERENCES CITED Aki, K., 1966, Generation and propagation of G waves from the Niigata earthquake of June 16, 1964. Part 2. Estima- tion of earthquake moment, released energy, and stress drop from the G wave spectrum: Tokyo Univ. Earth- quake Research Inst. Bull., v. 44, p. 73—88. 1967, Scaling law of seismic spectrum: Jour. Geophys. Research, v. 72, p. 1217—1231. Allen, C. R., Grantz, A., Brune, J. N., Clark, M. M., Sharp, R. V., Theodore, T. G., Wolfe, E. W., and Wyss, M., 1968, The Borrego Mountain, California, earthquake of 9 April 1968: A preliminary report: Seismol. Soc. America Bull., v. 58, p. 1183—1186. Anderson, D. L., Ben-Menahem, A., and Archambeau, C. B., 1965, Attenuation of seismic energy in the upper mantle: Jour. Geophys. Research, v. 70, p. 1441—1448. Archambeau, C. B., 1968, General theory of elastodynamic source fields: Rev. Geophyics, v. 6, p. 241—287. Ben-Menahem, A., 1961, Radiation of seismic surface waves from finite moving sources: Seismol. Soc. America Bull., v. 51, p. 401-435. Ben-Menahem, A., Smith, S. W., and Teng, T. L., 1965, A pro- cedure for source studies from spectrums of long-period seismic waves: Seismol. Soc. America Bull., v. 55. p. 203—235. Berckhemer, H., and Jacob, K. H., 1968, Investigation of the dynamical process in earthquake foci by analyzing the pulse shape of body waves: Final Sci. Rept. Contract AF61(052)—801, Institute of Meteorology and Geophysics, University of Frankfurt, Germany, 85 p. Brune, J. N., 1970, Tectonic stress and the spectra of seismic shear waves: J our. Geophys. Research, v. 75, p. 4997—5009. Brune, J. N., and Allen, C. R., 1967, A low stress-drop, low magnitude earthquake with surface faulting—The Im- perial, California, earthquake of March 4, 1966: Seismol. Soc. America Bull., v. 57, p. 501—514. Brune, J. N., Henyey, T. L., and Roy, R. F., 1969, Heat flow, stress, and rate of slip along the San Andreas fault, Cali- fornia: Jour. Geophys. Research, v. 74, p. 3821—3827. Burridge, R., and Knopoff, L., 1964, Body force equivalents for seismic dislocations: Seismol. Soc. America Bull., v. 54, p. 1875—1888. Chinnery, M. A., 1964, The strength of the earth’s crust under horizontal shear stress: Jour. Geophys. Research, v. 69, p. 2085—2089. DeNoyer, J ., 1959, Determination of the energy in body and surface waves (Part II): Seismol. Soc. America Bull., v. 49, p. 1—10. Gutenberg, Beno, and Richter, C. F., 1956, Earthquake magni- tude, intensity, energy, and acceleration: Seismol. Soc. America Bull., v. 46, p. 105—145. Haskell, N. A., 1964, Total energy and energy spectral density of elastic wave radiation from propagating faults: Seismol. Soc. America Bull., v. 54, p. 1811—1841. Julian, B. R., and Anderson, D. L., 1968, Travel times, appar- ent velocities, and amplitudes of body waves: Seismol. Soc. America Bull., v. 58, p. 339—366. Kasahara, K., 1957, The nature of seismic origins as in- ferred from seismological and geodetic observations: Tokyo Univ. Earthquake Research Inst. Bull., v. 35, p. 473—530. Keilis-Borok, V. 1., 1960, Investigation of the mechanism of earthquakes: Akad. Nauk. SSSR Geofiziki Inst. Trudy, no. 4 (166), Soviet Research in Geophysics, v. 4, (trans. by American Geophysical Union, Consultants Bureau En- terprises, New York). . King, C. Y., and Knopoff, L., 1968, Stress drop in earthquakes: Seismol. Soc. America Bull., v. 58, p. 249—257. Knopoff, Leon, 1958, Energy release in earthquakes: Geophys. Jour., v. 1, p. 44—52. Maruyama, T., 1963, On the force equivalent of dynamic elastic dislocations with reference to the earthquake mechanism: Tokyo Univ. Earthquake Research Inst. Bull., v. 41, p. 467—486. Thatcher, W. R., and Hamilton, R. M., 1971, Spatial distribu- tion and source parameters of the Coyote Mountain after- shock sequence, San Jacinto fault zone [abs], in Geol. Soc. America Abs. with Programs, v. 3, no. 2, 229 p. Tsai, Y. B., 1969, Determination of focal depths of earthquakes in the mid-oceanic ridges from amplitude spectra of sur- face waves: Massachusetts Inst. Technology Ph. D. thesis. Wu, F. T., 1966, Lower limit of the total energy of earth- quakes and partitioning of energy among seismic waves: California Inst. Technology, Pasadena, Ph. D. thesis. Wyss, M., 1970, Observation and interpretation of tectonic strain release mechanisms: California Inst. Technology. Pasadena, Ph. D. thesis. Wyss, M., and Brune, J. N., 1968, Seismic moment, stress, and source dimensions for earthquakes in the California-Ne- vada region: Jour. Geophys. Research, v. 73, p. 4681-4694 AFTERSHOCKS OF THE BORREGO MOUNTAIN EARTHQUAKE FROM APRIL 12 TO JUNE 12, 1968 By ROBERT M. HAMILTON U.S. GEOLOGICAL SURVEY ABSTRACT Epicenters determined for 533 of the Borrego Mountain aftershocks form a complex pattern consisting of a main band 60 km long, subparallel and near the Coyote Creek fault, and two clusters centered 16 km northeast and 22 southwest of the main band. Foci range in depth from near surface to 12 km; the zone of shallow foci is partly associated with the zone of surface fractures. All well—located aftershocks with a magni- tude over 3.5 occurred below 6 km depth. The southeast fault strand was the site of much of the aftershock activity; the epicenter pattern along this strand exhibited a southeastward extension of about 3 km after the first week. Focal mechanisms are mostly of the strike-slip type. In the region of faulting, the right-lateral nodal plane is subparallel to the fault trend. At individual recording sites, average magnitude residuals are directly related to P-wave traveltime delays and thus to base- ment depth. INTRODUCTION Aftershocks of the Borrego Mountain, California, earthquake were monitored by 20 portable seismo- graphs of the U.S. Geological Survey (USGS) from April 12, 3 days after the magnitude 6.4 event, to June 12, 1968. Also in operation in the aftershock region during this 2-month period were four seismo- graphs of the California Institute of Technology (Caltech). The main goal of the monitoring was to determine the relationship between the spatial dis- tribution and focal mechanisms of the aftershocks, on the one hand, and the surface breakage and mapped faults, on the other. In conjunction with the aftershock monitoring, a seismic refraction experiment was conducted involv— ing three explosions that were recorded at the port- able seismograph stations and at 10 truck-mounted seismic refraction units. The purpose of this experi- ment was to provide information about P-wave velocities in the upper crust in the Borrego Mountain region in order to locate the aftershock hypocenters more accurately. Details of this refraction study were reported previously (Hamilton, 1970). The combination of a large number of seismo— graphs deployed, their careful distribution with re- spect to the aftershock zone, and the special seismic velocity study made possible unusually precise hypo- center and focal mechanism determinations. The only previous study of similar scope was done after the Parkfield-Cholame earthquake of 1966 (Eaton, O’Neill, and Murdock, 1970). ACKNOWLEDGMENTS The portable seismographs were installed and serviced by J. M. Coakley, E. E. Criley, and J. 0. Ellis. Valuable help in selecting sites and in estab- lishing the first seismograph stations was received from C. R. Allen, California Institute of Technology, who, with J. N. Brune of University of California at San Diego and Max Wyss of Lamont-Doherty Geo- logical Observatory, provided arrival times and mag- nitude estimates for many aftershocks. J. C. Savage and Wayne R. Thatcher, U.S. Geological Survey, made helpful suggestions concerning the interpreta- tion of the aftershock pattern. SEISMOGRAPH STATIONS The Borrego Mountain earthquake occurred at 6:29 pm. local time (P.s.t.). The deployment of the Caltech seismograph stations during the first day of the aftershock sequence has been described by Allen and Nordquist (this volume). The U.S. Geological Survey dispatched 11 portable seismographs on the day after the main shock, and they arrived in the area of the earthquake a day later. At that time, the approximate extent of ground breakage along the Coyote Creek fault was known. It was assumed that the aftershocks would occur in the vicinity of this fracture zone, so some of the seismograph stations were placed along the zone and the others were distributed to surround it (sites 1—11, fig. 12). About 10 days after the main shock, the seismo- graph at site 5 was moved to site 12, and four other seismographs that had become available were in- stalled at sites 13—16 to expand coverage. Near the end of April, 3 weeks after the main shock, the density of the network was increased by station additions at sites 17—19. About 3 weeks later, after two earthquakes were detected southeast of the net- work, the seismograph at site 13 was moved to site 31 32 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL_ 9, 1968 45 30 I5' ll6°OO' r l 33°00' 45 n FIGURE 12. ——-Map of the Borrego Mountain region showing location of seismograph stations (triangles), epicenter of main shock (Allen and Nordquist, this volume), aftershock epicenters (crosses), and active faults. Faults are solid where there is evidence of recent or historic surface breakage, dotted where evidence is less conclusive. 20, and station 21 was established. The final change in the network was the reoccupation of site 5, bring- ing the number of stations to a total of 20. The coordinates of the USGS seismograph stations have been published elsewhere (Hamilton, 1970), except for those of station 13, which are 32050.06’ N., 116° 09.33' W. Table 7 gives the period of operation for each station. The portable seismographs installed in the Borrego Mountain region are the same units that were used by the US. Geological Survey to record the Parkfield— Cholame aftershocks of 1966. The recording system, which uses magnetic tape, the playback system, and the seismograms, have been described in detail by Eaton, O’Neill, and Murdock (1970). Briefly, the portable units all employ a short-period, vertical- component seismometer; six of them are also equipped with two horizontal-component seismom— eters. The system is particularly sensitive to fre- quencies ranging from 1 to 25 Hz (Hertz), with peak response at 17 Hz. Maximum magnification, for a typical attenuation setting, is about 5 million. THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 33 TABLE 7. — Seismograph station data [The station correction is subtracted from the observed arrival time; thus a positive correction means waves are typically late. A positive magnitude residual means that for a particular earthquake the specified station yields a magnitude that is larger than the average magnitude using all stations. P and X magnitudes are based on the maximum signal amplitude in the P wave and the other waves, respectively.] . Recording Period Station P-magnitude X7 Station from to cor(r:ecct;on residual ni'iiiiildglie 11 ............ 4/11 6/12 —0.20 —0.25 —0.18 12 ............ 4/11 6/12 -— .15 — .30 — .27 13 ............ 4/11 6/13 — .36 — .28 — .27 14 ............ 4/11 6/12 .39 .41 .42 4/12 4/18 — .23 —- .01 .06 5/24 6/13 .................. 4/12 6/12 .37 .16 .13 4/12 6/12 .10 .22 .08 4/12 6/12 — .14 — .24 — .16 4/12 6/12 .27 .21 .27 4/12 6/12 — .24 — .13 — .15 4/12 6/12 .09 .05 .09 4/18 6/13 —— .34 .25 33 4/18 5/18 2.00 ............ 4/18 6/12 — .26 .00 .08 4/19 6/12 — .06 — .09 — .19 4/20 6/12 — .42 — .05 .00 4/28 6/12 .52 .07 — .14 4/29 6/12 — .07 — .13 — .21 4/30 6/12 .20 .21 .13 5/19 6/12 .71 .45 .38 5/19 6/12 .89 .42 .18 4/09 6/11 — .27 ............ 13-component station. 2Station correction derived from average of P-wave arrival time residuals. Magnitudes not determined because seismograph was not satisfactorily cali- brated. aCaltech station. DATA PROCESSING In the 2 months that the seismograph network was in operation, well over 10,000 earthquakes were re- corded. Locating all of these events was not feasible, so it was decided to locate 500, a manageable number to process with current techniques. The principal goal in the analysis was to define the nature and extent of the aftershock activity. The more formid- able task of producing a uniform data set including, for example, all earthquakes above a certain magni- tude, was not undertaken. ‘ Attention in locating the aftershocks was focused on the larger events. Besides involving a dispropor- tionately large part of the energy released by the sequence, the larger aftershocks were recorded more clearly at more stations and therefore could be located more accurately than the smaller ones. Also, the direction of first motion on each record was generally clearer. The earthquakes were selected from slow-speed (37.5 mm/ min) playbacks of the tapes from station 1. Station 1 was chosen as the primary source be- cause it was located to the side of the main after- shock band; distance effects were thereby minimized, and it was easier to distinguish the larger from the smaller earthquakes. Earthquakes were selected for analysis if their maximum amplitude exceeded a certain value; this value was varied with the S-P interval or with the shape of the signal envelope to allow for amplitude attenuation with epicentral dis- tance. The amplitude threshold was set to limit the number of events selected to about 500. Slow-speed (60 m/min) playbacks from stations 6, 9, 12, and 15 were used to fill periods when no record was obtained at station 1. Although most of the larger aftershocks were selected for analysis by this procedure, not all of those selected were successfully located. A poor or nonexistent radio-time-code signal from National Bureau of Standards station WWVB and interfer- ence from an earlier earthquake were two reasons for not determining a hypocenter. The times of the events selected were noted to the nearest minute. High-speed playbacks (75 cm/min) were then made covering a span of several minutes including each time. From these playbacks, readings were made of the P-wave arrival time, direction of first motion, maximum amplitude in the P-wave group, and maximum amplitude in the signal after the P-wave group. Sharp onsets of P could, in gen- eral, be read With a precision of 0.02 and 0.03 sec. The readings were made for all events that appeared on the playbacks and seemed large enough to be locatable, even for smaller events that had not been previously selected but that occurred within several minutes of a selected event. Extensive use was made 0f readings from the Caltech station at site 22 (fig. 12). Coordinates for this station have been given previously (Hamilton, 1970), and they are also listed in Allen and Nordquist (this volume). The aftershock parameters (origin time, focal coordinates, and magnitude) were c0mputed using a program written by Eaton (1969). Features of this program are also discussed by Eaton, O’Neill, and Murdock (1970). Locations, derived from P—wave arrival times, were determined only for those events read at six or more stations. Aftershock parameters and data indicating the reliability of the parameters are given in table 8. . The precision of location of each, earthquake de— pends mainly upon where the event occurred with respect to the seismograph stations. Epicenters for aftershocks within the net are thought to be accurate within 1 km; this opinion is based on comparison of the computed and actual epicenters of the explosions (Hamilton, 1970) and on success in defining narrow lineations in epicenter patterns and relating them to recognized faults in a variety of similar studies elsewhere. As the margin of the net is approached, the epicenter precision declines rapidly. Experience indicates that just outside a net like the one in the Borrego Mountain area, epicenter errors of several kilometers are not unusual. Within the net, uncer- tainty in focal depth, in general, is probably about flmmuw UWUUN NNNNO— ~00on 00000 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 TABLE 8.—Bor’rego Mountain aftershocks [For explanation of column heads see notes at end of table] SEC LAT N LONG H DEPTH MAG N0 GAP DMIN ERT ERH ERZ MD 1.0 33-10.1 116- 4.2 5.0* .3.4 6 177 15.3 0.03 0.2 0.02 33.3 33- 6.3 116- 2.8 2.1 1.8 6 147 8.3 0.01 0.0 0.1 0.00 16.8 33-14.9 116- 6.2 5.0* 0.8 7 199 15.0 0.06 0.4 0.03 37.2 33- 4.2 116-12.7? 5.0‘ 1.8 6 119 6.2 0.48 4.2 0.51 52.6 33- 1.5 116-15.7? 5.0* 1.1 6 224 4.7 1.08 7.8 0.50 24.7 33- 9.2 116- 6.4 2.0* 2.1 7 130 12.5 0.10 0.9 0.14 46.3 33- 9.0 116- 6.2 1.0 .2.8 8 123 12.3 0.03 0.1 0.4 0.02 19.8 33-16.1 116-12.3 7.5 1.3 8 186 2.5 0.38 1.8 3.8 0.14 59.7 33-14.9 116-13.8 4.6 0.9 7 116 2.0 0.08 0.5 1.2 0.04 30.0 33- 3.4 116- 1.1 1.5 1.9 7 159 4.6 0.02 0.2 0.4 0.02 55.2 33- 0.9 115-58.6 7.7 1.2 6 269 7.8 0.16 0.7 0.8 0.01 39 7 33- 3.6 116- 1.3 3.7 1.1 7 151 4.8 0.01 0.1 0.4 0.01 9.9 33- 2.6 116- 2.5 6.5 1.3 7 161 2.1 0.01 0.1 0.1 0.00 47.3 33- 8.5 116- 4.9? 0.6 1.6 7 122 10.6 0.04 0.2 0.5 0.02 31 1 33- 8.9 116- 6.2 2.0* 1.6 6 126 12.3 0.09 0.8 0.10 39.7 33- 7.7 116- 4.2? '1.0 0.8 6 112 10.0 0.08 0.5 1.0 0.05 23.2 33- 5.0 116- 3.4 4.4 1.9 6 91 5.9 0.04 0.2 1.0 0.01 59.1 33- 3.7 116- 1.3 2.9 0.8 6 147 4.9 0.07 0.6 0.9 0.03 26.7 33- 7.5 116- 5.7 1.3 1.6 6 111 11.1 0.00 0.0 0.1 0.00 58.7 33- 6.0 116- 2.0 7.6 1.9 6 98 8.1 0.05 0.2 0.7 0.01 15.9 33- 1.8 116-13.4 5.0’ 1.2 6 170 1.9 0.21 1.7 0.17 30.1 33- 4.0 116- 1.1 2.6 1.3 6 144 5.4 0.02 0.2 0.3 0.01 54.3 33- 6.1 116- 1.7 8.0 1.4 6 100 8.3 0.05 0.2 0.6 0.01 4.3 33- 2.8 115-58.8 7.1 1.3 6 200 7.6 0.20 0.8 1.8 0.03 1.7 33-12.2 116-10.2 5.0# 1.2 6 149 15.9 0.05 0.5 0.06 53.1 33-11.3 116- 8.1 5.0* 0.8 6 149 15.2 0.06 0.5 0.07 53.3 33- 3.3 [IS-58.6 3.1* 2.3.2.9 7 189 7.0 0.07 0.5 0.05 49.5 33-16.9 116-15.8 2.6 2.6.3.3 9 159 3.4 0.04 '0.4 0.3 0.04 38.2 33- 5.3 [IS-55.9 5.0* 1.7 6 149 2.0 0.17 1.4 0.11 45.5 33- 1.0 115-58.6 7.9 2.5.3.2 7 229 7.8 0.11 0.5; 0.8 0.02 59.1 33- 4.4 116- 2.6 6.8 .3.1 9 111 5.0 0.02 0.1 0.2 0.01 0.9 33- 2.4 116- 0.4 2.0 2.3.2.5 7 200 10.1 0.04 0.4 0.7 0.01 49.3 33-16.6 116-14.2 1.3 . .3.3 13 148 1.2 0.06 0.4 0.5 0.09 10.0 33- 7.7 116— 5.4 3.0 2.0 10 73 1.1 0.03 0.3 0.6 0.05 15.5 33- 7.6 116- 5.4 4.4 1.9 10 72 1.0 0.03 0.2 0.6 0.05 13.8 33-14.2 116-10.8 9.2 1.9 10 103 5.8 0.07 0.3 0.7 0.04 2.5 33-13.7 116- 1.9 8.1 1.7 10 186 8.6 0.16 0.6 1.7 0.06 55.0 33-14.4 116- 0.9 5.5 1.4 10 206 8.8 0.14 0.6 2.2 0.06 5.9 33- 6.0 116— 2.9 1.5 .3.2 13 89 4.9 0.06 0.4 0.7 0.09 37.6 33- 2.0 115-59.2 1.2 2.2.2.7 10 212 6.7 0.07 0.4 0.5 0.05 20.6 32-57.6 116-18.9? 12.7 1.8 10 272 11.3 0.18 1.0 0.3 0.05 57.2 33- 6.1 116- 2.9 1.7 2.7.2.9 10 88 4.8 0.03 0.3 0.5 0.05 3.2 33- 3.2 116- 1.3 1.8 1.4 9 161 4.3 0.07 0.5 0.7 0.07 2.4 33- 4.9 116- 2.7 8.1 1.5 10 104 5.8 0.07 0.3 0.7 0.03 18.2 33- 3.3 116- 1.0 2.4 .3.2 11 163 4.7 0.09 0.7 1.2 0.06 53.1 33- 3.4 116- 1.1 4.3 0.3 6 159 4.7 0.30 1.5 6.1 0.09 38.3 33-11.9 116- 9.2 5.2 1.7 10 80 3.1 0.07 0.4 1.2 0.07 56.6 33- 6.1 116- 2.1 1.7 1.2 10 96 6.0 0.09 0.7- 1.2 0.13 57.7 33- 8.9 116- 6.2 1.2 2.0 10 80 3.2 0.04 0.3 0.5 0.05 37.9 33-15.3 116- 5.7 8.2 1.0 8 144 4.2 0.10 0.4 0.9 0.03 11.5 33-13.9 116-13.5 7.5 1.6 10 88 3.9 0.07 0.3 1.0 0.04 50.2 33- 8.4 116- 5.2 5.6 0.9 10 80 2.5 0.05 0.3 0.8 0.05 18.3 33-11.6 116- 8.3 9.5 2.0 10 85 3.0 0.07 0.3 0.8 0.04 24.5 33- 6.4 116- 3.6 8.7 .3.3 13 78 3.5 0.09 0.4 1.0 0.07 53.6 33-10.0 116- 6.3 2.7 1.8 11 90 4.6 0.05 0.3 0.5 0.07 32.6 33- 8.3 116- 6.0 5.1 1.9 11 76 2.1 0.04 0.3 0.7 0.05 54.1 33-20.7 116-17.7 13.9 1.1 7 252 5.0 0.50 2.7 2.9 0.06 2.7 33- 7.1 116- 6.1 8.9 1.8 10 120 0.6 0.21 1.0 2.1 0.12 2.3 33- 5.9 116- 2.9 11.1 1.7 10 90 4.9 0.08 0.3 0.8 0.04 4.2 32-57.8 116-15.0 11.2 v3.2 11 266 6.8 0.19 0.9 1.0 0.04 m>m>n n>aa>n manna) DDWWD nnmnn OWDUO trauma nnmmn 31¢an 000070 0 PUDDDD ODCIOD THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 35 TABLE 8.—Bowego Mountain aftershocks—Continued D A >mw>> 1968 HR MN SEC LAY N LONG H DEPTH MAG N0 GAP DMIN ERT ERH ERZ MD 1 24 36.4 33-15.2 116- 5.7 10.2 0.5 8 143 4.4 0.18 0.7 1.7 0.07 B 1 25 53.1 33- 9.2 116- 6.9 3.4 '3.0 7 79 4.1 0.03 0.2 0.8 0.03 8 1 26 4.8 33-17.2 116-15.3 2.1 0.4 8 156 3.1 0.10 0.8 0.9 0.10 B l 26 18.3 33- 8.5 116- 5.9 4.9 g3.0 11 77 2.5 0.03 0.2 0.5 0.03 A 1 27 34.6 33- 8.4 116- 5.9 2.9 0.6 9 77 6.2 0.06 0.5 5.1 0.07 C 1 29 22.7 33- 7.8 116- 6.7 2.6 1.0 10 69 1.9 0.05 0.4 0.4 0.06 3 30 52.0 33- 8.4 116- 5.8 4.6 1.6 11 88 2.2 0.04 0.3 0.7 0.06 3 51 4.6 33- 5.4 116- 2.8 1.4 1.6 10 89 5.6 0.02 0.2 0.3 0.04 7 14 45.3 33-15.4 116—13.9 3.5 0.3 6 71 1.0 0.04 0.3 0.6 0.02 7 15 34.8 33-11.6 116- 8.6 6.6 1.8 11 82 2.7 0.05 0.3 0.8 0.05 7 45 21.1 33- 7.7 116- 6.6 0.5 1.2 10 112 1.6 0.07 0.4 0.9 0.08 A 12 55 58.7 33-14.2 116-11.4 10.8 .4.3 12 108 5.1 0.09 0.6 0.8 0.08 8 13 13 7.4 33- 3.4 115-59.1? 2.9 2.2.2.6 10 102 6.7 0.04 0.6 1.1 0.05 B 13 14 47.6 32-53.7 116-18.7 5.0. 1.5 7 301 28.0 1.53 7.1 0.10 D 16 46 30.4 33- 9.4 116- 6.0 4.9 p3.0 8 111 4.2 0.07 0.5 1.3 0.06 A 21 6 22.8 33- 7.2 116- 5.6 2.9 0.5 6 80 0.2 0.15 1.5 1.2 0.09 8 21 6 32.6 33- 7.1 116- 4.8 2.0 1.6 7 107 1.4 0.10 0.9 1.1 0.12 8 9 50 11.9 33- 4.6 116- 2.2 1.3 1.7 9 115 5.5 0.04 0.3 0.6 0.05 8 10 7 11.6 33-18.9 116-13.3 3.04 .3.3 12 207 5.4 0.08 0.5 0.07 C 12 28 7.5 33-17.4 116-15.6 3.0 p3.5 11 210 3.7 0.12 0.7 2.4 0.08 C 15 50 55.5 33- 9.1 116- 5.9 3.9 1.9 9 110 3.5 0.04 0.3 0.9 0.05 A 21 56 50.3 33- 2.7 115-59.3 7.3 .3.4 10 219 8.4 0.09 0.5 1.1 0.04 8 21 58 0.2 33- 8.6 116- 5.7 1.2 8 84 2.7 0.14 1.1 2.4 0.12 B 22 7 6.1 33- 2.9 115-59.0 8.2 13.3 11 186 7.3 0.18 0.9 1.7 0.08 C 3 30 29.9 33- 2.9 115-59.2 8.3 14.8 13 186 7.0 0.14 0.6 1.4 0.08 C 4 56 45.5 33— 2.4 116- 0.3 2.9 1.8 11 197 5.2 0.04 0.3 0.3 0.03 8 6 35 31.3 33-11.7 116-10.1 4.5 1.1 9 71 3.1 0.05 0.3 1.1 0.06 A 6 37 33.4 33- 8.9 116- 3.5 4.7 2.0 10 106 4.7 0.07 0.4 1.4 0.06 A 9 49 37.6 33-12.1 116- 9.3 8.3 2.2 10 77 3.6 0.04 0.2 0.5 0.03 A 10 51 34.0 33- 6.0 116- 2.1 6.7 1.8 9 142 6.0 0.06 0.3 0.8 0.03 B 2 43 47.5 33- 3.2 115-59.6 7.9 o3-7 12 178 6.6 0.07 0.3 0.7 0.04 B 3 14 25.8 33-14.0 116-11.3 11.4 g3.4 12 80 5.4 0.04 0.3 0.4 0.04 A 22 1 17.2 33- 8.3 116- 4.9 7.0 2.2 9 96 7.5 0.04 0.2 0.6 0.03 8 3 21 9.1 33- 2.7 115-59.4 9.2 2.1 12 102 5.5 0.13 0.4 1.1 0.05 A 6 8 25.6 33- 6.8 116- 5.0 3.4 0.2 8 72 1.3 0.02 0.2 0.4 0.02 A 6 8 43.6 33- 5.4 116- 4.0 9.6 1.5 12 65 4.3 0.06 0.2 0.6 0.04 A 6 17 25.9 33- 2.1 115-58.3 8.0 1.3 12 133 4.1 0.53 1.9 4.4 0.10 8 6 18 10.5 33- 7.6 116- 5.3 10.0 1.6 12 71 1.0 0.12 0.5 1.3 0.08 8 8 2 21.7 33- 6.8 116- 3.8 10.8 1.7 12 65 3.0 0.04 0.1 0.4 0.02 A 9 9 13.8 33- 2.4 116- 1.3 8.0 1.5 12 116 3.6 0.06 0.3 0.6 0.03 A 9 28 26.2 33- 2.6 115-59.6 7.9 1.8 12 98 5.5 0.10 0.4 1.0 0.04 A 9 31 47.9 33-15.6 116-17.0 0.1* 0.6 6 155 4.4 0.14 1.0 0.12 C 17 46 46.4 33-11.8 116-10.0 5.6 1.5 10 84 3.3 0.04 0.3 0.7 0.04 A 19 28 8.8 33- 9.4 116- 3.6 5.6 1.6 10 96 5.2 0.03 0.1 0.6 0.03 A 21 23 35.7 33- 8.1 116- 1.9 1.2 1.6 13 78 6.2 0.04 0.3 0.5 0.07 B 23 24 15.9 32-56.2 116-16.3 10.4 1.1 13 201 10.2 0.10 0.5 1.2 0.06 B l 46 15.5 33- 9.1 116- 4.9 7.8 1.8 14 87 3.8 0.04 0.2 0.6 0.05 A 6'11 51.1 33- 1.5 116-13.6 0.2* 1.4 14 159 1.7 0.07 0.5 0.10 C 6 12 50.6 33- 0.5 115-57.5 4.3 1.2 14 84 1.7 0.03 0.? 0.5 0.04 A 6 54 7.0 33- 2.8 115-59.2 4.8 0.6 13 62 5.6 0.03 0.2 0.8 0.04 8 6 55 20.3 32-58.7 116-20.1 11.0 1.2 15 212 12.2 0.10 0.5 1.4 0.0 8 7 7 27.0 33- 0.9 115-56.6 7.5 1.3 15 96 3.2 0.06 0.3 0.7.3.06 A 8 7 13.5 33- 7.8 116- 2.1 1.7 1.3 15 73 5.6 0.03 0.2 0.4 0.06 B 11 34 4.1 32-56.8 116-17.1 5.3 0.8 8 204 10.1 0.10 0.6 2.5 0.05 C 11 35 57.7 33- 7.7 116- 3.8 0.1* 1.2 15 73 3.1 0.04 0.3 0.09 B 12 14 48.2 33- 1.2 116-11.2 3.4 0.9 15 87 2.5 0.04 0.3 1.1 0.07 A 12 32 28.0 33- 2.2 115-59.2 8.4 1.9 15 65 4.6 0.04 0.2 0.4 0.03 A 12 34 32.0 33- 8.2 115-58.8 4.5 1.1 15 121 3.7 0.07 0.4 1.3 0.06 8 14 33 19.9 33-15.5 116- 0.8 0.3* 1.7 15 218 9.5 0.09 0.6 0.09 C 14 34 15.4 33- 4.6 116- 6.9 10.1 0.6 15 52 5.2 0.05 0.2 0.6 0.05 A moooawm r4:- "—0.000: THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 TABLE 8.——Borrego Mountain aftershock—Continued SEC LAT N LONG H DEPTH MAG N0 GAP DMIN ERT ERH ERZ MD 43.3. 33-17.1 116-13.8 7.4 2.0 13 202 2.2 0.09 0.4 0.7 0.05 36.8 33- 2.2 116- 0.9 2.7 1.2 11 146 4.0 0.07 0.5 0.7 0.07 3.6 33-14.4 116-11.4 8.1 1.4 13 94 4.9 0.06 0.3 0.8 0.06 5.4 33- 7.4 116- 1.7 6.2 2.1 13 66 7.3 0.04 0.2 0.9 0.04 21.6 33- 6.6 116- 4.0 1.8 1.4 14 64 2.9 0.03 0.2 0.4 0.06 2.1 33- 6.8 116- 2.7 2.0 1.2 14 64 4.7 0.03 0.2 0.3 0.04 44.3 33- 4.9 116- 2.8 7.2 1.5 14 54 5.8 0.08 0.5 1.3 0.08 49.1 33- 1.8 116-11.0 2.5 1.2 14 77 3.1 0.05 0.3 0.6 0.06 12.6 33- 1.6 115-57.6 9.2 2.2 15 83 3.4 0.03 0.2 0.3 0.04 30.9 33- 1.3 116-11.1 3.7 0.5 8 113 2.8 0.11 0.6 1.7 0.04 53.6 33- 5.1 116- 2.7 8.0 0.9 15 55 6.1 0.03 0.2 0.5 0.04 55.7 33- 7.1 116- 5.5 2.9 1.7 13 67 0.4 0.03 0.2 0.3 0.04 37.6 33-22.0 116-14.6 10.1 2.6.3.2 15 114 7.2 0.05 0.4 0.4 0.07 29.4 33- 7.2 116- 1.9 2.7 1.5 15 65 6.0 0.04 0.2 0.4 0.06 21.8 33- 9.3 116- 5.7 2.3 1.4 15 86 3.9 0.04 0.2 0.4 0.05 53.7 33- 1.6 115-56.3 4.5 0.6 11 101 4.4 0.06 0.4 1.2 0.06 15.9 33- 8.4 116- 3.6 5.2 1.7 15 83 4.0 0.03 0.2 0.6 0.05 47.9 33- 5.4 116- 1.9 7.5 1.3 15 62 6.7 0.04 0.2 0.6 0.05 35.2 33-16.7 [lb-15.4 1.7 1.3 15 85 2.7 0.04 0.3 0.5 0.06 17.6 33-16.6 116-15.6 1.1 1.7 15 87 2.8 0.03 0.2 0.3 0.05 52.9 33-16.7 116-15.6 0.9 1.3 14 87 2.9 0.05 0.3 0.6 0.07 3.6 33- 5.5 116- 0.9 6.6 1.8 15 71 7.8 0.04 0.2 0.6 0.05 47.4 33-12.3 ll6-10.5 3.7 1.5 13 85 4.3 0.03 0.3 1.5 0.07 34.1 33- 9.9 116- 1.6 3.4* 2.0 13 110 5.0 0.04 0.3 0.08 49.6 33- 1.3 116-14.2 7.2 1.3 15 164 2.3 0.04, 0.3 0.6 0.04 21.2 33- 4.1 116- 2.4 8.5 2.3 15 53 4.6 0.03 0.2 0.4 0.04 7.1 33-11.6 116- 9.3 7.8 1.6 15 81 2.7 0.04 0.2 0.5 0.05 38.5 32-59.6 116-14.1 6.1 0.7 11 171 3.2 0.07 0.4 0.9 0.05 10.1 33-14.0 116-11.4 11.0 1.9 15 80 5.3 0.03 0.2 0.4 0.05 0.1 33- 4.1 116- 2.0 1.1 0.3 6 118 4.8 0.06 0.5 0.9 0.04 13.4 33- 7.1 116- 5.1 2.7 1.4 15 67 0.9 0.03 0.2 0.3 0.05 1.8 33- 7.1 116- 2.5 1.4 0.9 14 66 5.0 0.04 0.3 0.5 0.06 35.5 33- 8.4 116- 4.6 2.6 1.6 14 80 2.8 0.05 0.3 0.5 0.06 48.9 33-16.1 116-16.1 2.1 1.8 13 91 3.4 0.24 1.1 2.0 0.17 28.0 33- 6.8 116- 5.2 4.9 1.4 14 64 8.9 0.08 0.5 2.9 0.09 33.4 33- 7.6 116- 2.5 1.2 0.4 8 102 7.9 0.07 0.5 1.1 0.06 33.9 33- 1.7 115-59.3 8.0 0.2 7 73 3.8 0.09 0.5 1.2 0.05 30.8 33- 2.6 [IS-59.0 9.2 1.7 13 65 5.2 0.03 0.2 0.4 0.03 11.1 33- 7.7 115-58.5 5.9 0.6 12 127 4.6 0.07 0.4 1.1 0.05 32.6 33- 2.7 116- 1.3 2.2 1.4 14 49 3.8 0.03 0.2 0.3 0.05 55.1 33- 1.8 115-59.2 6.0 1.3 15 66 3.9 0.03 0.2 0.5 0.04 12.8 33- 8.5 116- 6.1 6.0 1.4 14 78 5.8 0.03 0.2 0.7 0.04 28.5 33- 2.9 115-59.3 6.0 1.1 11 60 5.8 0.03 0.2 0.5 0.03 45.0 33- 1.5 115-49.4 1.1 0.5 10 208 7.2 0.09 0.5 0.6 0.06 47.2 33- 1.3 115-49.6 1.4 2.5.2.7 12 205 7.0 0.08 0.5 0.5 0.07 6.0 33- 2.5 116- 0.1 5.8 2.2.2.7 15 56 5.3 0.03 0.2 0.6 0.05 0.1 33- 7.8 116- 6.2 3.1 1.2 15 72 1.5 0.02 0.2 0.5 0.04 14.3 33- 3.7 116- 1.3 3.9 1.6 14 56 4.8 0.02 0.1 0.8 0.04 35.7 33- 3.5 116- 1.8 7.2 1.0 13 96 4.1 0.04 0.? 0.5 0.05 55.5 33- 7.5 116- 5.4 3.8 1.1 14 71 0.8 0.02 0.2 0.5 0.04 35.4 33- 8.6 116- 4.5 4.6 1.4 15 83 3.2 0.02 0.2 0.6 0.04 26.9 53- 1.8 116-11.0 2.5 0.9 14 70 3.2 0.05 0.3 0.6 0.07 30.3 33-11.6 116- 9.3 7.9 1.2 16 77 2.5 0.04 0.2 0.6 0.06 41.1 33- 1.7 115-56.1 4.3 0.4 16 105 4.8 0.05 0.3 1.2 0.07 4.4 33- 9.0 116- 7.4 2.7 0.7 16 75 3.6 0.04 0.2 0.4 0.07 2.3 0 3 1 7 16 178 8.1 0.05 0.4 0.07 6.8 1 4 1 1 16 166 2.6 0.05 0.3 0.4 0.05 21.7 33— 1.6 116-14.2 0.2* 0.9 15 164 2.5 0.08 0.6 0.11 7.0 2 l 1 3 16 166 2.5 0.06 0.3 0.5 0.07 5.9 5 1 .2.9 16 56 4.7 0.02 0.1 0.5 0.04 D >>>>> >>>>> Nab») DWDCDCC mn>>> QPWDD (papa-Va: >>>>> >m>>> >>>>> >m>mm >WOCDO THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 ~l~l~l~lb ##J‘é“ OOQCDN 19 20 22 4 32 9 41 50 11 SEC 42.6 54.4 23.1 48.2 22.8 41.6 48.7 46.1 16.3 30.7 36.6 38.4 24.4 26.3 45.1 12.3 27.8 30.0 10.4 24.2 32.7 13.3 476-246 0 - 72 - 4 LAT N 33- 2.6 33-16.2 33- 2.9 33- 0.2 33- 8.1 33- 9.6 33-16.5 33- 1.8 32-56.5 33- 2.4 33-10.3 33- 7.6 33- 7.6 33- 9.1 33- 7.6 33-18.2 33- 0.6 33- 6.0 33- 0.4 33- 0.3 33- 0.1 33- 8.1 33- 6.9 33-16.6 33-11.9 33- 7.6 33-16.9 33-17.1 33- 6.4 33- 8.0 33- 8.0 33- 8.1 33- 8.1 33- 1.5 33-17.5 33-17.6 33- 3.4 33- 3.4 33- 3.1 33- 8.6 33- 0.8 33- 8.3 33-19.2 33-16.8 33- 5.2 33- 9.9 33-10.8 33- 7.8 33- 7.7 33- 1.9 33- 8.1 33- 1.7 33- 8.4 33- 5.3 33- 5.5 33-11.9 33-15.1 32-56.1 33- 2.5 33- 6.1 LONG w DEPTH MAG 115-59.7‘ 5.6 0.7 116- 0.0 0.4 0.9 115-55.4 5.0 1.7 116-14.9 0.0: 0.8 116— 2.0 2.0 0.3 116- 6.7 6.0 1.4 116-16.2 3.6 0.8 115-58.o 8.7 0.4 116-15.7 10.5 1.5 115-59.7 5.3 1.7 116- 8.3 5.1 0.9 116- 2.2 6.7 1.5 116- 2.2 1.8 1.5 116- 4.2 8.1 1.9 116- 2.6 1.2 0.5 116-15.8 10.0 1.8 115-55.7 7.4 0.7 116- 3.4 2.8 1.0 116—15.0 6.1 0.7 116—14.8 2.4 0.2 116—14.9 2.5 1.2 116- 4.9 5.2 1.5 116- 5.1 2.6 0.9 116-15.2 2.8 0.5 116- 9.6 5.4 1.6 116— 2.2 7.1 0.9 116-13.5 1.2 1.9 116—13.3 0.2: 0.7 116- 3.2 3.7 0.3 116- 5.3 5.7 1.2 116- 5.3 5.7 .3.4 116- 5.5 5.3 2.0 116— 5.1 5.5 1.3 116- 1.6 7.8 1.6 115-57.9 2.0: 1.6 115-58.0 2.0: 0.5 116- 1.1 5.3 1.7 116- 1.7 1.9 .3.0 116— 1.9 1.4 1.9 116— 3.9 1.9 0.1 116-13.8 2.8 1.0 116— 4.7 3.2 0.0 116- 9.8 10.0 1.6 116—14.5 3.2 2.2 115—49.9? 2.0: 0.9 116— 6.4 2.7 1.4 116- 2.2? 1.8 1.6 116- 2.8 8.6 0.3 116— 6.1 2.1 1.7 116- 1.1? 7.8 1.6 116- 4.7 5.9 0.5 115-57.0 6.4 1.5 116— 4.3 2.9 -0.0 116- 3.1 2.6 1.5 116- 3.0 1.6 .2.9 116- 9.4 5.1 2.3.3.0 115-58.3 0.3* .3.1 116-15.6 2.8* 1.0 116- 1.1 2.1 1.9.2.9 116- 1.9 7.6 1.8 TABLE 8.—Borrego Mountain aftershocks—Continued- N0 GAP DMIN 16 16 14 16 16 15 16 16 10 11 16 16 15 16 15 15 14 59 189 119 248 116 83 92 81 197 59 201 WJ‘O‘NF‘ J‘“UIU‘~ I...- ‘ONO‘UIU'I WW§§UI 00.00 NwONO‘ J‘O‘r-‘UJO OD-‘WCDU'I ,_. wt-#>> m>>m> CDWUW >m>x>n n>b>> bbfibb Dbbbm mm>>> >>w>> DWW>> moan» )bfifip 38 MAY J‘J‘J‘WW WWWWU‘ WNNNN NF‘F‘F‘F‘ VWF‘F‘“ #«£‘#‘-bb 25 26 27 49 37 58 33 36 42.2 16.5 24.7 37.1 12.7 0.2 19.1 39.7 56.7 24.1 6.1 37.2 20.9 6.4 18.0 30.0 17.7 44.6 45.0 38.0 48.1 THE LAT N 33-1 33- 33- 33- 4 33- 1 33-15.1 33-15.2 33-17.3 33-15.4 33-17.4 33-15.5 33- 9.0 33- 7.7 33- 3. 33- 2. 33- 1. 33- 1. 33- 6. BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, TABLE 8.-——Borrego Mountain aftershocks—Continued LONG H 115—59.8 115-56.0 116— 5.7 116- 5.7 116- 5.5 116- 1.1 116- 4.7 116- 3.4 116- 5.4 116- 3.2 115-57.0 116- 3.2 116- 5.3? 116‘ 1.7 116-14.1 115-56.4 116- 4.6 116- 2.8 116- 6.3 116- 6.5 115‘58.3 115-57.9 116- 3.2 116-16.4 [IS-55.6 115-55.7 116- 3.0 116- 9.9 116— 4.2 116' 4.2 116-10.8 115-55.9 116- 4.7 115-56.8 116- 1.4 116- 2.1 116- 1.5 115-58.0 115-58.0 116-15.5 115-58.2 116-15.6 115-58.3 116- 4.8 116- 7.0 116- 0.9 116- 0.7 115-59.? 115-59.8 116- 1.9 115-56.2 116-18.5 115-59.6 116- 2.3 116-15.7 116- 9.7 116-13.3 116— 2.3 116-15.1 116- 0.1 DEPTH WNNWW D I o a o WJ‘flk'flO CO‘ONUI no... “0000 p. ‘l’ QQOO‘O o o a o o J‘U‘U’OU 'I‘ NW§U§ no... GIO\H‘JO a~m~fl-u NUIUOQ 0.... MAG bur-00rd:— u o o o a UNOWN WFflJ‘fl l v—OONO 0000. ,3.0 ,2.7 NO 15 14 9 15 7 14 16 15 16 GAP 188 104 98 87 150 56 66 113 85 124 88 100 110 65 80 100 97 59 84 71 193 195 51 131 114 DMIN I-I-I‘J‘UIJ‘ OI... \DrdI-INO ##Nflm on... too. WNV‘VDO obomw NI—UID-‘O wmmuo OU'ID-‘J-‘D wUJNN“ on... U'INNWO ERT 0.09 0.06 0.04 0.03 0.05 0.02 0.02 0.03 0.03 0.05 0.05 0.04 0.16 0.03 0.04 0.03 0.04 0.02 0.03 0.03 0.08 0.12 0.03 0.03 0.06 0.04 0.06 0.04 0.03 0.03 0.10 0.05 0.05 0.10 0.02 0.04 0.04 0.07 0.08 0.03 0.08 0.03 0.07 0.03 0.03 0.05 0.02 0.04 0.03 0.05 0.08 0.07 0.03 0.03 0.05 0.03 0.08 0.04 0.11 0.04 00000 WNNNN 00000 I...- uaN-buéw o g o o o F'O‘WW“ NNdfi’N WNNOUI NNNNN 00000 00000 00000 NUIJ‘N“ 00000 no... NNU‘NUI NNNNU .000. 00000 00000 wnz~:~# 1968 ERZ 00000 00°00 00°00 00°00 MD 0.07 0.05 0.04 0.05 0.04 0.04 0.04 0.04 0.05 0.04 0.05 0.03 0.10 0.05 0.05 0.04 0.05 0.05 0.06 0.05 0.08 0.09 0.04 0.04 0.07 0.04 0.08 0.07 0.06 0.07 0.06 0.05 0.11 0.13 0.04 0.09 0.05 0.06 0.08 0.05 0.08 0.04 0.07 0.05 0.04 0.06 0.05 0.04 0.05 0.06 0.06 0.07 0.05 0.05 0.06 0.04 0.10 0.05 0.04 0.05 >>1>cnw >>>>> D>O>fi DOODW DUUDW >>>a7> )tflbnfi >>>>> >>mu=> m>>>> EDDDG O bmbcflb THE BORREGO MOUNTAIN EARTHQUAKE OF ‘APRIL 9, 1968 39 TABLE 8.—Borrego Mountain aftershock—Continued 1968 HR MN SEC LAT N LONG H DEPTH MAG N0 GAP DMIN ERT ERH E82 MD 0 MAY 4 23 50 45.7 33- 2.2 116- 2.5 4.5 1.8.2.6 18 56 1.8 0.02 0.2 0.4 0.04 A 5 5 8 50.8 33-16.5 [lb-15.2 3.1 1.8 17 83 2.2 0.03 0.2 0.7 0.05 A 5 5 43 50.9 33-17.7 116-13.7 5.0 2.2.2.9 18 89 3.1 0.04 0.3 0.8 0.05 A 5 10 14 56.0 33- 3.1 116- 1.4 1.1 1.5 17 69 2.2 0.04 0.3 0.5 0.09 A 5 13 34 10.8 33- 8.8 116- 6.7 4.0 1.2 17 49 4.7 0.02 0.2 1.2 0.05 A 5 16 47 37.3 33- 7.3 116- 3.0 1.7 1.8.2.5 18 54 4.2 0.02 0.1 0.2 0.04 A 6 2 0 5.0 33- 2.5 115-56.8 6.4 0.9 18 152 5.4 0.08 0.4 0.9 0.07 B 6 2 3 42.4 33- 2.1 115-57.0 7.1 1.8.2.2 18 153 4.6 0.11 0.6 1.1 0.08 C 6 2 4 48.5 33-16.4 116- 0.2 0.3* 0.8 15 194 4.5 0.10 0.6 0.09 C 6 2 42 31.2 32-59.9 115-58.5 8.7 1.8 17 86 0.4 0.04 0.2 0.3 0.04 A 6 6 24 9.7 33- 2.4 115-56.9 6.3 1.7 17 152 5.1 0.06 0.3 0.7 0.05 8 6 7 34 26.0 33- 2.4 115-56.8 6.4 1.8 17 153 5.2 0.08 0.4 0.8 0.06 B 6 8 32 28.8 33- 2.2 116- 2.3 5.3 1.4 16 58 2.0 0.03 0.2 0.4 0.04 A 6 10 53 36.9 33- 2.0 116- 0.1 9.2 .3.5 18 87 4.7 0.05 0.3 0.5 0.05 A 6 12 42 21.0 33- 2.4 115-57.1 6.3 1.6 17 149 5.1 0.05 0.3 0.6 0.04 8 6 17 31 47.6 33- 2.4 115-56.9 6.7 .4.0 18 152 5.0 0.07 0.3 0.7 0.05 8 7 1 7 16.9 33- 9.9 116- 5.0 10.3 1.5 19 48 5.1 0.05 0.2 0.6 0.07 A 7 1 26 38.6 33- 2.8 116- 1.1 7.3 1.9.2.6 19 51 2.7 0.04 0.2 0.5 0.05 A 7 3 17 16.7 33- 8.4 116- 0.3 4.8 1.6 18 83 4.3 0.04 0.2 0.8 0.05 A 7 7 49 41.8 33- 3.2 116- 1.2 2.3 1.6 17 72 2.0 0.02 0.2 0.2 0.05 A 7 7 50 39.8 33- 9.0 116- 3.6 10.0 0.5 14 52 4.8 0.07 0.3 0.8 0.06 A 7 7 56 37.8 33-16.3 116- 6.0 2.9 .3.3 17 133 2.2 0.05 0.3 0.5 0.07 8 7 16 17 3 6 33- 2.4 115-56.9 7.0 1.8 16 152 5.2 0.08 0.4 0.8 0.06 8 8 10 0 46.2 33-14.4 116-11.7 10.8 1.9 17 58 4.6 0.03 0.2 0.3 0.04 A 8 10 54 46 2 33-10.3 116- 2.5 6.3 1.7 16 74 6.4 0.14 0.8 2.4 0.15 C 8 13 21 39.4 33- 2.4 116- 2.4 5.1 1.5 17 51 2.0 0.02 0.2 0.4 0.04 A 8 14 54 22.4 33-12.2 116-11.7 1.0 -0.1 9 98 5.3 0.05 0.3 0.6 0.06 8 8 14 54 50.9 33-11.4 116- 9.8 3.8 1.8 17 63 2.5 0.03 0.2 0.8 0.06 A 8 16 23 38.1 33-11.5 116- 9.8 3.1 .3.3 17 62 2.5 0.02 0.2 0.8 0.05 A 8 17 28 15.6 33- 8.3 116- 5.5 6.0 1.8 18 75 2.2 0.03 0.2 0.5 0.05 A 8 22 30 6.3 33-10.7 116- 7.6 6.9 2.1.3.0 17 63 2.7 0.03 0.2 0.5 0.05 A 9 0 16 54.1 33-17.9 116-14.3 4.4 2.2.2.5 17 88 3.5 0.03 0.2 0.6 0.04 A 9 6 47 57.8 33— 2.2 115-57.1 6.2 1.8 18 88 4.7 0.04 0.2 0.5 0.04 A 9 10 21 46.2 33- 6.1 116- 1.9 7.3 .3.6 18 74 3.4 0.04 0.2 0.5 0.05 A 9 20 16 1.5 33- 6.9 116- 3.1 1.3 1.0 18 75 4.0 0.04 0.3 0.5 0.08 A 9 23 35 17.8 33- 7.0 116- 6.7 10.6 1.6 17 84 1.5 0.05 0.3 0.6 0.06 A 10 4 32 25.2 33- 2.1 116- 2.4 5.3 .3.4 18 47 1.8 0.03 0.2 0.4 0.05 A 10 5 28 7.8 33- 2.0 116- 2.4 4.8 2.2.3.3 17 56 1.7 0.03 0.2 0.4 0.04 A 10 6 28 24.8 33- 8.4 116- 3.5 4.4 0.5 17 48 4.2 0.03 0.2 0.8 0.05 A 10 6 28 46.6 33- 7.8 116- 5.0 0.7 1.4 18 44 1.6 0.03 0.2 0.4 0.06 A 10 7 24 9.3 33- 7.8 116- 5.1 0.3* 0.9 17 48 1.5 0.03 0.2 0.07 8 10 7 57 49.1 33- 9.4 116- 5.6 0.5 1.1 17 70 4.1 0.03 0.2 0.4 0.05 A 10 20 16 23.5 33- 5.6 116- 3.1 7.7 1.4 15 83 3.4 0.04 0.3 0.5 0.05 A 10 20 17 49.6 33- 7.7 116- 2.9 4.6 *1.5 15 74 4.5 0.02 0.2 0.6 0.04 A 11 8 10 4.0 33- 2.4 116- 0.3 8.8 .4.2 18 56 4.0 0.05 0.3 0.5 0.05 A 11 8 13 14.0 33- 1.1 115-55.9 6.4 0.7 15 106 4.3 0.09 0.5 1.3 0.11 8 11 8 13 33.7 33- 2.4 116- 0.2 7.3 1.4 17 56 4.0 0.04 0.3 0.6 0.06 A 11, 8 15 10.2 33- 0.1 116- 1.0? 11.6 0.5 8 122 5.0 0.59 2.7 6.1 0.33 C 11 8 15 19.2 33- 2.5 116- 0.1 7.9 0.5 8 138 4.0 0.13 0.7 1.3 0.08 8 11 8 45 55.2 33-14.3 116- 0.7? 0.2 0.6 12 180 0.8 0.10 0.6 0.8 0.07 C 11 8 46 3.6 33- 2.4 116- 0.3 8.5 .3.5 18 56 4.0 0.04 0.3 0.5 0.05 A 11 10 38 31.7 33-16.7 115-58.1 1.3 2.1.3.0 18 199 6.8 0.11 0.5 0.6 0.07 B 11 15 22 26.0 33- 8.3 116- 1.8 6.7 1.2 17 62 6.2 0.04 0.2 0.6 0.06 A 11 15 22 38.2 33- 7.4 116- 4.0 1.4 0.9 9 83 2.7 0.03 0.3 0.8 0.06 A 12 8 43 31.1 33- 2.5 116- 0.2 7.0 0.1 7 109 3.9 0.12 0.6 1.3 0.06 B 12 8 44 54.3 33- 7.9 116- 2.2 1.3 1.0 18 60 5.6 0.03 0.2 0.4 0.06 E 12 8 47 51.4 33- 7.9 116- 2.6 0.1* 0.4 15 56 5.1 0.06 0.4 0.13 C 12 8 48 35.6 33- 4.6 116- 1.9 1.5 0.7 18 58 0.9 0.03 0.2 0.3 0.07 A 12 9 55 21.2 33- 5.0 116- 2.9 7.7 1.7 18 53 2.4 0.03 0.2 0.5 0.05 A 12 14 24 16.3 33-27.7 116-24.4 6.3 2.4.3.3 19 252 13.3 0.20 1.0 0.7 0.08 C rmwww wwwr—p-a cocoa)».- THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 TABLE 8.—Borrego Mountain aftershock—Continued SEC LAT N LONG H DEPTH MAG N0 GAP DMIN ERT ERH ERZ MD 49.6 33- 6.1 116- 2.8 4.2 1.3 18 57 3.9 0.02 0.2 0.6 0.05 8.3 33- 6.1 116- 2.8 2.2 0.9 16 57 3.9 0.02 0.1 0.2 0.04 7.6 33-11.8 116-10.6 3.8 1.9 18 48 1.9 0.02 0.2 0.6 0.04 28.0 33- 9.2 116- 6.5 3.0 1.2 19 47 4.0 0.10 0.6 0.9 0.12 59.7 33- 7.4 116- 5.4 3.2 1.2 18 44 0.5 0.02 0.2 0.5 0.05 19.5 33- 9.6 116- 3.8 6.9 1.4 19 56 5.3 0.03 0.2 0.5 0.05 12.5 33- 4.1 116- 1.9 1.6 1.5 19 40 0.5 0.03 0.2 0.3 0.06 39.5 33- 4.0 116- 2.0 1.6 1.0 19 41 0.6 0.02 0.2 0.3 0.06 4.6 33- 4.1 116- 2.1 0.9 -0.0 8 109 0.7 0.08 0.6 0.5 0.07 11.9 33- 9.9 116- 6.3 2.8 1.0 18 47 4.6 0.03 0.2 0.3 0.05 26.9 33-10.0 116- 6.3 2.6 1.4 16 48 4.6 0.04 0.2 0.4 0.06 45.6 33- 1.2 115-56.3 5.6 0.5 12 101 3.9 0.06 0.4 1.0 0.06 38.9 33-11.0 116- 1.9 4.4 0.1 8 93 5.7 0.08 0.5 1.8 0.06 52.2 33-18.6 116-15.7 8.5 2.4.3.3 18 87 5.5 0.03 0.2 0.4 0.04 47.5 33- 2.0 116- 2.3 4.6 1.6 17 57 7.5 0.03 0.1 1.0 0.03 22.9 33-19.2 [lb-17.8 10.4 0.8 14 117 5.7 0.07 0.5 0.6 0.08 24.3 33-15.6 115-57.3 2.9 1.4 17 199 6.6 0.11 0.5 0.7 0.07 27.0 32-59.4 116-18.7 7.0 1.2 17 201 9.6 0.06 0.4 1.0 0.05 45.7 33- 2.5 116- 1.0 1.9 2.0.2.4 18 52 4.1 0.03 0.2 0.3 0.06 23.8 33-11.8 116-11.4 5.1 1.8 10 137 0.8 0.06 0.6 0.8 0.07 7.0 33- 7.9 115-59.5 4.5 1.7 19 103 4.5 0.03 0.2 0.8 0.05 53.4 33- 8.1 116- 6.6 4.0 1.3 19 49 2.2 0.02 0.1 0.5 0.04 42.0 33- 8.5 116- 4.6 1.1 0.8 18 83 3.1 0.03 0.2 0.3 0.05 18.8 33-11.1 116- 1.4 6.6 0.3 9 137 4.9 0.08 0.5 1.3 0.07 13.9 33- 0.7 116-17.7 3.6 1.0 19 189 7.7 0.05 0.3 2.3 0.06 35.7 33- 2.1 116- 2.3 5.0 1.5 17 50 1.9 0.03 0.2 0.4 0.05 59.8 32-59.7 116-15.6 9.4 0.4 17 180 4.9 0.07 0.4 0.8 0.06 17.0 32-59.6 116-15.6 9.5 1.3 19 180 5.0 0.06 0.4 0.6 0.07 24.8 33- 1.4 116- 6.0 4.2 0.8 19 70 4.0 0.03 0.2 1.3 0.07 45.8 33- 1.6 116- 5.9 3.3 1.0 19 69 3.8 0.04 0.3 1.8 0.09 36.8 33- 9.1 116- 7.1 2.8 1.4 17 50 3.9 0.03 0.2 0.3 0.05 39.1 33- 1.1 115-56.3 7.2 1.5 19 101 3.9 0.04 0.2 0.5 0.04 24.0 33- 9.0 116- 6.5 0.1* 0.6 18 48 3.6 0.03 0.3 0.10 30.4 33- 1.5 116- 6.0 4.1 1.6 16 70 3.9 0.03 0.2 1.5 0.07 49.1 33-15.3 115-57.9 0.3* 1.7 19 195 5.4 0.08 0.5 0.08 34.5 33- 3.5 116- 1.6 6.4 1.2 18 64 1.3 0.03 0.2 0.4 0.05 44.0 33- 3.2 115-42.3 2.0* 2.1.3.0 18 267 14.8 0.18 0.9 0.06 26.6 33— 9.6 116- 6.1 1.9 1.4 18 45 4.5 0.03 0.2 0.3 0.05 22.4 33-12.0 116-10.2 2.9 0.4 14 78 2.5 0.03 0.2 0.3 0.05 36.4 33-15.4 115-58.1 0.3* 1.7 19 179 5.3 0.06 0.4 0.07 0.7 33- 5.9 116- 1.9 6.1 1.4 15 75 3.0 0.03 0.2 0.5 0.05 19.6 33- 3.4 116- 0.9 2.8 1.4 17 90 1.9 0.03 0.2 0.7 0.06 47.2 33- 2.6 116- 0.3 8.1 1.7 18 111 3.8 0.03 0.2 0.5 0.05 43.4 33-18.1 116-11.7 6.5 0.5 12 101 5.1 0.07 0.4 1.1 0.08 14.8 32-59.3 116-13.9 6.1 0.4 13 201 16.7 0.06 0.3 1.7 0.03 50.6 33-16.5 115-59.7 3.1* 1.2 19 177 5.2 0.06 0.4 0.07 1.9 33-11.7 116-10.1 7.1 -0.1 10 84 2.8 0.05 0.3 0.6 0.05 11.4 33- 8.7 116- 4.4 2.2 1.2 18 46 3.5 0.02 0.1 0.2 0.05 59.6 33- 8.7 116- 4.3 2.0 1.0 17 46 3.5 0.03 0.2 0.4 0.07 42.5 33-17.4 116-15.2 4.8 -0.1 9 84 3.3 0.03 0.2 0.6 0.03 0.3 33- 8.6 116- 4.5 1.7 0.2 15 46 3.3 0.03 0.2 0.4 0.06 47.0 33- 3.1 116- 1.1 2.2 0.2 16 94 2.3 0.02 0.2 0.2 0.05 26.6~ 33- 3.7 116- 0.8 2.9 0.2 15 80 1.7 0.03 0.3 0.9 0.06 25.5 33- 2.7 115-59.9 3.6 0.3 17 100 3.9 0.05 0.3 1.1 0.08 17.9 33- 4.3 116- 8.2 8.7 13.2 20 74 6.6 0.04 0.2 0.6 0.05 9.7 33- 3.5 116- 0.9 2.8 1.5 20 58 1.8 0.02 0.2 0.5 0.05 10.5 33- 3.4 116- 0.8 2.9 1.7 19 60 2.0 0.03 0.3 0.8 0.07 39.7 33- 3.2 116- 0.6 4.1 1.8 20 65 2.5 0.02 0.2 0.5 0.05 25.2 33- 3.3 116--0.6 3.5 0.3 13 69 2.4 0.03 0.2 0.8 0.06 36.9 33- 3.? 116- 0.5 3.8 0.5 15 70 2.6 0.03 0.3 0.8 0.07 r. >z>>x>l> l>x>>>fi <91>1>>> D>>O> Ohm)» bbmmb Ombb> m>mmx> mbmbb Dbbb‘) bebbb bbbb) JUN THE 14 14 14 14 22 WOOD BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 41 TABLE 8,—Borrego Mountain aftershocks—Continued SEC LAT N LONG H DEPTH HAS NO CAP DMIN ERT ERH ERZ MD Q 38.4 32-57.3 116-14.7 10.3 1.3 19 209 7.3 0.08 0.5 0.5 0.07 B 32.3 33-17.4 [lb-15.3 3.9 1.6 19 85 3.4 0.03 0.2 0.7 0.04 A 48.4 33- 4.2 116- 8.2 8.6 1.2 18 75 8.5 0.04 0.2 0.6 0%05 A 16.8 33-15.0 115-56.9 0.3* 1.6 19 181 6.7 0.07 0.4 0.09 C 39.0 33- 3.4 116- 1.8 6.5 1.7 19 51 1.7 0.03 0.2 0.5 0.06 A 31.3 33- 4.2 116- 1.6 1.1 .3.2 19 58 0.2 0.04 0.3 0.4 0.10 A 25.2 33- 3.9 116- 1.6 0.6 0.7 15 79 0.6 0.07 0.4 0.6 0.10 A 21.3 33- 4.1 116- 1.6 0.9 1.4 18 78 0.3 0.04 0.3 0.4 0.09 A 15.3 33- 4.0 116- 2.1? 0.5 0.2 10 109 0.8 0.07 0.6 0.7 0.11 B 56.1 32-59.8 116-15.1 7.0 1.3 18 200 4.2 0.14 0.8 1.3 0.09 8 39.4 33- 2.1 115-59.6 4.8 73.0 19 85 4.7 0.03 0.2 0.6 0.05 A 4.3 33- 3.5 116- 1.0 2.9 1.6 15 81 1.6 0.02 0.1 0.2 0.04 A 55.4 33—18.6 116-13.4 7.5 .4.4 13 112 4.9 0.08 0.5 1.2 0.07 A 23.2 33-18.8 116-13.2 5.6 1.4 15 113 5.3 0.05 0.3 1.1 0.06 A 38.9 33- 1.1 116-16.1 0.3* 0.9 12 198 5.2 0.14 0.9 0.11 C 2.9 33° 9.2 116-24.4 9.8 1.7 13 227 7.1 0.07 0.4 0.4 0.04 C 0.1 32-58.8 116-15.8 7.8 1.3 18 206 6.0 0.10 0.5 1.0 0.07 B 22.9 32-53.5 116-13.l? 7.5 1.7 19 221 13.7 0.13 0.8 1.7 0.10 C 47.5 32-57.1 [lb-18.2 10.1 1.4 19 224 11.0 0.11 0.6 0.6 0.07 8 58.7 33- 0.8 [lb-16.8 1.3 0.9 18 203 6.3 0.11 0.5 0.7 0.08 8 4.6 32-59.4 116-16.8 9.5 1.4 19 209 6.9 0.08 0.5 0.6 0.07 8 10.7 33- 4.7 116— 1.9 4.8 0.3 17 58 0.8 0.05 0.4 0.9 0.11 8 43.8 33- 6.4 116- 2.6 2.4 1.0 20 55 4.3 0.03 0.2 0.3 0.06 A 25.6 33- 6.9 116- 2.6 6.5 1.5 19 54 5.1 0.04 0.2 0.8 0.07 A 57.1 33- 3.1 115'59.1 4.9 2.1.2.6 19 69 4.5 0.02 0.1 0.5 0.04 A 22.2 33- 3.1 [IS-59.0 5.1 2.0.2.5 19 70 4.6 0.04 0.3 1.0 0.07 8 55.8 33-18.8 116-13.2 3.0 1.8 20 101 5.4 0.03 0.2 2.1 0.05 B 48.5 32-59.4 [lb-17.0 9.3 1.4 19 210 7.1 0.08 0.5 0.6 0.06 8 16.5 33- 0.1 116-16.5 7.5 1.0 19 204 6.0 0.12 0.6 1.3 0.09 B 41.5 33- 2.6 116-11.9 6.5 1.2 20 104 3.5 0.06 0.3 0.8 0.06 A 36.8 33-14.5 116-11.8 8.1 1.7 18 59 4.3 0.04 0.2 0.5 0.05 A 37.2 32-57.1 116-18.3 10.3 1.8 20 225 11.1 0.13 0.7 0.7 0.08 C 32.7 32-57.1 116-18.4 9.9 1.1 20 225 11.2 0.20 1.1 1.2 0.10 C 36.3 33- 7.4 116- 2.5 6.8 1.9 20 52 5.1 0.03 0.2 0.5 0.05 A 56.8 33- 6.9 116-16.4 2.8 1.6 20 129 4.0 0.05 0.4 0.7 0.07 B 28.1 33- 2.4 116- 2.4 4.4 y3.2 17 91 3.6 0.02 0.2 0.6 0.04 A 8.6 33-17.3 116-15.3 4.7 1.6 19 85 3.2 0.02 0.2 0.5 0.04 A 17.9 33- 8.4 116- 3.6 4.4 0.1 17 57 3.9 0.04 0.2 1.0 0.06 A 25.8 33- 4.6 116- 1.7 2.1 1.1 18 72 0.6 0.02 0.2 0.2 0.05 A 27.3 33- 2.9 115-59.8 7.0 1.3 19 73 3.8 0.04 0.2 0.6 0.06 A 57.3 33- 2.6 116‘ 0.2 7.4 1.9 21 77 3.8 0.03 0.2 0.4 0.04 A 26.6 33- 9.4 116- 5.8 6.9 1.7 20 44 4.] 0.03 0.2 0.5 0.05 A 13.1 33- 7.9 116- 5.1 7.7 1.3 19 48 1.7 0.03 0.2 0.5 0.05 A 30.3 33-17.2 116-15.4 3.9 2.1 18 86 3.3 0.02 0.2 0.6 0.04 A 55.1 33- 7.1 116- 4.3 2.1 1.0 18 56 2.2 0.03 0.2 0.4 0.07 A 15.4 33- 1.2 [lb—13.6 3.6 -0. 6 209 1.4 0.34 1.6 2.8 0.03 C 58.2 33- 2.6 116- 0.3 7.6 1.9 17 78 3.7 0.05 0.2 0.6 0.05 A 2.1 33- 3.6 115-55.8 3.5 .3.2 18 55 3.2 0.02 0.2 0.8 0.04 A 6.0 33-10.0 116- 3.6 6.3 1.7 17 60 6.2 0.04 0.2 0.7 0.05 A 3.3 33- 3.7 115-55.7 3.4 1.8 17 56 3.0 0.03 0.2 0.8 0.04 A 32.4 33-10.0 116- 6.3 1.2 1.2 18 48 4.6 0.03 0.2 0.4 0.06 A 50.1 32-57.7 116-15.6 5.6 1.2 18 211 7.4 0.09 0.5 1.4 0.07 8 7.2 32-58.1 116-15.3 6.6 1.1 15 208 6.6 0.14 0.8 1.9 0.09 C 43.4 33- 9.7 116- 5.6 8.8 0.4 16 57 5.7 0.05 0.2 0.7 0.06 A 60.0 33- 9.6 116* 5.6 8.9 1.4 16 57 5.7 0.03 0.2 0.5 0.04 A 22.1 33- 6.4 116- 1.3 2.6 1.5 19 67 4.1 0.02 0.1 0.2 0.04 A 43.9 33- 2.6 116— 0.2 7.2 .3.3 19 84 3.8 0.03 0.2 0.5 0.05 A 13.5 33- 2.6 116- 0.2 7.8 1.7 20 77 3.7 0.04 0.3 0.6 0.05 A 38.8 33- 5.1 116- 2.9 5.1 1.3 20 54 2.5 0.04 0.2 0.7 0.06 8 57.8 33- 5.2 116- 2.8 6.2 1.4 19 55 2.4 0.03 0.2 0.6 0.06 A 42 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 TABLE 8.-——Borrego Mountain aftershock—Continued 1968 HR MN SEC LAT N LONG H DEPTH MAG N0 GAP DMIN ERT ERH ERZ MD JUN 3 6 2 19.6 33- 3.6 [IS-58.5 4.6 1.2 21 66 5.1 0.08 0.5 2.0 0.09 3 6 4 56.6 33-18.9 116-13.2 3.0* 2.1.2.5 21 102 5.5 0.03 0.2 0.05 3 6 6 26.1 33- 1.4 [IS-55.0 7.3 0.6 20 68 5.7 0.05 0.3 0.8 0.07 3 9 10 22.4 33-11.7 116- 9.7 7.7 .3.3 20 47 3.0 0.04 0.2 0.5 0.06 3 9 11 42.4 33-11.9 116- 9.9 6.4 0.2 13 79 2.9 0.08 0.5 1.1 0.08 3 10 47 17.4 33- 4.2 115-57.9 5.5 1.7 21 77 5.3 0.03 0.2 0.7 0.05 3 12 31 37.3 33-11.7 116- 9.5 7.7 2.1.2.6 21 47 .2.9 0.04 0.2 0.6 0.06 3 14 16 56.9 32-52.0 116- 1.0? 5.0* 2.0 15 155 7.1 0.33 2.6 0.47 3 21 28 34.9 33- 7.8 116- 2.1 1.8 1.1 20 52 6.7 0.03 0.2 0.3 0.06 3 21 31 36.8 33- 7.9 116- 2.1 2.0 1.1 21 53 5.8 0.03 0.2 0.3 0.06 3 21 32 31.4 33- 7.8 116- 2.1 1.7 0.2 16 61 5.7 0.02 0.2 0.3 0.05 3 21 34 56.4 33-11.7 116- 9.9 7.0 0.8 19 47 3.0 0.03 0.2 0.4 0.05 3 21 36 0.9 33- 7.8 116- 2.0 1.7 0.6 20 53 5.8 0.03 0.2 0.3 0.06 3 22 25 51.0 33-16.1 115-59.5 3.4 1.6 20 176 4.6 0.06 0.4 1.9 0.07 4 2 14 16.7 33- 2.5 116- 2.3 5.3 1.6 20 67 2.3 0.03 0.2 0.5 0.05 4 3 48 19.7 33- 2.4 116- 2.3 5.1 93.2 21 69 2.2 0.03 0.2 0.4 0.04 4 13 37 42.2 33-16.4 115-59.6 0.9 2.0 19 176 5.0 0.09 0.4 0.7 0.07 4 13 38 44.4 33-15.8 115-58.3 1.9 0.5 16 180 5.4 0.10 0.4 0.6 0.07 5 1 18 49.1 32-58.8 116-15.0 7.6 0.8 16 203 5.2 0.07 0.4 0.8 0.05 5 23 23 39.2 33- 0.9 115-55.6 9.6 1.7 20 76 7.7 0.05 0.3 0.6 0.07 6 5 27 46.4 33- 8.4 116- 0.4 4.7 1.6 20 67 4.4 0.03 0.2 0.7 0.06 6 6 13 48.5 33- 2.4 116- 2.4 4.6 1.8 20 66 2.1 0.03 0.2 0.6 0.05 6 9 12 2.0 33- 0.6 115-55.7 6.4 .3.2 18 77 8.3 0.04 0.2 0.8 0.04 6 13 18 4.9 33- 0.8 115-55.6 7.6 .3.1 20 76 8.0 0.04 0.2 0.7 0.04 6 16 28 11.2 32-57.9 116-12.5 2.4 0.4 8 200 5.5 0.29 1.6 1.8 0.13 6 16 29 6.9 32-58.7 116-12.6 2.7 0.5 11 194 4.1 0.12 0.6 0.9 0.07 6 22 41 24.9 33- 7.8 116- 6.0 1.5 1.0 19 56 1.1 0.03 0.2 0.3 0.07 7 2 18 13.1 33- 8.7 116- 3.6 6.1 1.3 20 50 4.3 0.03 0.2 0.6 0.06 7 13 1 56.8 33- 1.4 115-56.4 7.7 1.7 21 68 4.1 0.03 0.2 0.5 0.04 7 13 36 50.1 33- 6.3 116-11.3 8.2 1.6 20 86 7.8 0.04 0.2 0.6 0.05 7 13 46 39.2 33-27.4 116-22.0 9.8 1.7 19 237 10.8 0.23 1.3 0.7 0.12 7 16 13 55.3 33— 8.3 116- 3.6 4.6 1.7 19 47 3.9 0.02 0.2 0.6 0.05 8 16 14 10.0 33- 1.4 115-56.2 6.4 2.0.2.6 20 69 4.2 0.03 0.2 0.5 0.04 8 16 19 6.3 33- 1.5 115-56.2 6.2 1.6 20 68 4.4 0.04 0.2 0.6 0.05 8 20 51 18.8 32-58.5 116-15.4 7.9 1.5 21 207 6.1 0.09 0.5 1.0 0.07 8 21 37 0.9 33-15.9 116- 2.5 1.3 2.3.3.1 21 162 4.1 0.05 0.3 0.4 0.07 9 0 15 11.8 33-14.1 116- 0.6 0.4* 1.5 21 162 0.8 0.04 0.3 0.07 9 6 0 49.1 33-25.2 116-21.3 10.6 2.0 17 290 6.6 0.31 1.6 0.8 0.09 9 14 20 15.6 33-16.8 115-59.5 2.9 .3.1 18 211 5.6 0.14 0.6 0.8 0.09 9 14 48 29.6 33- 4.6 [IS-55.5 5.1 93.0 19 84 1.6 0.03 0.2 0.5 0.04 11 2 22 7.0 33- 1.8 115-58.6 6.6 -0.2 11 92 3.5 0.08 0.4 1.0 0.06 11 2 22 34.7 33- 2.6 116- 2.3 5.3 1.8 20 63 2.4 0.03 0.2 0.5 0.05 11 5 32 17.3 33-20.9 116-20.0 10.1 2.5.3.1 21 185 2.5 0.10 0.6 0.5 0.10 11 5 33 37.8 33-21.3 116-20.8 13.0 0.6 8 194 2.8 0.20 0.8 1.5 0.06 11 19 28 11.1 32-57.2 116-17.5 11.5 1.3 11 220 10.1 0.11 0.7 0.8 0.07 11 21 41 7.6 33- 2.1 115-59.2 8.3 1.5 20 84 4.5 0.05 0.3 0.6 0.06 12 1 9 47.9 33-17.1 116-16.3 6.3 0.1 12 75 2.7 0.04 0.2 0.5 0.04 12 1 9 52.5 33-16.3 115-59.6 0.4 1.6 21 176 4.9 0.08 0.3 0.6 0.07 12 12 31 8.0 33- 0.1 116-14.8 6.0 0.4 15 196 3.5 0.09 0.5 1.1 0.09 12 12 31 33.3 32-57.0 116-15.4 5.2 0.6 18 213 8.3 0.09 0.5 1.8 0.06 12 12 34 26.4 33-17.1 115-57.2 1.1 1.0 19 219 8.4 0.19 0.7 1.0 0.10 12 13 30 29.3 33- 2.1 115-56.6 4.9 0.9 33 53 2.0 0.02 0.2 0.4 0.05 12 13 38 11.6 33-10.6 116- 5.5 9.8 1.2 21 54 5.9 0.04 0.2 0.5 0.06 WEI-TOD) banana m>>>n >>>>co OWG‘DP >mma=> >WW>D coma)» abbom 0 man)» l>>n THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 TABLE 8.—Borrego Mountain aftershocks—Continued EARTHQUAKES SOUTHEAST 0F PORTABLE SEISMOGRAPH NETWORK 1968 HR MN SEC LAT N LONG W DEPTH APR 18 0 0 25.2 32-55.6 115-40.5 5.0* 23 15 42 41.1 32-45.9 115-32.2 5.0* 23 16 23 52.4 32-48.2 115-38.7 5.0* 23 16 24 09.1 32—46.3 115-35.8 5.0* 23 16 50 41.1 32146.8 115-33.3 5.0* 24 12 32 31.1 32—49.2 115-39.5 5.0* MAY 14 9 11 18.7 32-57.6 115-30.3 5.0* 43 NOTE. —List of Borrego Mountain aftershocks for which loca- tions were determined. For each event the following data are given: Origin time in Greenwich mean time: date, hour (Hr), minute (Mn), and second (Sec). Epicenter in degrees and minutes of north latitude (Lat (N.)) and west longitude (Long (W.)). Poor convergence of the epi- solution is indicated by “7”. DEPTH=depth of focus, in kilometers. Assumed depth is indi- cated by “*”. X-MAG=earthquake magnitude. Value on left determined from USGS data (see discussion in text); value on right quoted from Caltech bulletin (Nordquist and others, 1972). NO=number of stations used in locating earthquake. GAP=1argest azimuthal separation, in degrees, between stations. DMIN=epicentral distance, in kilometers, to the nearest station. ERT=standard error of the origin time, in seconds. ERH=standard error of the epicenter, in kilometers. [= vml, where SDX and SDY are the standard errors in latitude and longtitude, respectively, of the epicenter. ERZ=standard error of the depth, in kilometers. MD = mean deviation of the time residuals. [=JN—loei], where R; is the observed seismic wave arrival time less the computed time at the ith station. 2 km, but this uncertainty depends strongly on the distance from an epicenter to the nearest station. The crustal model used in locating the aftershocks, which was derived from the special refraction survey (Hamilton, 1970), is summarized below: Velocity (km/sec) 2.5 Thickness Depth to top m} (kin) 0.4 0.0 2.5 .4 1.1 2.0 1 0 14.0 25.0 T1995” {DF‘CDPd MAG N0 ERT ERH MD ,3.2 14 0.11 1.4 0.11 16 0.43 3.0 0.09 ,3.1 18 0.14 1.4 0.14 ,4.1 17 0.15 1.4 0.15 16 0.60 4.0 0.11 ,3.4 18 0.22 2.1 0.26 ,2.6 21 0.15 1.7 0.23 Q = solution quality of the hypocenter. This measure is intended to indicate the general reliability of each solution: Q Epicenter Focal depth A ............ Excellent ............... Good B ............ Good ...................... Fair C ............ Fair ....................... Poor D ............ Poor ...................... Poor Q is based on both the nature of the station distribution with respect to the earthquake and the statistical measures of the solution. These two factors are each rated independently ac- cording to the following scheme: Station distribution N0 CAP DMIN 2 8 S 120° S DEPTH or 5 km ...2 6 S 150° S 2XDEPTH or 10 km 2 6 S 225° S 50 km 2 4 S 180° S 50 km D ............... Others ..................... SIuIiinral metuurex ERHHm) ERZtkm) MD(sec) RMAX (.w(‘)' A ....................... S 1.0 S 2.0 S 0.10 S 0.25 B ....................... S 2.5 S 5.0 S 0.20 S 0.50 C ....................... S 5.0 ........ S 0.30 S 0.75 D ....................... Others .......................... 'RMAX is the maximum residual. Q is taken as the average of the ratings from the two schemes; that is, an A and a C yield a B, and two B’s yield a B. When the two ratings are only one level apart, the lower one is used; that is, an A and a B yield a B. Station time corrections against this model are given in table 7; a correction is subtracted from the ob- served arrival time. The corrections were applied to all arrivals, a procedure that is not strictly accurate because the corrections were obtained from refracted Pg rays. Thus, direct P-wave arrivals would not be appropriately corrected; therefore, this method leads to errors mainly in focal-depth estimates for earth- quakes above the Pg refractor and in the vicinity of a station having a large correction. For example, shallow shocks near station 17 probably are actually 44 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 somewhat deeper than was computed. It does not appear, though, that the method of applying station corrections seriously distorted the hypocenter pat- tern. AFTERSHOCK DISTRIBUTION The general distribution of the Borrego Mountain aftershocks in relation to mapped faults is shown in figure 12. The aftershock distribution with respect to surficial fracturing is shown in figures 13 and 14. In these three figures, epicenters are plotted for all earthquakes with a mean deviation of the time resid- uals (MD) 50.05 sec that are given in table 8. Figure 12 shows that a large part of the after- shock activity occurred in a band subparallel to the northwest-trending fault system. The epicenter of the main shock is near the center of this band, although many more aftershocks occurred to the southeast of it than to the northwest. The concen- trated activity in the band southeast of the main shock ends at a point about 5 km east of station 10 and about 28 km southeast of the main shock. North- ' west of the main shock, the band extends to near station 12. The epicenter 3 km west-southwest of station 12 lies 28 km from the main shock. At its southeast end, the main seismic zone swings toward the Superstition Hills fault, where movement exceeding 1 cm occurred near the time of the main earthquake (Allen and others, this volume). Three earthquakes occurred on the Superstition Hills fault. Five approximately colinear epicenters were located about 5 km southwest of the Superstition Mountain fault. The seven epicenters in the azimuthal quadrant southeast of station 15, outside the portable seismo- graph net, were located using readings from both the portable stations and Caltech regional stations. The time corrections for the Caltech stations that were derived by Allen and Nordquist (this volume) were used. The earthquakes are given at the end of table 8. No earthquakes were detected near the San Andreas fault; this fault also moved on the order of 1 cm (Allen and others, this volume). Most of the aftershocks that are not in the main band lie in two clusters centered about 16 km north- east and 22 km southwest of the middle of the main band. A more detailed View of the epicenter pattern and its relation to the surface faulting is presented in figure 13. The surface faulting associated with the Borrego Mountain earthquake can be divided into two more or less continuous strands that come to- gether in a complex pattern of faulting at the Geo- tillo Badlands, a group of low hills in the vicinity of station 9. Initial maximum horizontal displacement, which was in the right—lateral sense, was 38 cm on the northwest strand, near the main-shock epicenter, and 20 cm on the southeast strand (Allen and others, 1968; Clark, this volume). Movement on the south- east strand continued Iong after the main shock: postearthquake displacement of 8 cm occurred be- tween April 25, 1968 and May 4, 1969 (Burford, this volume), and surface cracks continued to de— velop at least until January 1970 (Clark, “Surface Rupture Along the Coyote Creek Fault,” this vol- ume). No evidence of continued movement was found on the northwest strand. The aftershocks in the main band are mostly along the surface fracture or to the northeast of it. The concentrated activity along the southeast fault strand extends about 4 km beyond the surface break. To the northwest, the earthquake 3 km west—southwest of station 12 lies about 20 km from the nearest surface fracture. The southeast fault strand was far more active seismically than the northwest strand during the period of this study. Along the southeast strand, the epicentral zone varies in width from over 5 km, near the region of complex faulting, to about 1 km at the center of the strand (near station 17). Farther to the southeast, the zone is several kilometers wide. On its southwest margin, the seismic zone along the southeast strand is rather sharply defined. Seismic activity on the northwest fault strand is concentrated primarily in one cluster near the north- west end (near station 18). Another cluster of epi- centers lies about 10 km northwest of the limit of recognized surface faulting, and there was scattered activity even farther to the northwest. Relatively few aftershocks were located near the epicenter of the main shock. Variations in the epicenter pattern with time are shown in figure 14. The pattern was fairly stable; however, two variations are evident, both of which occurred after the first week of monitoring. One is the increased level of activity in the areas northeast and southwest of the main aftershock band; the other is the eastward extensions of the southeast end of the main band by about 3 km. The depth distribution of the aftershocks is shown in figures 15 and 16, where earthquakes having ERH <1 km and ERZ <2 km (table 8) are plotted in ver- tical sections along the lines indicated in figure 13. The aftershocks are distributed from near surface to about 12 km. Depth determinations for after- THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 45 33 23.0 33 23.0” g 74 n2 , i z x ,6 Z X X X x x x x "‘ )‘Q ‘5 x ‘ x II A x: x u‘ H x x ’2‘ x X X X X ‘( ‘ y“ x x X X 9 + ‘6 x + X + Xxx X x x x Xx x i )‘Q‘ i x xx x x g‘ x x Ab . -r\ \ \. mg; "gar, MAIN SHOCK )3 ‘" \ x a xxx./ x \\ x x xx QV 8A\\X ‘ X xx x X X X A? x X,“ x \/ + \R» xxxyfikg' ’5‘ x x + )\ "k ’5‘ x x x” X ’K . X # X \gg‘&‘§'&x Xxx XXX ’“ x ’ $ A2 x9 \ I x x ,2‘ ) x V { \IxKx x xx X XXX X X ”,8 g x w x " 4A x I, x Q X X +‘( I xx‘x’éfiéz x )‘ {wk x x x it X“ x 5‘ xx x «3 Sx‘fi’r’x‘ + I + ‘22 #:A‘xxg Xxx xx x if X xx )ll‘ xx q x O x X X & xx AI \f\ X I: ~( x 3 x * x \, XAIO x O x m x y x x x x 65 x x Xxx x )l x x x 0 5 '0 KM ... "' I__A_I_I_L_H_A_J _. E; x x m m X x :- 632 55.0 32 55.09 FIGURE 13.— Detailed map of aftershock epicenters (crosses), ground breakage (after Clark, this volume), epicenter of main shock (Allen and Nordquist, this volume), and lines of vertical sections for figure 15 (A—B, C—D) and figure 16 (E—F, etc.) THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 #33 23.0 33 23.0__ #33 23.0 x 3323.0 '5; )( G 3 G 53 ii." R’ i .o x o c: X x .0 x XX x X X X xnxfl x x xx x x xi 3“ x 50‘ x x § \. ‘\ .‘\\x ‘.\x‘ \ x \\ g X x ‘2 x x5(&x K x x waif; if} xx “V" «Ax X x 0 & xkx x X xx 15.)“ "ch"?! x u 1* .2 . v w x \ x Xxx x Z Z Z X '“ 03 x x L71 (7) tn IN) 01 '0 0'! M J- N X J: 'o X — / Q: 'o x —‘ '0 3255.0 4/12 4 18 3255.0 3255.0 4/19 5/2 3255.0 33 23.0 3323.0 3323.0 33 23.0 S 53x 5 N MN x 01 N x an r. 2:, x 6:: 'o x )3 x xx x xi x ‘ x X x xx x 1i X x *x x X x x .I I\ - 1 ~\ \\\| \‘\\ 5x336? ‘\\\~ ‘ x“ \ \ X x \ x \ \ \ xgxx x \‘ xx‘ x . r x x \- \ x X i x x x u‘ § x x 'xx x XXX )1 ‘ ff X! )s \\ 9“ x . \ V x ‘\ X! X ‘\ x ~)(‘\ ‘\\ x\X \uk X ' I! \ x “43* ~ * x ‘ l i Q r ‘ ~. Xx X X ‘ ‘\ ix X \_\ “fix I K . x X X X X ‘ x X "‘ ‘ Z : x X Z :3 UV tn Xx x <11 x x 2' £ 1 “ >x :2 :3 — :2 :3 / — /12 ‘o 3255.0 5/3 5/23 -3255.0 3255.0 5 2H 6 3255.0 FIGURE 14. — Variation of epicenter pattern with time. THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 47 DISTANCE; (KM) 30 4O 50 B DEPTH (KM) FIGURE 15.—Vertical sections through aftershock zone. Arrows indicate the approximate position of ground breakage on the line of section. Maximum projection distances of foci are: A—B, 5.00 km; C—D, 30.00 km. Lines of section are shown in figure 13. shocks in the main band of activity along the fault are generally good; they are thought to have a precision of better than 2 km. The depths for the shocks in the side clusters, on the margin of the seismograph net, are not so reliable. The minimum depth of aftershock occurrence varies and is related to the surface faulting. On the southeast fault strand (fig. 15, section A—B, distance range 30—50 km), where continued movement was observed, aftershocks occurred near the surface. The aftershocks beyond the southeastern limit of ob— served faulting occurred only as shallow as about 4 km. The aftershocks just beyond the northwestern limit of observed faulting are below 7 km. Farther northwest (between stations 5 and 6), foci are again near the surface, but no surface breakage was ob- served. This zone of activity lies in a thick section of Tertiary sedimentary rocks. At the northwest boundary of this zone the minimum depth of foci sharply increases to about 10 km under Clark Valley. The vertical sectionsperpendicular to the main epicenter trend (fig. 16) reveal several interesting characteristics of the aftershock distribution. The aftershock zone of each section generally lies toward the right, that is, toward the northeast of the surface break, and usually ascends to its minimum depth near the surficial fractures. In section I—J, which crosses the zone of complex surface faulting, the aftershock zone is about 4 km thick and dips about 60° NE. In section M—N, except for the small cluster of earthquakes at 5 km depth near the left-hand margin, the aftershock zone is within about 1 km of a vertical line extending to a depth of 9 km. A histogram of the depth distribution of all the aftershocks in table 8 that are‘assigned a quality of A or B is shown in figure 17A. An A or B rating means that the focal depth is relatively well deter- mined, probably with a precision better than 2 km. The histogram shows that the foci are fairly evenly distributed to a depth of about 8 km, below which there is a steady decline to none below 13 km. The relatively low number of foci above 1 km could be the result of a systematic error in the hypocenter determinations. , The possibility that larger and smaller earth- quakes have different depth distributions is examined 48 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 DISTANCE (KM) O E 2 4 .6 8 IO I2 F ( 2 4 6 8 IO I2 I. 1' x x v x xxx x 2 ' . “05‘ a. I, x x‘v‘ ‘ xx x ‘( , x 4 x x I; r)( x! ‘I x ”(x x 6 x ‘ xl ‘ x xx ‘1 x x x xx“ xx x s ‘ .x' ‘ ‘1‘ “f. x)( ,( x I0 I i X x X x x I! x 1 I2 x I4 06 V H M V N I 2 x 339%. {x 1:)? E 4 x)! 5 . 3w %. 2' E 6 ,3: ’I I x”; x g 8 ‘k, . . X D x x IO :1 I2 I4 Xx; ’13": x 2 x “(xh’fi xx x “ax . . . ‘2 x g x ‘ 2 x ”K lg xx x x x I x‘ xi lgIx :X x 6 5:51 x x x 5:" x {k x xx x " 3“ ”I 8 . ‘ I" x * ‘I‘Ix‘xx‘? ‘ * *9‘& IO x x x? I I2 I4 FIGURE 16.—Vertical sections through aftershock zone. Ar- rows indicate the approximate position of ground breakage on the line of section. Maximum projection distances of foci are: E—F, 11.42 km; G—H, 4.24 km; I—J, 3.92 km; K—L, 2.47 km; M—N, 2.40 km; and 0~P, 4.73 km. Lines of section are shown in figure 13. in figures 17B and C. In figure 173, the A— or B-rated earthquakes that are assigned a magnitude by the U.S. Geological Survey are plotted in four magnitude categories. No strong trend is apparent. The A- or B-rated earthquakes that are assigned a magnitude by the California Institute of Technology are plotted in figure 170. Although there are few earthquakes in each category, the data show an increase in average depth with magnitude. Most impressive is the fact that none of the nine earthquakes of magnitude 3.5 or larger occurred shallower than 6 km. One must question whether any variation observed could result from some systematic error in the hypo- center determination procedure that depends on mag— nitude. The main difference in the data between NUMBER OF EARTHQUAKES {A/ 29 . 49 6.0 I2] {B/ML <. LO LO SIML<-o.4- MR-0.02+O.28 TC | I I 1 I 1 1 -o.4 o 0.4 0.8 Station time correction (sec) FIGURE 21. — Average P-magnitude and average X-magnitude residuals (MR) plotted against station time correction (TC). A large time correction means that waves arrive late. The lines were fitted by the least-squares method. slopes of 0.46 and 0.28 for the P and X magnitudes, respectively. Variations in the station-time correction are re- lated directly to variations in basement depth in the Borrego Mountain region (Hamilton, 1970). The relation in figure 21 suggests that the magnitude variations may also be related to basement depth. Although depth to basement is not the sole control on magnitude anomalies, it is nevertheless a very important one. The importance of the effects of near-surface lithologic variations on controlling the excitation in the coda of the seismic signal was demonstrated by Aki (1969). Variations in thickness of sediments in the San Francisco Bay region have been related directly to variations in amplification of horizontal and, to a lesser degree, vertical motion from underground nuclear explosions at the Nevada Test Site (Borcherdt, 1970). 52 RATE OF AFTERSHOCK OCCURRENCE VERSUS TIME The rate of earthquake activity in the Borrego Mountain sequence was analyzed by using a continu- ous slow-speed (37.5 mm/min) playback of the tapes from station 1. This station was chosen because it had three components'of motion recorded, thus making S-wave identification easier, and because it was situated to the side of the main aftershock area (fig. 12), thus minimizing effects arising from varia- tions in epicentral distance. Almost all seismic activ- ity in the sequence was within 40 km of station 1 and, except for the cluster of activity southwest of the main zone, was at distances beyond 15 km. The earthquakes counted had an S-P time between 1 and 10 seconds. This restriction eliminated events that occurred outside the aftershock region and those that occurred very near station 1. Events that pro- duced a signal so large that the S waves could not be recognized were counted; though some of them may have had an S-P time outside the range specified, such shocks made up less than 1 percent of the total. The events counted were also required to have an amplitude above a specified level; the amplitude threshold corresponded to a magnitude 1 earthquake at an epicentral distance of 20 km, the distance to the main aftershock area. Owing to recorder stoppages, a continuous record was not obtained. The longest stoppage was for 5 days; other stoppages were less than 1 day’s dura- tion. The total time lost was 10.6 percent of the recording period. If less than 24 hours of record was obtained in a day, the count was adjusted by dividing by the fraction of the record obtained. A total of 13,412 earthquake signals was counted. There were 529 other signals that may have been caused by earthquakes, but they also could have been noise; therefore, they were not included. Plots were made of log 71 against t, where n is the number of earthquakes per day and t is the day after the main shock, and of log 7; against log t. Straight lines fitted by least squares to these plots had the equations 112836 exp (—0.185t) , and (1) n:5,130 t‘°~93. (2) The observed variations in the rate of aftershock occurrence lies between these two relations, but the second one is the better approximation of the two. The data for equation 2 are shown in figure 22. Such a time dependence has been observed for the rate of aftershock occurrence in numerous other sequences. RELATED STUDIES IN OTHER AREAS Detailed earthquake studies using dense seismo— graph networks have been carried out in recent THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 l I lllll] I ll IOOO > _ o _ 'o _ \ _ m _ a: x _ o a u- - .c 4— I- O -1 a; : I00 llllll 1 11 I0 I00 1*, days I l FIGURE 22. —— Rate of earthquake occurrence, n, plotted against time after the main shock, t. Open circle indicates that n is based on less than half a day’s record and is therefore of lower reliability. years along a number of other sections of the San Andreas fault system in California from the Imperial Valley to north of San Francisco Bay. The maximum depth determined for hypocenters in all but one of these studies is within a few kilometers of the 12-km maximum depth found for the Borrego Mountain aftershocks. The areas studied include (1) southeast of Anza (Arabasz and others, 1970), (2) Lake Hughes and Cajon Pass (Brune and Allen, 1967), (3) Parkfield-Cholame (Eaton, O’Neill, and Mur— dock, 1970), (4) Bear Valley (Eaton, Lee and Pakiser, 1970), (5) Watsonville (Udias, 1965), (6) Corralitos (McEvilly, 1966), (7) the San Andreas, Sargent, Hayward, and Calaveras faults in the San Francisco Bay area (Eaton, Lee, and Pakiser, 1970), (8) Antioch (McEvilly and Casaday, 1967) , and (9) Santa Rosa (Unger and Eaton, 1970). Exceptional focal depths were found in the San Gorgonio Pass area, where earthquakes occur as deep as 21 km (Hanks and Brune, 1970). The previously studied area nearest the Borrego Mountain area lies at the trifurcation of the San Jacinto fault about 10 km southeast of Anza and 50 km northwest of the main Borrego Mountain earthquake (Arabasz and others, 1970). One hun— dred earthquakes, located by using as many as six THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 53 seismograph stations, occurred in a dispersed pattern extending several kilometers outside the area out- lined by mapped faults. The complexity of the epi- center pattern indicates activity on more than one fault in the area. The aftershock and creep patterns in the Borrego Mountain area reveal a similar be- havior. The dispersed tectonic activity observed near Anza and Borrego Mountain reflects the locally complex and splaying nature of the San Andreas fault system in southern California. In contrast, the aftershocks near Parkfield in central California (Eaton, O’Neill, and Murdock, 1970) mostly oc- curred within 1 km of a near-vertical plane; the intersection of this plane with the surface almost coincides with the surface faulting along the rela- tively simple and single-stranded trace of the San Andreas fault in that area. SUMMARY AND TECTONIC IMPLICATIONS The epicenters of the Borrego Mountain after— shocks define a rather complex pattern consisting of a main band subparallel to the regional fault trend and two side clusters. Most of the earthquakes in the main band occurred in a distance interval of 56 km and are approximately centered on the epicenter of the main shock. The symmetrical extent of the epi- center band with respect to the main-shock epicenter suggests that the Borrego Mountain earthquake re- sulted from bilateral faulting (Allen and N ordquist, this volume). It also seems that the rupture started near the bottom of the fault: the aftershocks suggest that the fault extends to a depth of about 12 km, and the computed depth of the main shock was 11.1 km (Allen and Nordquist, this volume). This depth de- termination cannot be considered very reliable be- cause the nearest station was 47 km from the main shock; however, it is supported by the observation that all nine aftershocks with magnitudes over 3.5 and with well-determined focal depths occurred below 6 km deep. Although the extent of the aftershock zone was apparently symmetrical with respect to the main shock, the level of aftershock activity was not. Most of the aftershocks occurred southeast of the main shock, where part of the zone of activity was closely associated with surficial faulting. The southeast sec— tion of the surficial faulting was the site of continued displacement long after the main shock (Burford, this volume; Clark, “Surface Rupture Along the Coyote Creek Fault,” this volume). The continued fault displacement following the 1966 Parkfield earthquake was thought by Smith and Wyss (1968) to result from the viscous properties 476-246 0 - 72 - 5 of the near-surface materials. This explanation could probably also be applied to the Borrego Mountain earthquake; however, the southeasterly growth of the aftershock zone, with earthquakes in the depth range of 4 to 10 km, indicates further fault move- ment in the basement. The additional displacement on the surface fractures could then have resulted from growth of the original faulting. Although part of the aftershock activity was asso- ciated with surficial fractures, most of the earth- quakes occurred northeast of the fractures. To a certain extent, the surficial fractures bound the main aftershock band on the southwest. Wyss and Brune (1971) found that the Borrego Mountain aftershocks had higher stresses associated with them than did the Parkfield aftershocks. Noting the dispersed Bor- rego Mountain aftershock pattern, they inferred that at Parkfield the fault surface is a well-developed zone of weakness, whereas at Borrego it is less well defined. Relatively few aftershocks were detected in the vicinity of the main shock. The surficial fracturing near the main-shock epicenter exhibited the greatest initial displacement but showed no later displace- ment. Beyond the northwestern extent of surface fault- ing, only a few aftershocks were detected. However, approximately a year after the Borrego Mountain earthquake, on April 28, 1969, a magnitude 5.8 earthquake occurred near the site of station 12 (fig. 12). Its aftershocks, monitored in a joint project by California Institute of Technology and the US. Geo- logical Survey (Thatcher and Hamilton, 1971), de- fine a zone that is essentially a northwestward extension of the Borrego Mountain aftershock zone. Thus, aftershock studies reveal that the initial bilateral faulting from the Borrego Mountain earth- quake extended in markedly different ways at its two ends. On the southeast end, fault creep and gradual growth are indicated, whereas on the northwest end, sudden extension more than a year later appears to have occurred. The curious distribution of the main aftershock band and the epicentral clusters northeast and south- west of it suggests that the zones may be related in a special way. One possibility is that the clusters are associated with a fault system that is conjugate with the northwest-trending system. However, the lack of continuity between the two clusters detracts from this suggestion, and such a system of active faults is unknown in the cluster areas or in any other nearby part of southern California. The shear-stress changes resulting from the main shock may have controlled the aftershock pattern. The stress changes 54 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 from a strike-slip dislocation in a halfspace were calculated by Chinnery (1963). His results show that the greatest increase in shear stress in the horizontal direction parallel to the dislocation plane occurs in a complicated pattern near the ends of the dislocation. However, the shear stress also increases in areas centered on the order of one dislocation length perpendicular to the center of the dislocation. Such a stress redistribution resulting from the Bor- rego Mountain faulting may account for the unusual epicenter pattern. An important result of this study is the correlation between the aftershock distribution and the surface faulting. In the area where the aftershock zone is narrowest, along the southeast fault strand near station 17 (fig. 12), the surface faulting runs through the epicenter region (fig. 16, M—N). Near station 9, where the fracture zone is widest, the aftershock zone is wide (fig. 16, I—J). Beyond the southeastern limit of faulting, the minimum depth of foci increases to below 2 km (fig. 16, O—P). The minimum depth of foci also increases at the north- western limit of faulting (fig. 15, A—B). The fault- plane solutions also correlate well with the aftershock distribution. Where the aftershock zone is narrowest, near station 17, the fault—plane solutions are all essentially identical, and one of the nodal planes passes through the aftershock zone. There is good evidence that aftershocks occurred in the sedimentary rocks above the granitic base— ment. Because of the uncertainties inherent in focal- depth determination, such a conclusion would generally be difficult to draw with confidence. For the Borrego Mountain aftershock sequence, though, the close-in distribution of seismographs and the explo- sion-refraction study (Hamilton, 1970) provided critical information. Along the southeast fault strand, foci were computed to lie within 1A2 km of the surface. The explosion data indicate basement depths of about 13/; km below station 9, 2% km below station 17, 214 km below station 4, and 31/4. km in the area several kilometers northeast of the surface faulting. REFERENCES CITED Aki, Keiiti, 1969, Analysis of the seismic coda of local earth- quakes as scattered waves: J our. Geophys. Research, v. 74, p. 615—631. Allen, C. R., Grantz, A., Brune, J. N., Clark, M. M., Sharp, R. V., Theodore, T. G., Wolfe, E. W., and Wyss, M., 1968, The Borrego Mountain, California, earthquake of 9 April 1968: a preliminary report: Seismol. Soc. America Bull., V. 58, p. 1183-1186. Arabasz, W. J., Brune, J. N., and Engen, G. R., 1970, Loca- cations of small earthquakes near the trifurcation of the San Jacinto fault southeast of Anza, California: Seismol. Soc. America Bull., v. 60, p. 617—627. Borcherdt, R. D., 1970, Effects of local geology on ground motion near San Franscisco Bay: Seismol. Soc. America Bull., v. 60, p. 29—61. Brune, J. N., and Allen, C. R., 1967, A microearthquake survey of the San Andreas fault system in southern California: Seismol. Soc. America Bull., v. 57, p. 277-296. Chinnery, M. A., 1963, The stress changes that accompany strike-slip faulting: Seismol. Soc. America Bull., v. 53, p. 921—932. Eaton, J. P., 1969, HYPOLAYR, a computer program for de- termining hypocenters of local earthquakes in an earth consisting of uniform flat layers over a half space: U.S. Geol. Survey open-file report, 106 p. Eaton, J. R, Lee, W. H. K., and Pakiser, L. C., 1970, Use of microearthquakes in the study of the mechanics of earth- quake generation along the San Andreas fault in central California: Tectonophysics, v. 9, p. 259—282. Eaton, J. P., O’Neill, M. E., and Murdock, J. N., 1970, Detailed study of the aftershocks of the 1966 Parkfield-Cholame, California, earthquake: Seismol. Soc. America Bull., v. 60, p. 1151—1197. Hamilton, R. M., 1970, Time-term analysis of explosion data from the vicinity of the Borrego Mountain, California, earthquake of 9 April 1968: Seismol. Soc. America Bull., v. 60, p. 367—381. Hanks, T. C., and Brune, J. N., 1970, Seismicity of the San Gorgonio Pass [abs.]: EOS (Trans. Am. Geophys. Union), v. 51, p. 352. McEvilly, T. V., 1966, The earthquake sequence of November 1964 near Corralitos, California: Seismol. Soc. America Bull., v. 56, p. 755—773. McEvilly, T. V., and Casaday, K. B., 1967, The earthquake sequence of September, 1965 near Antioch, California: Seismol. Soc. America Bull., v. 57, p. 113—124. Nordquist, J. M., Allen, C. R., Brune, J. N., Richter, C. F., and Taylor, V., 1972, Local bulletin of earthquakes in the southern California region, 1 January 1968 to 31 Decem- ber 1969: California Inst. Technology, Pasadena, 38 p. Smith, S. W., and Wyss, Max, 1968, Displacement on the San Andreas fault subsequent to the 1966 Parkfield earth- quake: Seismol. Soc. America Bull., v. 58, p. 1955-1973. Thatcher, W. R., and Hamilton, R. M., 1971, Spatial distribu— tion and source parameters of the Coyote Mountain after- shock sequence, San Jacinto fault zone [abs.]: Geol. Soc. America Abs. with Programs, v. 3, no. 2, p. 208. Udias, Augustin, 1965, A study of the aftershocks and focal mechanism of the Salinas-Watsonville earthquakes of August 31 and September 14, 1963: Seismol. Soc. America Bull., v. 55, p. 85—106. Unger, J. D., and Eaton, J. P., 1970, Aftershocks of the Octo~ ber 1969 Santa Rosa, California, earthquakes [abs] : Geol. Soc. America Abs. with Programs, v. 2, no. 2, p. 155. Wyss, Max, and Brune, J. N., 1971, Regional variations of source properties in southern California estimated from the ratio of short- to long-period amplitudes: Seismol. Soc. America Bull., v. 61, p. 1153—1168. SURFACE RUPTURE ALONG THE COYOTE CREEK FAULT By MALCOLM M. CLARK U.S. GEOLOGICAL SURVEY ABSTRACT The nearly linear 31-km-long surface rupture that accom- panied the Borrego Mountain earthquake of April 9, 1968, along the Coyote Creek fault comprised three well-defined bands of fractures—the north, central, and south breaks —each of which consisted of en echelon, complex, and simple right-lateral fractures that cut Pliocene-Pleistocene and Holocene sediments at the surface. Associated minor diverging and isolated fractures and zones of fractures de- veloped as far as 3 km from the well-defined breaks. A few of these displayed dominant vertical or left-lateral displace- ment. Maximum right-lateral displacement was 380 mm on the north break and, including postearthquake creep into 1971, 250—300 mm on the central break, and 80—140 mm on the south break. Subordinate vertical slip (230 mm maximum) accompanied right—lateral slip along nearly half of the north and central breaks and without exception was in the same sense as earlier uplift of scarps and hills. Some of the outlying fractures that displayed significant left—lateral displacement have orientations that are com- patible with some theories and observations of secondary fracturing associated with strike-slip faulting. However, others of these that also display large vertical displacement appeared to be normal faults trending obliquely to the local direction of tectonic extension. Surviving older scarps and vegetation boundaries extended along roughly one-third of the entire length of the 1968 rupture and were in all places coincident with new fractures. However, where vertical displacements and vegetation bound— aries were absent, other surface evidence that would have betrayed the position of the fault was scanty. Few ofi‘set drainage channels survived from previous displacements, be- cause erosion in the poorly consolidated sediments proceeds more rapidly than tectonic displacement. Along about 50 percent of the length of the rupture, the position of the main fractures could have been predicted before the earthquake to within :100 m or closer. The width of the band of fracturing along the three main breaks exceeded 50 m along about 35 percent of the length of the 1968 rupture and exceeded 100 in along 25 percent of the length of the rupture. However, most of the ground dis- placement along the main breaks occurred within a band less than 20 m wide. Associated fractures with potentially damaging displacements extended more than 100 m from the main breaks in some locations and as much as 500 m from the main breaks in a few locations. Because not all these fractures followed obvious lines of earlier faulting, sub- surface investigations should be mandatory around the site of any proposed large or multistory structure within about 1 km of the 1968 rupture. Fault creep and rain-induced collapse maintained or en- larged some fractures as late as 1971, more than 31/2 years after the earthquake. No significant postearthquake creep took place on the north break, but simple measurements made at irregular intervals since April 1968 reveal that postearth- quake movement along the central and south breaks has in- creased original displacements by as much as 100 percent. Some of this creep was apparently continuous, beginning immediately after the earthquake, but several parts of the central break did not start moving until at least several months after the earthquake, and the south break may not have started moving until about 1 year after the earthquake. Reconstruction of the creep history since the earthquake at one place on the central break employed repeated measure- ments of displaced tire tracks and new ofl’sets in crusts formed in material filling 1968 fractures. The reconstruction is compatible with the exponential decay in creep rate re— ported for other earthquakes, except for an apparent delay of several months after the earthquake before the onset of creep. The net result of creep on the central break has been to increase initial displacements to amounts comparable to those on the north break at the time of the earthquake. The south break not only experienced smaller displacements and a greater delay (perhaps more than 1 year) before the onset of creep than did the central break, but it also includes four sets of tectonic fractures that predate the 1968 earth- quake by a few months to more than 15 years. This activity, combined with postearthquake creep and a recent history of earthquakes that includes one of magnitude 6.5, suggests that the south break may behave more like the Superstition Hills and Imperial faults than adjacent parts of the Coyote Creek fault. INTRODUCTION This paper describes the surface rupture and dis- placements that formed along the Coyote Creek fault during and after the Borrego Mountain earthquake of April 9, 1968. The fault is part of the San Jacinto fault zone, a major branch of the San Andreas sys~ tem of right-lateral strike-slip faults. (See Sharp, “Tectonic Setting of the Salton Trough,” this vol- ume.) Conditions along the 1968 rupture have re- vealed some characteristics of surface faulting that are not obvious at many other places along the San 55 56 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 FIGURE 23.—Vertical aerial photograph of the region of surface faulting associated with the Borrego Mountain earthquake. The Lake; CM, Coyote Mountain; NB, north break; OB, Ocotillo Badlands; OW, Ocotillo Wells; SB, south break. Other features Andreas fault system. The combination of scant vegetation, low rainfall, low relief, and predomi- nantly fine—grained sediment along the 1968 rupture has resulted in preservation of many details of the surface fractures and allowed continuing study of changes in the rupture more than 3 years after the earthquake. Many aspects of the surface faulting are similar to patterns produced by earlier earthquakes along the San Andreas system. Most of the rupture consisted of right-lateral en echelon fractures that coincided with locally surviving traces of earlier faulting. The investigation also documented the asso- ciation of significant and consistent vertical and minor left-lateral components of displacement on fractures associated with the main breaks, charac- teristics not generally recorded or observed after earlier earthquakes in the San Andreas system. Fur- thermore, the investigation has pieced together a complex ‘history of postearthquake creep that con- trasts with the simpler relations found between initial displacements, creep, and aftershocks of the San Andreas fault at Parkfield-Cholame after the earthquake of 1966. Field investigation of the surface rupture started about 8 hours after the earthquake and is still being conducted at this writing (November 1971). The initial effort to record fractures began at 2 a.m. (P.s.t.) April 9, 1968, when G. R. Allen of California Institute of Technology measured the rupture where it crossed Highway 78. By April 14, most of the fractures had been mapped. Allen was joined in mapping the rupture by several US. Geological Sur- THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 57 surface ruptures are shown as rows of open circles. BM, Borrego Mountain; BS, Borrego Springs; CB, central break; CL, Clark can be identified from plate 1. USGS aerial photographs USAF 665V 037 and 039, taken by U.S. Air Force, November 1967. vey geologists (Arthur Grantz, T. G. Theodore, R. V. Sharp, E. W. Wolfe, and myself). Later in April and in May, Survey geologists E. W. Wolfe, R. 0. Castle, and T. L. Youd recorded additional fractures at some distance away from the main breaks. In June 1968, I noticed the first obvious evidence of continued movement along the rupture and mapped additional fractures that were either overlooked or not present in April. Thereafter, I was accompanied at times by A. Grantz, R. V. Sharp, M. Rubin, M. G. Bonilla, and J. Kahle in recording increases in dis- placement, new fractures, and other changes in the surface rupture during visits in October 1968, March, April, and October-November 1969, January and December 1970, and March and November 1971. Allen also monitored creep during other visits. Many people in addition to the aforementioned have contributed to this investigation. J. P. Lock- wood helped identify and record ground ruptures during the first few days of field study. Light Photo- graphic Squadron 63 of the U.S. Navy at Miramar, Calif, made excellent large-scale aerial photographs of the surface rupture in 1968 and 1969. I especially thank Commanders Hegrat and Dunkin for their cooperation and expertise in getting these photo- graphs. Mr. and Mrs. Earl Cartier and Mr. and Mrs. John Balch of Ocotillo Wells provided important information and assistance during our fieldwork. Bob Densmore of Borrego Springs supplied logistic support and took photographs of creep localities in March 1970. H. B. Goldis helped compile many of the illustrations of this report. Arthur Grantz, R. V. 58 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 Sharp, M. G. Bonilla, and R. F. Yerkes offered help- ful suggestions regarding the text. Field parties used a variety of base materials for mapping fractures. They recorded most of those found in April and May 1968 on aerial photographs taken by the US. Department of Agriculture in 1953 at 1:20,000 scale; they plotted the remaining frac- tures found during this period on 1:25,000-scale US. Geological Survey photographs or 1:24,000- scale topographic maps. Subsequent investigations used aerial photographs at 1 : 10,000 and larger scales taken in May 1968 and April 1969 by US. Navy aircraft from Miramar, Calif., and 1:8,000- scale US. Geological Survey color aerial photographs taken in April 1969. DESCRIPTION OF THE SURFACE BREAKS The principal surface rupture defines a narrow roughly linear band about 31 km long (pl. 1; fig. 23). The band consists mainly of right-lateral en echelon fractures and trends roughly N. 40° W. across lower Borrego Valley and along the northeastern base of Borrego Mountain. Minor isolated and diverging fractures formed as much as 3 km away from the main zone of breakage. The surface fractures formed entirely in Holocene alluvium and lake beds, except in the vicinity of Ocotillo Badlands and at the ex- treme north end of the main rupture. In both these places, fractures traversed uplifted and folded Plio- cene-Pleistocene sediments. Two interruptions in the continuity of the rupture divide it into three main breaks (fig. 23), the north (locs. 1—14, pl. 11), central (locs. 14—26), and south (locs. 26—33). The north and central breaks are en echelon and are separated by the uplifted and tightly folded sediments of Ocotillo Badlands. The central and south breaks are connected by a narrow zone of Widely separated small fractures. Displacements on each of these three breaks reached a maximum near the center and diminished toward the ends. Maxi- mum right-lateral displacement on the entire rupture was 380 mm (15 in.) on the north break near the base of Borrego Mountain. Fractures of the 1968 rupture can be divided into three categories: (1) nearly all fractures of the three main breaks, (2) all diverging and isolated tectonic fractures that lay more than about 100 m from the centers of the three main breaks, and (3) nontectonic fractures produced by differential com- paction of sediment and by local slumping and sliding on steep slopes bordering washes or in hills. Although we intended to map only tectonic fractures, 1Throughout this and several subsequent chapters, locations along the fault zone will be given by the horizontal coordinate system (kilometers) of pl. 1. a few of the fractures on plate 1 (for example, those 1 km east of the north break at 10c. 3 and within Ocotillo Badlands at Ice. 15) may actually be in the nontectonic category. Fractures of the three principal breaks (the first category) shared characteristics that differed from most of the fractures of the second category. All fractures of the principal breaks showed right-lateral displacement, except for minor extension and com- pression fractures oriented nearly perpendicular to the trend of the zones. Many fractures of the prin- cipal breaks also exhibited vertical displacement that was nearly always smaller than the associated hori- zontal movement. The principal breaks trend more or less uniformly N. 30°-50° W. and displayed an essential continuity of fracturing within each break; they also contain prominent scarps from previous faulting, indicating continuity of position with re- spect to past tectonic events. In contrast, outlying and diverging fractures (the second category) as a group showed different char- acteristics. Although some showed right-lateral off- set, others showed vertical offset only (loc. 3.4) or vertical offset greater than horizontal (10c. 3.6), and on a few, displacements were left lateral (locs. 3.6, 12.4, 14.0, 15.2, 15.8, 17.4). The outlying and diverg- ing fractures also displayed a greater range of orien- tations and, obviously, a lesser degree of continuity. Maximum displacements in this group of fractures were far less than the maximum displacements along the main breaks. Many of the outlying and diverging fractures had large vertical components of displace- ment but did not follow older scarps, suggesting randomness of position with respect to earlier frac- turing. Despite these differences, most of the diverging and outlying fractures appeared to be tectonic, rather than the secondary effects of shaking. Most either connected with fracures of the main breaks or bordered low uplifted hills and ridges underlain by tectonically deformed Pleistocene and Holocene strata, such as those southwest of the principal rupture zone at locations 3 and 15 to 16, or northeast of the zone at locations 13 to 14, 22, and 27. Some of the isolated and diverging fractures of the 1968 rupture probably escaped detection. Many shown on plate 1 had small displacements (less than 20 mm) and were quite inconspicuous. These are not visible on our largest scale vertical aerial photo- graphs and were found only during careful traverses on foot. In addition, the wide bands flanking the 1968 rupture were not inspected as thoroughly as the immediate areas of the main breaks and the larger diverging fractures. THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9. 1968 59 Nontectonic fractures (the third category) were common. The 140-m-high hills of Ocotillo Badlands, made up of unconsolidated and poorly consolidated muds, sands, and gravels, were shattered by hun- dreds of slide and slump fractures from a few meters to hundreds of meters long. Small subsidence fractures also surrounded the many irrigated alfalfa fields south and southeast of Ocotillo Wells. PATTERNS OF GROUND BREAKAGE Displacement along the three principal breaks was commonly distributed along a zone of fractures, en echelon, parallel, or complex, rather than along a single fracture. Most of the north break and much of the central and south breaks consisted of en eche- lon fractures from a few meters to tens or hundreds of meters long that occupied a band ranging in width from less than 1 m to more than 100 m. A small part of the three main breaks consisted of bands of com— plex fractures as much as 10 m wide or two or more subparallel fractures in a zone as much as 100 m wide. Other parts of the central and south breaks consisted of single fractures as much as hundreds of meters long. As noted by Allen, Grantz, Brune, Clark, Sharp, Theodore, Wolfe, and Wyss (1968) , even in the wide zones of complex fractures most of the displacement took place within a narrower zone that was generally less than 20 m across (figs. 24—28). This band of maximum displacement did not occupy a consistent position within the rupture zones and would have been the location of greatest damage had manmade structures been present along the rupture. The en echelon pattern of fracturing persisted on various scales in remarkable similarity to patterns of surface faulting reported from such places as Parkfield-Cholame on the San Andreas fault (Brown and others, 1967) and Iran (Tchalenko and Abraseys, 1970). En echelon separation ranged in size from 2 km between the north and central breaks to a few millimeters between en echelon hairline cracks developed locally on smooth parts of the surface. The geometry of en echelon pattern requires com- pression and distortion in the area between frac- tures, which is shown clearly by the folded sediments of Ocotillo Badlands, lying between the north and central breaks. (See Sharp and Clark, this volume.) However, compression was also displayed over much smaller distances by regular patterns of thrusting between en echelon fractures developed in surface flakes and slabs of the muddy silt playa of Benson Lake. Here fractures about 3 m apart dis- played a notable regularity of spacing and behavior; each was connected to the next by paired S-shaped .0 ,_ I l I l l ‘l ‘ FIGURE 24.—Central break at location 19.7 showing en echelon fractures as much as 20 m long and 2 m apart within a zone about 2 m wide (sets of vehicle tracks are about 1.6 m wide). The fractures follow an old scarp that is 0.5—1 m high (southwest side up). Displacements on this segment of the break at the time of the earthquake reached 160 mm right lateral and 100 mm vertical. Photo- graph by B. W. Troxel, California Div. Mines and Geology, April 19, 1968. overthrusts (figs. 28, 29). Thrusting here appeared to be limited to a relatively brittle layer about 0.1— 0.2 m thick at the surface. Presumably, compression below this layer caused distortion or flowage of the material between the en echelon fractures. In general, individual en echelon fractures trended 5°—40° clockwise from the strike of the breaks (pl. 1; fig. 24), with 30° the most common difference. Thus, along the predominant N. 40° W. strike of the north and central breaks, individual fractures , trended nearly north-south. However, in the vicinity of Fish Creek Wash (10c. 25), where the central break runs nearly north-south, individual fractures strike N. 5°—30°E. Although the position and relation of the three main breaks probably reflect the configuration of the fault in bedrock, the factors that control the detailed appearance of the surface rupture itself remain un- known. Magnitude of displacement does not obvi- ously control the character of surface fracturing, inasmuch as both small and large displacements occurred across single, multiple, complex, and en echelon fractures that occupied either narrow or wide bands. Depth to bedrock may slightly influence 60 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 FIGURE 25.—Band of complex fractures about 1 m wide across hardened silty sand crust in a wash at the eastern base of Borrego Mountain (near loc. 6). Right-lateral dis- placement across the rupture reached 300 mm in the fore- ground; right (southwest) side was raised about 100 mm. A subsidiary fracture is in the right center. Photo- graph by C. R. Allen, April 13, 1968. the surface fractures, because most long single frac- tures were limited to the central break, which lies entirely in a 2—4-kni thickness of late Cenozoic sedi- ment (Hamilton, 1970; Sharp, “Tectonic Setting of the Salton Trough,” this volume) , whereas the north break, which essentially lacked long single fractures, presumably crosses very shallow sediment alongside Borrego Mountain. In contrast, however, en echelon and complex breaks similar to those on the north break next to Borrego Mountain were abundant in the thick sediment of the central break. It seems more reasonable to suppose that patterns of surface fracturing tens to hundreds of meters wide are con- trolled by physical properties within a surface layer of comparable depth, rather than to suppose that such patterns are transmitted unchanged upward from bedrock through sediments tens to hundreds of times thicker than the dimensions of these pat- terns. An explanation for the varied appearance of the surface trace probably requires further study of subsurface material of the fault zone. DISPLACEMENTS Nearly all the displacements recorded on plate 1 were measured directly from irregularities on or in FIGURE 26.—En echelon, parallel, and complex fractures in a zone 6—10 m wide on the north break at location 6.4. Cumulative horizontal displacement here was 380 mm, the maximum measured along the 1968 rupture. Photograph by C. R. Allen, April 13, 1968. the ground itself, rather than from manmade objects such as fences, pipelines, or buildings, few of which are present in the fault zone. Vertical measurements were simple to make because most of the rupture crossed nearly flat terrain, but horizontal displace- ments were more difl‘icult to determine. We were able to measure a few such displacements in places where matching irregularities across fractures sur- vived; small channels across fractures in some other places recorded horizontal offsets. However, the best and most abundant indicators of horizontal displacement were tire tracks. Hundreds of motor- cycles, dune buggies, and four-wheel-drive vehicles frequent this region every weekend and holiday period during the cooler seasons. Tracks laced the fault at the time of the earthquake and became one of our most common indicators of lateral displace- ment (figs. 30, 31). Unfortunately, drivers continued to drive on and across the break after the earth— quake, often obliterating evidence of displacement. FRACTURES WITH RIGHT-LATERAL COMPONENTS OF DISPLACEMENT Horizontal displacement along any individual frac- ture that does not connect with other fractures increases from zero at each end to a maximum value somewhere in between, usually near the center. This distribution of displacement holds for each of the three major breaks as well as for diverging or out- THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 61 D FIGURE 27. —Time sequence on central break at location 22.5. Here the central break was a single fracture with very few en echelon fractures. For 100 m along the single fracture at this location, the vertical displacement in 1968 and in the past raised the northeast side of the fault as shown here, in contrast to adjacent sections of the break, along which the southwest side has moved up. (See fig. 24.) A large amount of postearthquake creep occurred on this section of the break. A, April 12, 1968. Horizontal displacement about 180 mm, vertical, 100—150 mm. B, June 20, 1968. Hori- zontal displacement has increased to 250—300 mm and vertical displacement to 150 mm at the shovel handle and 230 mm between the shovel and jeep. No further displacement occurred after this date. C, March 1969. Drifting sand fills the fracture and is banked against the scarp. D, October 1969. Scarp is gradually being destroyed by erosion and deposition. lying zones and individual en echelon fractures. (See fig. 32.) Maximum displacement on the north break, 380 mm, occurred in the central third of that zone at the base of Borrego Mountain. Similarly, the maxi- mum displacements measured on the central and south breaks, 230—300 mm and 50—80 mm, respec- tively, occurred in the central part of each. Some of the pronounced variability in displace- ment shown in figure 32 along each break probably results from measurement errors. Generally, the larger and smaller measurements of displacement on each break are fairly accurate. The larger displace- ments received special attention because we were interested in the maximum displacements caused by the earthquake. Smaller displacements (less than 30——50 mm) nearly always occurred on clean single fractures and were easy to measure accurately. How- ever, some intermediate values of displacement may be too low, because total cumulative slip was not recognized and measured, particularly across Wide zones of multiple, complex, or en echelon fractures. The central and south breaks appeared to have a continuous yet poorly developed connection, shown on plate 1 between locations 25 and 28 as a series of widely spaced fractures with small displacements. There were probably more fractures in this zone immediately after the earthquake than the map shows. Unfortunately, strong winds drifted sand along this and other parts of the central and south breaks before mapping was completed. The sand almost certainly filled some unmapped fractures at the junction of the central and south breaks because 62 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 FIGURE 28.—En echelon fractures at the north end of Ben- son Lake (loc. 10.2). Fractures are connected by S-shaped overthrusts shown in detail in figure 29. Photograph by Arthur Grantz, April 10, 1968. it filled some fractures in nearby areas that had already been mapped. Thus, we probably missed many of the small fractures in this region. However the small displacements that characterize this sec- tion also indicate that the connection at the surface between the central and south break was poorly developed, despite the fact that sand probably buried much of the evidence. Mapping to the north and south before the wind filled some of the fractures and later to the south in areas not greatly affected by blown sand indicated that displacements diminished to very small values (about 10 mm) toward the junction of the central and south breaks. Horizontal slip along nearly all right-lateral frac- tures of the three main breaks appeared to be parallel to the N. 30°~40° W. trend of the entire rupture, regardless of the orientation of individual fractures. Thus, fractures trending N. 30°—40° W. showed only strike slip, but those with increasingly divergent trends showed an increasing component of opening normal to the trend of the fracture. Fractures ori- ented more than 45° from the general trend of the rupture showed larger components of opening than components of strike slip. Inasmuch as most parts of the three main breaks nearly parallel the N. 40° W. trend of the 1968 rupture, most individual right- lateral fractures trend about 30° clockwise from this direction and show roughly twice as much strike slip as opening slip. However, near location 24, where the central break runs roughly north-south, individual fractures diverge about 60° from the general N. 40° W. slip direction and show twice as much opening slip as strike slip. 0-5 2 1 FIGURE 29.—Detail of symmetrical S-shaped overthrusts in the clay-silt crust of Benson Lake. These thrusts connect en echelon fractures of the main rupture shown in figure 28. Sketch by R. V. Sharp. FRACTURES WITH VERTICAL COMPONENTS OF DISPLACEMENT Nearly all significant vertical displacement across tectonic fractures occurred at locations showing clear evidence of vertical displacement in the past. Evidence of such displacement was found along scarps, at the base of hills or mountains, or along- side uplifted beveled strata. Moreover, vertical dis- placement in 1968 everywhere continued the sense of past vertical movements. For example, vertical displacement on the north break occurred only along the base of Borrego Mountain (mountain side up) and along Ocotillo Badlands (badlands side up). On the flat terrain north of Borrego Mountain and at Benson Lake, the north break showed no vertical THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 FIGURE 30.-—Tire tracks at location 22.5 displaced about 180 mm across the single fracture at this location. Photo- graph by C. R. Allen, April 10, 1968. displacement. The central break displayed vertical offset of 20—100 mm (southwest side up) nearly continuously for 8 km from its north end south- eastward along the base of Ocotillo Badlands and beyond (description of an exception follows) along a 1/2—1-m-high east-facing scarp as far as location 23, where the break diverged from the scarp. Far- ther southeast, the break crossed flat terrain and showed no vertical offset. The south break followed no significant scarps and showed very little vertical displacement. A localized zone of anomalous vertical displace- ment occurred along the scarp of the central break at location 22.5. Along this prominent east-facing scarp, vertical displacement in 1968 reversed along a segment about 100 m long and created a west- facing scarp that was 100—150 mm high on April 8 but rose to a height of 230 mm by June 1968 (fig. 27). N 0 old west-facing scarp existed here, but color aerial photographs taken in April 1969 and ground examination revealed uplifted and beveled strata on the east side of the break exactly at this location. Evidently, there has been reversed uplift on this short segment many times in the past. New fractures were associated with upwarped sediments in two other places, both east of the main rupture. Near location 22, 1 km east of the central break, a prominent secondary break traversed the steeper, north slope of low hills formed by uplifted unconsolidated sediments in an area of otherwise 63 FIGURE 31.—Dune buggy crossing the north break in Ben- son Lake 6 days after the earthquake. Desert vehicles supplied many tire tracks for measurements of horizontal displacement; however, they also caused a small but sig- nificant part of the destruction of the surface rupture. undisturbed surficial strata. Gentle dips evident on some of the strata reveal a pattern of upwarping and beveling very similar to that of the reversely uplifted section of the central break just described. Near location 27.4, 5 km farther to the southeast and some 3 km distant from the south break, the sand-filled remnants of fresh cracks traverse the scarp on the southwest side of a low hill. Again, gently dipping beds are present at this scarp. The fractures most likely formed during or after the April earthquake, and as at the two locations just mentioned and at Ocotillo Badlands, the vertical displacements of the earthquake were continuing to uplift a hill here that evidently owed its existence to episodic tectonism. (See Sharp and Clark, this vol- ume.) FRACTURES WITH LEFT-LATERAL COMPONENTS OF DISPLACEMENT A fascinating aspect of the outlying fractures was that some of them displayed left-lateral offset. We found this apparently anomalous displacement at four separate locations: (1) a series of fractures lying as much as 1 km west of Ocotillo Badlands (locs. 12—12.4; 15—16), (2) several fractures north- east of Highway 78 and 1—2 km from Ocotillo Bad- lands (locs. 13—14), (3) two small zones at the northeast base of Ocotillo Badlands (locs. 15.1, 17.5) , and (4) a series of fractures near the north end of the north break at the base of Borrego Mountain (locs. 3—5). Unfortunately, the first two groups of left-lateral fractures, recorded by R. V. Sharp, R. 0. Castle, and E. W. Wolfe, did not survive for further study. However, the left-lateral fractures at the other two locations did survive (those at the north- east base of Ocotillo Badlands were not discovered 64 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 EXPLANATION 0 APRIL I968 A JUNE I968 E400? 0 MARCH I969 E V OCTOBER I969 (EstimoIed total ‘0 displacement) L” 350- K g ———-EsIImated displacement )— 2? at 300- cf LL 8 o 250- o: o <: r— 200- z LLI E Lu I50- 0 <1 _1 o. (2 I00- o M A 50. t1 O 02/ 3 (5 s E) {a v.4, ls l8 £0 éa 24 26 as K———- NORTH BRI~:/.\K————*‘>T~L CENTRAL. BREAK——>I<— HORIZONTAL DISTANCE (KM) KEYED TO GRID ON PLATE I FIGURE 32. — Total right-lateral displacement along the north, central, and south breaks measured at different times (data from pl. 1). The fact that June measurements near loca— tions 18-20 are smaller than measurements made the pre- ceding April has no significance. The same features were not measured each time, and the differences are within the until June 1968), and much of the discussion of left-lateral displacement is based on study of these fractures. Three generalizations can be made about the frac- tures showing left-lateral offset. First, the three groups in the vicinity of Ocotillo Badlands trended northeast, rather than northwest or north as did the right-lateral fractures in the same area. Second, at least some of the fractures in each group displayed a left-lateral en echelon arrangement. Third, all fractures except those northeast of Highway 78 and those southwest of the highway at locations 12—124 displayed about as much vertical as left-lateral dis- placement. Indeed, most of the left-lateral fractures at the base of Borrego Mountain had more vertical than horizontal displacement. In contrast, almost nowhere along the right-lateral fractures did vertical displacement exceed horizontal. Those left-lateral fractures that trend northeast are compatible with patterns of faulting reported by Dibblee (1954) and Tchalenko (1970) and with accuracy of measurements of this amount of displacement (roughly :10 mm). However, comparable differences be- tween measurements made in April and June between loca- tions 23 and 26 are significant because the measurement error is smaller (<5 mm) for small displacements. a theoretical analysis by Chinnery (1966). Dibblee ascribed the prevalence of minor left~lateral faults in this region to clockwise rotation of northeast- trending blocks, caused by drag along northwest- trending right-lateral faults. The northeast-trending left-lateral faults also fit the category of “conjugate Riedel shear” fractures as summarized from labora- tory and field observations by Tchalenko (1970), although the northeast-trending faults along the 1968 rupture are not so widespread as those de- scribed by Tchalenko. Chinnery’s (1966) model of the stress system at the end of a strike—slip fault also predicts left-lateral fractures. Left-lateral fractures in the vicinity of 0c0tillo Badlands (those of the first three locations de- scribed) fit Chinnery’s model in that they lay near the ends of two strike-slip faults (the north and central breaks), and their directions (roughly north- east to southwest) generally fit the predicted trends of secondary fractures showing displacement oppo— site in sense to that of a primary break. THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 65 ' \ “z 'N ‘0 ‘t— l“ .W. e. \i / , 44/ FIGURE 33.—Segment of an extensional fracture showing dip slip at location 3.3, 1 km southwest of the north break. Compass is parallel to the direction of extension, N. 20° E. Vertical displacement is roughly 50 mm, and opening is about 20 mm. Pencils and thermometer case point to matching irregularities. Those to the left of the compass, on a segment of the crack trending N. 10° W., show left- lateral displacement, whereas those to the right, on a segment trending N. 75" E., show right—lateral offset when viewed normal to each segment. An alternative explanation for left-lateral dis- placement is suggested by the evidence at the north end of the north break, where vertical displacement exceeded left-lateral displacement on most breaks that survived for further study two months after the earthquake. Here fractures trended north or north- west, so that explanations based on the northeast strike of left-lateral fractures do not apply. These fractures separated a steep mountain front from an alluvial basin, and at least some of them resembled normal faults because they were not straight and they displayed a significant amount of displacement normal to the trend of the fractures, in addition to vertical and left-lateral displacement. If normal faults are assumed to accommodate ex- tensional strain, then a normal fault that is not perpendicular to the direction of extension will ap- pear to have a strike-slip component in addition to the dominant vertical displacement. Figure 33 ShOWS part of a normal fault trace in the zone of left-lateral fractures near the base of Borrego Mountain. This fracture trended generally perpendicular to the ex- tension direction measured across it, about N. 20° E. In detail, however, the fracture followed a zig-zag course and showed left-lateral displacement in some places and right-lateral in others, depending on the trend of the fracture at each place. Other nearby fractures showed the same phenomena. Unfortu- nately, all the fractures mapped in April 1968 show- ing left— and right-lateral displacements in this . )- iif‘ '4 we...” .. FIGURE 34. — Fractures showing left-lateral displacement at location 17.5, northeast of the central break. The en echelon arrangement of these fractures is opposite that of right-lateral fractures of the main breaks; that is, in- dividual fractures here are oriented about 30° counter- clockwise from the trend of this fracture zone. Vertical and left-lateral displacements are about equal (40—50 mm). location could not be found later to verify whether local changes in trend had caused the observed mix of readings (pl. 1). However, since vertical displace- ments were consistently larger than horizontal ones, it seems at least possible, if not likely, that local changes did occur. Thus, displacement on outlying fractures at the north end of the 1968 break may result primarily from vertical adjustments near the boundary between a young range of crystalline rocks (Borrego Mountain) and an adjacent deep alluvial basin rather than from true left-lateral faulting. If this explanation is correct, the presence of minor left- and right-lateral displacements on these essen- tially dip-slip faults need not have fundamental tectonic significance. 66 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 Some of the left-lateral fractures adjacent to the uplifted Ocotillo Badlands may also be due partly to the mechanism just described. These frac- tures lay at the northeast base of the badlands (10c. 17.5) and displayed roughly equal amounts of left- lateral and vertical displacement (about 20—50 mm). In places, they followed older scarps that were as much as 150 mm high. However, some of the frac- tures displayed the en echelon pattern of left-lateral faults (fig. 34), suggesting that they did not result from normal faulting. Location of these fractures in an area of local tectonic uplift, plus the presence of significant vertical displacement on them, favors the normal-fault explanation, whereas the northeast strike and en echelon pattern of the fractures imply that horiontal displacements were most important. The position of the remaining left-lateral frac- tures as much as 2 km away from Ocotillo Badlands, the northeast strike of these fractures, and the lack of older scarps adjacent to them suggest at least that they were created directly by horizontal displace- ments rather than as byproducts of normal faulting. ALTERATION OF THE 1968 RUPTURE WITH TIME Periodic reobservation of parts of the fault has revealed extensive modifications of the 1968 rupture. In general, blown silt and sand have filled fractures and smoothed out abrupt scarps (fig. 27). Fractures that were not filled or sealed by sand and silt were enlarged by the action of storm runoff, which caused walls to slump and collapse and carried the resulting debris deep into the fractures (fig. 42; Clark, “Collapse Fissures Along the Coyote Creek Fault,” this volume). Most of the places where 1968 frac- tures still survived 1 year later were the same places that displayed obvious scarps or lines of vegetation before the earthquake. By April 1969, most of the north break north of Highway 78, a region lacking older scarps, was obliterated by blown silt and sand or by the action of San Felipe Creek. The same was true of the central and south breaks between the southeast end of the low scarp (loc. 22.6) and col- lapsed fractures at location 29.8. In contrast, as late as November 1971, many of the 1968 fractures and others subsequently created or enlarged by creep were still present and visible on the central break along the low scarp and Ocotillo Badlands. Moreover, many of the large collapse fissures of both the central and south break, some of which were still forming late in 1970, remained open in March 1971, and some were still open in November 1971. (See Clark, “Collapse Fissures Along the Coyote Creek Fault,” this volume.) POSITION OF THE RUPTURE RELATIVE TO OLDER TRACES The position of the 1968 and later fractures with respect to older fractures is important because such information helps us to predict patterns of future ground breakage along active faults, which tend to break repeatedly along the same trace. (See Ross, 1969; Clark and others, this volume.) Accordingly, a map (fig. 35) was prepared showing preearth- quake evidence at the surface indicating the presence of earlier fault traces along the principal rupture of 1968. This map, based partly on interpretations of aerial photographs, is similar to a series of strip maps produced by the Geological Survey showing recently active ruptures along the San Andreas and other active faults of California. (See Ross, 1969; Vedder and Wallace, 1970; Sharp, 1970; Clark, 1971; Brown and Wolfe, 1970.) Sharp and Clark (this volume) discussed the implications of the structural evidence for past faulting on and near the 1968 rupture. Several different kinds of field evidence reveal the presence of traces of active faults. Physiographic evidence, such as scarps, offset channels, linear valleys, and other features caused by faulting, is the most common kind. These features are short lived and indicate fault movement no older than the life- time of the landscape. Other evidence includes vegetation boundaries (caused by the influence of faults on ground water or soil) and the lithologic and structural criteria that ordinarily identify faults of any age. This nonphysiographic evidence is not restricted to active faults, because it merely indicates the presence of faults, whose most recent movements may be nearly as old as rocks they displace. How— ever, most of the rupture of the Borrego Mountain earthquake is in Holocene deposits, and any fault that disrupts deposits this young must be considered active or potentially so. All these types of evidence of active faulting are shown in figure 35. Roughly one-third of the 1968 rupture coincided with vegetation boundaries or the bases of scarps made by previous displacements FIGURE 35,—Map showing the location of recently active traces of the Coyote Creek fault (heavy dashed line) in the vicinity of the surface rupture of 1968 inferred from visi- ble evidence that existed just before the earthquake of April 9, 1968, and from knowledge gained during the post- earthquake investigation concerning the association of this evidence with surface fractures. Prepared with the help of preearthquake aerial photographs. Collapse fissures lying 2—3 km southwest of main trace may not be tectonic. (See Clark, “Collapse Fissures Along the Coyote Creek Fault,” this volume.) Light lines show 1968 surface ruptures. THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 / f éé ~50 \\ \ 4‘” ~ 6 \ x s \i k \ \ t \ , , B Raised elon ate nd #3 \ \ Q \..3 flue“ gas MM\ ‘ Southwestern bounda _ ‘2 ‘ _ EIsolated elongate R35“ “”3“" Notc 0' clumps 0' V980“ :‘ e e \\ fiEroded SWJacmg scarp "mm“. dunes gravel hill ‘b \ ‘1 \ \ 0° \ \ , be: 6 \ \ \ 06 \ s: tacin ’ ‘ / - a scarp. Alined \g \ 50—150 mm high Q bushes Guinea \ & "r {k}? mesquite dunes Intermittent NE-laci ‘ ’1 {p Fault trace dained by l'tholog’ scarp m 501 ontra t \ Raised hills expose I IS '- | C S contrast or by vertical dips or m \\ '°' ' “1 sediments '5” mungu'agmunit "r Searps and beds adiacent t- thetrace rdy n m sou e aoeted spu - ‘ .—“/ Elongat.‘ " ‘ u ‘ ‘ ’__,,'~ bus %\ v 45---?“ e w (5 - loontrast ‘\ on NE-leung scar rominent E-tacing V: m high scerp Vz-le m high 5 that lorms a sharp gt: boundary between x.\ ‘55 :Z‘Azga‘g‘ 31:33:? Irregular eastern boundary \ and earl Mn." 0' zone of abundant n large mesquite . flats to east ‘9 dunes It 0 Collapse tissures 4 Q e a” evident on 1953 4:, 9t \ / photographs 4‘ 5‘ \ \ ‘ . . V J" .0” 4P 4’ ,0 *~ X a“ II / \ ' g V ~ i / (Raised hill ,9 x“ I O Raised hill* ,1 ‘ ’l -tacin scarp, ‘/2-3 m high, I that am; the southern Isolated cluster. ’ boundary to abundant [at larger mesquite mesquite dunes dunes ’\ \ «9 «a3? «\ Northeastern boundary ol mesquite dunes 09"? Alined bushes and 00"39“ 7'35“" sinks along tilled colla selissure {a d nes Alined bushes and u \\ sinks along filled _ ‘ collapse tis e ”,—’ ’ A' ed mesquite dunes ~‘ \ ’____ a lorrn s'outhmtnn boundary ~ ‘\\ Northeastern boundary 0 group 0 “M \ ol mesquite dunes ‘ ~~ Fault exposed 4 in wash \ Mined pushes and Q Approximate position of mesquite dunes ~> northeastern boundary ‘ ol mesquite dunes q, g \e Collapse lissures evident on \ 1953 aerial photograph , 3‘ X43 s“ 3,5» - if 7 .0” xi” ‘5‘: a" \ 68 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 FIGURE 36. — View southeast along the central break showing an old scarp 0.3—0.4 m high in alluvium at the northeast base of Ocotillo Badlands. A single fracture of the 1968 rupture forms the base of this scarp but is not visible in this picture. (figs. 36, 37). The scarps are best developed along the northwest and southeast flanks of Ocotillo Bad- lands and, as noted before, extend for more than 3 km beyond the southeast tip of Ocotillo Badlands. Less prominent scarps, also followed by isolated fractures, exist along and within some of the up- lifted hills east of the central break. There is no scarp along the base of Borrego Mountain, even though relatively large vertical dis— placements took place there in 1968 and Borrego Mountain itself may be a result of continuing episodic uplift along the Coyote Creek fault. The reason for this lack of a scarp almost certainly is that water in San Felipe Creek and other washes flows directly on the trace, leveling vertical displace- ments soon after they take place. By April 1969, flow in San Felipe Creek had effectively erased the scarp created in its channels in 1968. A prominent scarp has not formed immediately south of where San Felipe Creek leaves the trace and the mountain front (loc. 6.6) apparently because of a combination of circumstances: (1) Minor channels flow parallel to and on the break, (2) vertical displacement was distributed across a zone more than 15 m wide, and (3) between April and June 1968, drifting sand buried most of the trace and completely filled a nearby l-m-deep stream channel. Parts of the fault zone that showed no physio- graphic evidence of previous vertical displacement also lacked other significant visible geologic evidence of faulting. This observation emphasizes that in an area of active deposition on both sides of an active fault, sedimentation acts to conceal lithologic and structural changes caused by horizontal displace- ment, and horizontal faulting without vertical move- ment has virtually no topographic expressions. In addition, on a nearly featureless and horizontal surface there are no topographic features to be offset. These conditions obtain in parts of the north break and almost all the central and south breaks southeast of location 23.5. Offset channels are virtually absent from the zone of the 1968 rupture, despite clear evidence of repeated late Holocene movement (Clark and others, this volume). Large channels incised 3—6 m into the surface cross the fault at locations 24.8 and 27.8 (visible in fig. 23) with no apparent deflection. (The prominent right-lateral offset in one of the washes about one-half km downstream from the south break at location 28 has no apparent tectonic cause.) More- over, virtually none of the many small deep well— developed channels and gullies that cross the central break along the northeast base of Ocotillo Badlands are offset. Such offset is common elsewhere along the San Jacinto fault zone (Sharp, 1970) and along other active strike-slip faults in California, such as the San Andreas and Garlock faults. (See Vedder and Wallace, 1970; Clark, 1971.) The primary reason that channels along the 1968 rupture are not offset is apparently that they are being enlarged by erosion faster than they are being offset by tectonism. The large channels southeast of Ocotillo Badlands at locations 24.8 and 27.8 lay below the maximum level of Holocene Lake Cahuilla (roughly the 40-ft contour line on pl. 1) and must have developed after the lake dried up, about 800 years ago (see Clark and others, this volume), be- cause the channels cut deposits of that lake. Since the lake disappeared the total lateral displacement across the fault in this location has probably been less than 1 m, based on 1968 displacements (about 100 mm maximum) and probable recurrence in- tervals for 1968-type events (about 200 years). (See Clark and others, this volume.) Horizontal displace— ment of 1 m almost certainly would not be evident in walls of channels that have widened by 10—30 m in the same period. Channels in Ocotillo Badlands cross the fault above the highest levels of Holocene Lake Cahuilla and hence are much older. However, the significant vertical component of uplift of the soft sediments of the badlands that attended the 1968 earthquake (pl. 1) and earlier ones evidently has promoted deepening and widening of nearly all channels sufficient to erase horizontal offsets. The central and south breaks evidently act as ground-water barriers, and zones of relatively abun- dant mesquite on the southwest (upslope) side of THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 69 FIGURE 37.——Central break at Old Kane Spring Road. Dashed lines show position of breaks formed since the earthquake. The central break follows alined bushes (barely evident across the center of this photograph) along a northeast-facing scarp 1/2—1 m high. Small chan- nels incise the relatively upthrown block (upper half of picture). The central break also forms the northeast boundary of prominent mesquite-stabilized dunes. Light surface tone indicates the white shelly youngest deposit of Holocene Lake Cahuilla. Removal of this layer from the relatively upthrown blocks defines a graben between the 476-246 0 - 72 - 6 central break and a branching break from locations 20.1— 205. Southeast of location 20.7, the shelly layer is pre- served on both sides of the central break. Arrows indicate fractures north of Old Kane Spring Road that follow alined creosote bushes (Larrea), apparently the location of previous fractures or collapse fissures. Faint north- south lineaments across the central break from locations 20.8—21.4 are creosote bushes along old tracks of vehicles, probably from World War II maneuvers. Aerial photo- graph by Light Photographic Squadron 63, US. Navy, Miramar, Calif., May 1968. 70 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 FIGURE 38. — View westward over south break (open circles) in November 1956. Mesquite-stabilized dunes are concentrated on the southwest side of the fault, which acts as a barrier to ground water. Relict collapse fissures visible as lines of bushes are at a, b, and d. A prominent collapse fissure developed later (but before the 1968 earthquake) at 6. (See Clark, “Collapse Fissures Along the Coyote Creek Fault,” this volume.) U.S. Geological Survey photograph OAJ 5—129, November 1956. ' the fault reflect this (figs. 37, 38). Mesquite (Pro- are conspicuous, both on aerial photographs and on sopis juliflom), a phreatophyte whose roots may the ground, any linear arrangement of or boundary penetrate to 20 m (Kearney and Peebles, 1960, p. to the mesquite dunes is easy to recognize and to 402) , tends to accumulate blown silt and sand, creat- check for fractures. However, such linear aspects ing prominent dunes as much as 5 m high from of the mesquite dunes that exist away from the which unburied branches protrude. Because they rupture did not show fractures. Some may represent THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 71 formerly active traces, but others may well result from other causes. Creosote bush (Larrea) , another plant very abun- dant in the area of the 1968 rupture, also commonly grows or flourishes along breaks. Indeed, since 1968, many creosote bushes along collapsed fractures have grown markedly and have become much fuller and greener than others nearby because the adjacent fractures have captured surface runoff. Several collapse fissures in the vicinity of Old Kane Spring Road and on the south break followed old lines of creosote bushes, presumably former breaks" or col- lapse fissures (figs. 37, 38). In many areas where older sediments are exposed, the usual geologic evidence of faulting also coincided with the 1968 rupture. Lithologic changes and drag effects on structural trends along the northeast base of Ocotillo Badlands (described in detail in Sharp and Clark, this volume) clearly coincide with the position of the 1968 rupture. Along parts of the 1968 rupture, wide zones evi- dently must be considered vulnerable to future frac- turing and large displacements, despite the lack of surficial evidence of similar broad bands in previous episodes of faulting. Although the 1968 rupture closely followed virtually all scarps and other evi- dence of former fracturing along its path, many individual fractures of the rupture formed where no obvious surficial evidence of previous breakage existed. Some of these fractures formed in sections of the rupture that entirely lacked such evidence, but others, including many with displacements greater than 50 mm, extended away from or developed parallel to either side of obvious lines of earlier faulting for distances greater than 100 m. Moreover, some fractures, including those with displacements greater than 50 mm, formed along lines known to have remained unbroken during the last 300—1,600 years, yet they were adjacent to traces on which movement had occurred many times in that same period (Clark and others, this volume). Where a branching or subsidiary break intersects a primary break, the zone of fracturing may be much wider. In the region where the central break and its large subsidiary break cross Old Kane Spring Road, fractures occupy a zone as much as 1 km wide and more than 3 km long that trends roughly east- west across the fault (locs. 19.8—22.8). Although some of the fractures in this area followed older scarps or alined bushes (fig. 37), many appeared to be randomly oriented. With the information in figure 35 and knowing how the various types of surficial evidence generally are associated with faulting in this region, R. V. Sharp and I estimated how well the position of the main breaks of 1968 could have been predicted be- fore the earthquake (Youd and Castle, 1970, p. 1206—1207). The location of the principal fractures could have been predicted to within :3 m along about 10 percent of the length of the main 1968 rupture, within :10 m along an additional 20 per- cent of the length, within :100 m along another 20 percent of the length, but only within :500 In over most of the remaining 50 percent of the 1968 rupture. The predictions within :3 and 10 m were along the prominent scarps of the central and north breaks. Those within :100 m were on parts of the central break and north break along or near scarps. Those within :500 m included places along the remainder of the rupture that lacked obvious or unambiguous surficial evidence of the presence of the fault. ENGINEERING IMPLICATIONS 01‘ PATTERNS OF FRACTURING Patterns of fracturing and the distribution of dis- placement within those patterns must be considered in evaluating the potential hazard to manmade structures near the 1968 rupture. The preceding section describes the clear evidence of earlier fault- ing, such as scarps, that coincide with the main fractures along many parts of the 1968 rupture, but emphasizes that other fractures extend well beyond all surficial signs of earlier faulting This section describes the variation in the width of the band of fracturing that constitutes the 1968 rupture and the possible widths of bands in which damaging dis- placements might occur. The engineering implica- tions of other recently active faulting in terrain farther away from the 1968 rupture are discussed by Sharp and Clark (this volume). The variation in width of the band of fracturing along the 1968 rupture is shown in table 9. The large subsidiary break near 01d Kane Spring Road TABLE 9.—Width of the main rupture along the Coyote Creek faultl Width of Cumulative P 3303:2196 rupture! length“ length a f Cumulative percentage (m) (km) rupture‘ 0-50 ............... 1 9. 7 64 64 100 50—1 00 ............. 2.7 9 73 36 100—500 ............. 5.4 17 90 27 500—1000 ........... 1.0 3 93 10 > 1000 ............. 2.2 7 100 7 1Estimated from fractures recorded on pl. 1 between locs. 1.8 and 32.8, including all diverging ruptures, all fractures between the central and north breaks in Ocotillo Badlands, and all fractures between locs. 19.8 and 22.7. Isolated fractures or groups of fractures 1/2 km or more from the main breaks at locs. 2.9—3.8; 4.6—4.9: 13.1—14 to the northeast and 15.0—16.3 to the south- west of Ocotillo Badlands ; and 27.2 were not included in the rupture zone for these estimates. ”Measured normal to the coordinate axis of pl. 1. ”Measured parallel to the coordinate axis of pl. 1. 4Length assumed to be 31.0 km : rounded to nearest 1 percent. 72 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 is included as part of the band of fractures of the adjacent part of the central break for these estimates because it may connect with the central break at depth and the area between them contains scattered small fractures. Other isolated fractures more than one-half km from the main break have not been included in this band. (See table 9.) The estimates show that fractures occupied a band more than 50 m wide along about 35 percent of the length of the 1968 rupture, a band more than 100 m wide along about 25 percent of the length, and a band more than 500 m wide along 10 percent of the length. Bands of fracturing less than 500 m wide occur at many places along the 1968 rupture; wider bands occur in only a few places, and these generally displayed pre- 1968 geologic clues about the width of the zone of fracturing. For example, at locations 14—152 in Ocotillo Badlands and locations 20—23 along the central break, tectonically upwarped hills bounded on one or more sides by scarps mark the positions of these unusually wide bands of fracturing. Be- cause of their more widespread occurrence (table 9), the bands less than 500 m wide seem to be the most important for estimating the probability that fractures would appear at a given distance from any part of the Coyote Creek fault. On the basis of the fractures that formed in 1968 and later, a 100— 500-m-wide zone along the entire fault must be considered vulnerable to fracturing, inasmuch as 17 percent of the length of the 1968 rupture was be- tween 100 and 500 m wide and 27 percent of the length was more than 100 m wide. The fact that a band of fractures developed on one side or the other of the principal fractures at dif— ferent places along the rupture in 1968 (pl. 1) indicates that estimates of the extent of future frac- turing along an identified fault trace (generally a line only a few meters wide, such as a scarp) should include a band on each side that is as wide as the widest band of fracturing that could be anticipated from field evidence. Thus, the band vulnerable to future fracturing along a given fault will be twice as wide as the anticipated band of fracturing. At most places along the 1968 rupture, the zone of potential damage to structures is narrower than the zone of fracturing because the outermost frac- tures in a zone tend to have small displacements. If we assume that displacements of 50 mm (2 in.) may damage structures, however, then the zone of potential dama e along the main breaks, although narrower than the band of fracturing, is consider- ably wider than the 20—m—wide band that contained most of the displacement in 1968. Although most fractures with 50 mm or more of displacement lie within this 20-m band, enough extend beyond in different places along the rupture to create a signif- icant general risk to structures. In several places, fractures with 50 mm or more of displacement ex— tended more than 100 m from the main break (for example, locs. 18, 30, 31.3, pl. 1). In the vicinity of Old Kane Spring Road, such fractures were widely scattered over the 1 by 3 km area across the central break and the large subsidiary break to the east. The available observations, however, are far too meager to allow any quantitative estimates of varia- tion in width of the band within which displace- ments of 50 mm occur along the 1968 rupture. The degree of risk to structures near the Coyote Creek fault depends on how far fractures with some minimum amount of offset might extend from the principal line or lines of fracturing and displace— ment. To assess this risk we must know (1) the location of the most recently active principal traces, (2) the patterns of fracturing that are likely to occur during an earthquake, and (3) the distribution of displacement within these patterns. Lacking data of the third sort, we cannot make specific assessments of risk along this fault. How- ever, the prevalence of apparently active faults in the general region of the 1968 rupture (see Sharp and Clark, this volume), plus the information dis— cussed in this and the previous sections, indicates that before any large structure or multistory build- ing is built within .that general region, the sub— surface should be examined. Until we learn more about distribution of displacement next to the fault, such subsurface examination would seem to be mandatory before any such structures are built within about 1 km of the 1968 rupture or other identified recently active faults nearby. CREEP After the 1968 earthquake, displacement (fault creep) continued along parts of the central and south breaks but not along the north break. Through November 1971, this postearthquake movement had added as much as 50 percent to the displacements measured during the week after the earthquake (pl. 1; fig. 32). Fault creep has occurred along most of the central break (nearly continuously along some parts), beginning shortly after the earthquake. In contrast, creep apparently did not occur on the south break until about 1 year after the earthquake, and then it occurred only in the middle part of the break. Figure 39 shows areas affected by postearth- quake creep along the 1968 break, and table 10 gives approximate dates of this activity. THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 OCOTILLO WELLS . (. \./“'\I 02/11”; [x (For V” FIGURE 39.—Areas of postearthquake displacement (a, b, c, d) after April 14, 1968, along the surface rupture. Summaries of dates and total creep are in table 10. TABLE 10.—Summary of creep at locations indicated in figure 3.9 Estimated total Date creep . Date creep Date creep ceased horizontal creep Area April 14, 1968, first noted commenced (as of to March 1971 November 1971) (mm) Before July 5, 1968.. 80—120 a ........ June 1968 ...... Probably after April 14, and definitely before June 19, 1968. Probably between May 26 and June 14, 1968. After June 20, b ........ June 1968 iiiiii Probably after Jan. 1 , l ; definitely before Mar. 22, 1969 Approximately 120 c ........ March 1969... 1968, and before Still moving as of 80—110 Jan. 14, 1969. November 1971. d ........ October 1969 Probably after Latest movement 30—60 between Decem- ber 1970 and November 1971. Jan. 14, 1969. Possibly as late as August 1969. Fault creep first became apparent at several places during mapping along the central break in June 1968 and could still be detected by tape measurement in March 1971. Although a few precision strain- measuring instruments and stations were established by other investigators on the north and central breaks shortly after the earthquake, Only the array on the central break recorded large changes (Bur- ford, this volume). Observations of this geodetic station have not been so frequent as the far less precise observations reported here, which although based in part on indirect evidence, offer a more complete record of both timing and location of creep along the 1968 break. Most changes in displacement were measured with pocket tapes between matching irregularities on opposite sides of fractures. Periods of creep could be determined only within broad limits defined by our infrequent visits to the area and by 73 our ability to deduce when creep occurred with re- spect to sporadic heavy rains in the area. Three general types of evidence indicate con- tinued displacement along the rupture: (1) the appearance of new (postearthquake) fractures, (2) an increase in displacement measured at a specific point on the rupture (for example, at a certain tire track), and (3) a general increase in displacements measured along a segment of the rupture in which exact locations of previous measurements cannot be recovered. The first two types of evidence are best for documenting creep. However, the second depends on rediscovery of individual displaced features, few of which were suitably recorded during the week fol- lowing the earthquake. The creep discovered in June 1968 became evident only through observation of a general increase of 50—100 mm in displacements along specific areas of the fault (locs. 22.2, 24). As a result of the discovery of postearthquake creep, however, many individual tire tracks and fractures were photographed in June 1968, and some survived erosion to indicate the presence or absence of creep during ensuing months. New postearthquake fractures, the most reliable sign of continued movement, can commonly be dated with respect to infrequent periods of intense rain. Several times during the 3-year period after the 1968 earthquake, hard rain created Widespread sur- face runoff that eroded or caused collapse of the walls of many open fractures. (See Clark, “Collapse Fissures Along the Coyote Creek Fault,” this volume.) Fractures that developed after rain were easy to distinguish from those that had been open during heavy rains or runoff. The former had sharp, angular edges; the latter rounded, eroded edges. Intense rains also formed a crust on windblown silt and sand that filled many fractures, and creep after the rain broke this crust. RUNOFF ALONG THE 1968 RUPTURE In order to establish dates of runoff that limit episodes of observed creep, we must combine rain- fall records, which give dates when surface runoff was possible, with repeated observations along the fault to determine which storms produced runoff. (See fig. 40.) Most of the scant rainfall in this area (about 70 mm per yr) generally comes from wide- spread winter storms and is commonly of relatively low intensity; however, a few cloudbursts connected with thunderstorms usually occur during the sum- mer months. Rain of sufi‘icient intensity or amount to cause surface runoff and obvious modification of fractures falls only during the largest winter storms or during summer showers. Unfortunately, published 74 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 -30 ’ 7 Heavy runoff on all breaks 1.0— . . . w _ Runoff likely along entrre break : _ \ a - 20 ._ E Ocotillo Wells ——> (—Brawley 2 SW - o: -10 - < _ LLl o.murmur”......l.n.......|..l. "l'"""'ll'l' ...... ......... .'..... .0 JAN FEB MAR APR MAY JUN E JUL AUG I SEP 1 OCT I ._______ - . . I No record at Ocotillo Wells ‘—*l 1968 Pernod of field observation 130 1.01- : — 20 '_ <- Runoff probable on central and south breaks a . Lu ~10 I ‘2’ l | :0 Iv rw'rrrvrr vu-r-"rr‘r vylrv'rv" nnnnn vv r“ lvvvlvw—m—r—H—fi—rv—rq‘r'r vvvvvvvvvvvvvvvvvvvvvvv u 0 NOV DEC I JAN FEB MAR APR MAY l JUN 1 JUL r AUG 2 4 1968 1969 m _‘ l— E 10 'l 30 m - Heavy runoff on south break' E Z ' Heav runoff on o ' - 3 moderate on central break no Wynn" on cenirstgrggiak' / Runoff on central break: .1 m f—k—fi probable on south break _ 20 -_-‘ E - 10 >. 5 ll h o .. .v.l..l... ... 'Il .......... l ......... l .. 0 JAN FEB MAR APR MAY JUN 1969 1970 Lu 2 l 0 ‘ 3° 2 S R -‘ 20 E unoff on central break; 8 probable on south break\ f—Jl W1 - 10 0 l .V ......|., ......|. r ..... l ......... l at...” ......l.....v‘...l..”unruln 0 JUL AUG SEP OCT NOV DEC l JAN FEB MAR 10 1970 1971 : l1— No record at Ocotillo Wells l Runoff on central and south breaks I l l I 0 vvrlrlllllllllllrllllIxIIIIIIIIlllllllllllilllllllllllllll|vlllllllllr'llllllllll’ l APR MAY JUN JUL AUG SEP OCT NOV 1971 FIGURE 40. — Rainfall, estimates of dates of surface runoff on the central and south breaks, and dates of field'investiga- tions (heavy bars below time axis), January 1968 through November 1971. Rainfall is shown as cumulative totals for 3-day periods; from Environmental Data Service (1968—71), records for Ocotillo Wells (heavy line) and Brawley 2 SW (light line). data alone do not indicate when rainfall produced surface runoff. US. Environmental Data Service weather records for Ocotillo Wells list total daily precipitation but not intensity of precipitation for shorter periods. The nearest station that records hourly precipitation, El Centro, is too distant to yield reliable estimates of intensity in the area of the rupture. Thus, detailed before-and-after field ob- servations of channels and other surficial forms are necessary to establish timing and location of episodes of runoff in the fault zone. Rainfall reported at Ocotillo Wells is assumed to be representative of rainfall along the entire 1968 rupture. This assumption is supported by figure 41 and by the fact that vegetation is Virtually uniform along the fault. The record from Brawley 2 SW, about 40 km east of the south break, is also included in figure 40 because parts of the Ocotillo Wells record in winter months are missing, and Brawley rainfall is generally quite close to that of Ocotillo Wells in winter. Seven periods of runoff are crucial to our recon- struction of the timing of creep episodes after April 1968. Rain that definitely caused surface runoff and ensuing collapse of open fractures along the 1968 break fell on July 6, 1968, September 6—15 and November 9—10, 1969, February to March and August 1970, and August 1971. Rain on January 14, 1969, probably led to local runoff and collapse, but previous observations during the brief field investi— gation of October 1968 were insufficient to permit an unequivocal distinction between all effects of the THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 orregoN 0 Desir 3 OCOti Ilo Wells 6 —I968 break 3 ‘5’ . °_ ‘3 o BRAWLEY 33 Brawley 2 SW I0 3 ’3 2o 4 . EL CENTRO o O ——16 ‘I- — - M'Exico EXPLANATION 8 . |sohye1 ‘ . . Sham/lg Inga/z annual prewar/aim” m Inches O Weather-recording station ”6° 1 FIGURE 41.—— Mean annual precipitation in the region of the 1968 break, 1931—60. From Hely and Peck (1964, pl. 3). 1 in. = 25.4 mm. rains of July 6, 1968 and January 14, 1969. Precipi— tation on July 6, 1968 (1.05 in.) was the highest amount recorded in a single day from 1967 through 1971 at Ocotillo Wells and resulted in heavy rain throughout the region and widespread surface runoff along the entire 1968 break. CREEP ON AND NEAR THE CENTRAL BREAK AREA A (FIG. 39) Obvious signs of continued displacement along the 1968 break were first noted in June 1968 near loca- tion 22.3, the part of the central break on which vertical displacement was reversed in sense for a short distance along the prominent scarp southeast of Ocotillo Badlands. Maximum vertical displacement increased from 100 mm in April 1968 to 150—230 mm 2 months later, and horizontal displacement in- creased from about 180 mm to 250—300 mm (fig. 27). This same site was measured during all subsequent visits to the fault zone, but no further displacement could be detected. Creep also affected the central break from location 22.5 southeast as far as location 25.0. Although no repeat measurements were possible at individual sites (the only offset exactly located and measured in this section in April 1968 was obliterated by the following June), many measurements made in June were significantly larger than any made in April, 75 implying additional movement (pl. 1; fig. 32). Measurements in October 1968 and later at many sites marked carefully the previous June along this part of the fault showed no further displacement. The heavy rain of July 6, 1968, enlarged many fractures in this area by collapse (fig. 42). However, by March 1968, all the collapse fissures along this part of the fault had been filled with sand and silt, a further indication that movement had ceased. AREA B (FIG. 39) Persistent and widespread postearthquake creep has taken place on many fractures along the subsid- iary break east of the central break. This area pro- vides clear documentation of formation of new fractures after the earthquake and their subsequent enlargement by collapse. The history of development of fractures in this zone is not complete because mapping in April 1968 did not cover all ground in which fractures were later found. Some fractures discovered as late as January 1970 might have de- veloped as early as the time of the earthquake. For- tunately, detailed mapping in certain parts of this area in June 1968 documented later creation of new fractures. The first indication of continuing displacement in this area after the earthquake was reported by W. L. Cantrell, manager of a motel near Ocotillo Wells. On June 9, 1968, he discovered a prominent fracture on the low hill about 2 km east of the intersection of Old Kane Spring Road and the central break (loc. 21.9). He was certain that the fracture was not present 2 weeks earlier on May 26 when he visited the same hill (oral commun., 1968). On June 21, I mapped this creek and associated fractures north of Old Kane Spring Road and searched the area between it and the main break without finding other frac- tures. Subsequent field checks indicated that move- ment continued in this area east of the central break until sometime between January 14 and March 22, 1969. As determined from evidence of surface runoff, new fractures evidently formed in this area in each of three periods after June 21, 1968: June 21 to July 6, 1968, July 6, 1968 to January 14, 1969, and Janu- ary 14 to March 22, 1969. Since March 1969, no new fractures have appeared here, nor have old fractures opened further. Figure 43 shows new fractures that formed in this area after June 1968. Most individual fractures were oriented roughly north-south, as on the main break, but many others were oriented closer to N. 30° E. In accord with observations made on the main break in April, these fractures showed significant compo- nents of opening movement as a result of their 76 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 orientation 30°—60° from the regional trend of the central break. Maximum opening was about 50 mm; however, most fractures opened 20—30 mm. AREA 0 (FIG. 39) The most complete record of postearthquake creep on the central break comes from the section near Ocotillo Badlands. Creep commenced here between June 1968 and January 1969 and continued through November 1971. By March 22, 1969, displacement of tire tracks immediately north of the road that inter- sects the central break at location 19.6 had increased to 210—240 mm from the 130—160 mm measured in the same area in April and June of 1968 (fig- 44). FIGURE 42.—Time sequence showing collapse and eventual filling along a fracture of the central break at location 23.8. A, June 23, 1968. Opening of the tectonic fracture (be- tween arrows) has increased from about 10 mm in April to 20—30 mm. B, October 20, 1968. Surface runoff from the heavy rain of July 6, 1968, caused slumping and collapse of the walls of the fracture, followed by partial filling when water could no longer flow readily into the depths of the fracture. A tire track made June 23 became a chan- nel conducting water to the fracture. No measurable creep occurred at this location after June 23, 1968. C, April 29, 1969. Blown silt and sand fill most of the collapsed fracture and channel. Creep after the rains of January 1969 left a clear record of offset crusts and channel deposits through- out the length of this segment of the central break (figs. 45-47). As measured in March 1969, offsets across these fresh, postrain fractures diminished steadily northwestward from about 40 mm near location 20.5 to 5 mm near location 15 (fig. 47). Postrain displacements to the southeast along the central break diminished to zero between locations 20.5 and 22.3. Subsequent observations in 1969, 1970, and 1971 revealed evidence of further small displacements. Silt and sand filled these new fractures after March 1969; then the rains of September 1969 formed new crusts on this material. New channel deposits of hard silt at location 17.6 and new crusts that formed in the main fracture between there and location 19.6 showed new fractures in October 1969 with about 1—2 mm of right—lateral displacement. (Cracks with THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 77 ”6° 02' 30" L-——— HQ 32 County Wel , _ o(Dry) "s7=7e7nz.-« 33° 05 Bose from U. 3. Geological Survey l:24,000 Borrego Mountain SE, l958 FIGURE 43. — Fractures (heavy lines) that apparently formed near the central break between June 21, 1968, and March 20, 1969 (light lines show fractures that formed at the time of the earthquake and afterward until June 21, 1968). Many of the fractures probably formed after October 1968. Postearthquake fractures labeled “a” follow alined vegetation (evident on 1953 aerial photographs) that apparently grew on earlier fractures. Base and kilometer grid from plate 1. openings of less than 1 mm are easily Visible in the hard silt of these channels. The tectonic fractures described here are precisely on the 1968 rupture, show en echelon patterns, and generally have larger openings than shrinkage cracks in the same depos- its.) By January 1970, displacements on some of these same fractures had apparently increased by about 1 mm more. Heavy rains in February to March 1970 and high winds erased these small fractures by March 7, 1970. Whether or not new fractures formed between then and the heavy rains of August 1970 is unknown, but any displacement in that period probably did not exceed 10 or 20 mm. New fractures developed again along the central break after August 1970 between locations 17.6 and 20.6, with as much as 10 mm of displacement, as observed in December 1970. At location 17.6, this displacement increased by several millimeters by March 1971 and by several more before November 1971. Figure 48 summarizes the history of displacement measured near location 19.4 on the central break within area 0 of figure 39. These crude measure- ments of postearthquake creep are grossly com- patible, except for an initial delay, with the exponential decay of creep found by Wallace and Roth (in Brown and others, 1967, fig. 25) and by Smith and Wyss (1968) after the Parkfield-Cholame earthquake on the San Andreas fault. Furthermore, Burford (this volume) inferred exponential decay of creep for the central break from a nearby aline- ment array. The data for the central break contrast with those for Parkfield-Cholame in that creep on this segment of the central break evidently did not begin immediately after the earthquake. 78 THE BORREGO MOUNTAIN FIGURE 44.—Tire tracks at location 19.4 offset 240 mm horizontally. March 1969. Rain, blown sand, and newer tracks had virtually obliterated these tracks 1 year after this photograph was taken, preventing further measure- ment of total offset at this site. Photograph by C. R. Allen. CONTINUED DISPLACEMENT ON THE SOUTH BREAK AREA D (FIG. 39) Probably the most unusual zone of continued dis- placement along the 1968 rupture is on the south break near location 31. (See fig. 49.) Large displace- ments did not occur on the south break during the earthquake, nor did it show any signs of continued movement in the months immediately following. Yet significant creep commenced between January and September of 1969 and was still apparent in a few places in December 1970. This creep reopened 1968 fractures, created some new fractures, and reopened a relict fracture that predated the earthquake. Be- cause no displaced tracks or other evidence of the original offset of April 1968 survived after Septem- ber 1969, we have no measurements of cumulative displacement after the earthquake. Creep on the south break has added an estimated 30—60 mm of horizontal displacement to the maximum of 80 mm measured in April 1968. Careful examination during nearly every Visit to the 1968 rupture has enabled us to reconstruct the history of postearthquake creep on the south break. A detailed check of the entire south break in June 1968 revealed no indication of continued displace- ment. Indeed, north of location 31.1, many of the EARTHQUAKE OF APRIL 9, 1968 FIGURE 45. —- New breaks with 30 mm of right-lateral offset. March 1969. These fractures broke the crust that was created by rains of January 1969 on blown silt and sand filling a large 1968 fracture (indicated by arrows). Loca- tion 19.4. fractures were no longer visible 2 months after the earthquake because they were filled by blown sand and silt, and between locations 29.4 and 25 all signs of the 1968 rupture were obliterated by drifted sand. However, a field check in October 1969 revealed renewed displacement along the 1968 fractures of the south break and creation of two new fractures di- verging from it at location 31.4 (fig. 49, e). The rains of September 1969 had caused large runoff and initiated major collapse of many of the newly opened and reopened fractures, including the two new frac- tures. Both these new fractures increased in length by more than 30 m after the rains of September, as shown by sharp-edged postrain extensions of each across channels that carried water in September (fig. 50). Triangular arrays of stakes placed across THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 79 FIGURE 46.—Detail of the fracture shown in figure 45. March 1969. The pencil lies parallel to the direction of tectonic movement; displacement here is 30 mm. these two new fractures in October 1969 showed that the southernmost of the two fractures had opened another 10 mm by December 1970. In contrast, other nearby new and reopened fractures apparently ceased further significant movement late in 1969. Two additional new fractures (fig. 49, g) developed in this area after early November 1969 and con- tinued to open and extend between January and De- cember 1970. Movement also recurred on a prominent preearthquake fissure (fig. 49, 0, described next) after January 1970 and continued in the period August-December 1970. Total opening on these latest fractures did not exceed 15 mm, and movement apparently ceased by December 1970. The period during which the ruptures of April 1968 first reopened and the new fractures began to form on the south break is fairly well defined. These fractures were definitely not open at the time of the field check in June 1968. The field check of March 1969 did not include parts of the south break that displayed renewed displacement the following Octo— ber. However, large-scale aerial photos taken in April 1969 along the entire rupture show only one section of minor collapse on the south break, about 3 m long at location 30.4, in the center of the area of later extensive collapse. (This same series of photo- graphs clearly shows all the collapse fissures else- where along the 1968 break.) Lack of collapse in April 1969 means that the fractures had not re- opened at the time of the previous runoff, very probably January 1969 but certainly July 1968. Thus, renewed displacement on this part of the south break definitely began between July 1968 and Sep- tember 1969 and probably began between January and September 1969. The fact that the two new fractures (fig. 49, e) continued to extend after the rains of September 1969 suggests that the tectonic activity had begun shortly before those rains. In addition to renewed displacement after perhaps 1 year or more of stability subsequent to the earth- quake, the south end of the 1968 rupture is notable for its association with preearthquake fractures. A fresh collapse fissure and at least three prominent groups of relict collapse fissures nearby all predate the 1968 earthquake. Of these fissures, shown in figure 49, a—d, a and (1 developed before 1953, b developed after 1953—56 but ceased movement sev- eral years before 1968, and c evidently opened or reopened less than 1 year and possibly just a few months before the earthquake (Clark, “Collapse Fissures Along the Coyote Creek Fault,” this vol- ume). All these fissures almost surely developed from tectonic fractures. In addition, several small (less than several meters long) alined sinks that predated the earthquake were present on the south break at locations 28.4 and 32.0. The relict fissures, creep behavior, and small total displacement since 1968 suggest that the region at the south end of the 1968 rupture is one of more frequent tectonic activity than the areas to the north along the 1968 rupture. The older fissures at a and d appear to represent tectonic activity perhaps 15—20 year's before the earthquake, whereas b seems to have developed, or at least collapsed, after 1953 but well before 1968. Fissure c strongly indicates tec— tonic movement in this area shortly before the earth- quake. The reactivation of the 1968 break and creation of new fissures resulted from creep that has continued for more than 2 years after the earth- quake. PATTERNS OF CREEP ALONG THE 1968 RUPTURE Ready explanations are not apparent for the pat- tern of postearthquake creep indicated for the 1968 rupture. Creep did not simply migrate from one end Of the rupture to the other, as at Parkfield in 1966-67 (Smith and Wyss, 1968). It started first on the southeast end of the central break and on the outly- ing fractures along Old Kane Spring Road but then shifted by unknown steps to the northwestern part of the central break. The net result of creep on the central break has been to increase the amount of displacement closer to that of the north break than it was immediately after the earthquake. No clear relation exists between creep and the pattern of larger local aftershocks. Figure 51 shows the loca- tions of aftershocks of M230 through 1970. (See also table 11.) Although the delayed onset of creep along the south break might be connected with the 80 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 |l6°05' 33°07’30“ Base from US. Geological Survey |=24,000 Shell Reef, |959; Borrego Min 5 E, I958 FIGURE 47.—Horizontal displacement, in millimeters, measured March 21—22, 1969, on either (1) post-January 1969 fractures in crusted deposits filling the 1968 fracture, (2) new channel deposits laid across the 1968 fractures in January 1969, or (3) new post-January 1969 fractures in sediments that were not broken in 1968. Direction of offset shown by arrows. Base and kilometer grid from plate 1. enlarging “doughnut” of aftershocks noted by Allen creep along the central break is not obviously related and Nordquist (this volume), the persistence of to any pattern of recorded local aftershocks. THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 81 .4 C u: E 0 m w .5 “ ~ 2 .- m E c g a s = g s w 3 - E g m ._ x. ,9 _ u E c E v- .- c o ‘0 a In a; E 3 m '0 » a: -— , -— o s a u — E .E E g 3 E 3 2 g LO 3 g a d g ‘5 z 3 t3 1? 2 x g :3 2 m E c ‘6 3 '5 : z; E t: 5 .C .o n: O _‘ E ‘° 2 '5 '5 '— E 3 E a) :a “ c 2 ‘c o '5 "‘ G 5 D E G B E a E o D w m a 5 a E N 0‘: F m I9 2 E g a E 1:: 5 u. <3 5 _ 13 ‘0 3 5 E > “ T: u o r» m x -— o a: a) t: a; av _ 2 w 2 w u ."2 u v 3 - u — o g g .9 '6 8 ‘ L“ 5 E _ *- 3 5' ..' E 1: a B 3 250— E3 o“ ‘5: 2.3 02% ° 1'9 5% ° 0 '6 z ‘5 2 .o o v (2 _ z o a o _ T 2 T T r r l I I I I I _f’/’-—i——l ______ —i—————1—----t——I———-?-*———L— l I I I l l i 1 i l 200- T 4‘ E E I— , , ’ ’:/ I z , I ’ / / ‘L L” I / ’ ’ / / E //’ /// EXPLANATION u / // j ‘50— Total displacement measured from & offset tire tracks. Bar represents 0 estimated range of accuracy 2‘ l5 Difference between displacements l— measured MAW” and June 1968 is T Total displacement estimated from I00— not necessarIly sIgnIfIcant because i total measured on a diffe'enl date measurements may not have been 1 plus displacement measured or made 0" the same ""9 tracks J_ estimated in the intervening period Lines connecting successive measurements are dashed where slope is uncertain 50- Episode of rainlall Period of field observations i-IJ'.I.|...1.. .......'li|--l 1.111....WHf .fi APR JUN AUG ocT DEC IJAN MAR MAY JUL SEP NOV I JAN MAR MAY JUL SEP NOV [JAN MAR 1968 1969 1970 1971 FIGURE 48.—Displacement from April 9, 1968, to March 1971 near location 19.4. Ofi'sets were measured on tire tracks and rain-consolidated crusts along a segment of the central break about 100 m long. TABLE 11.—Ea7‘thquakes of M23 in the area from 32°45 to 33°15’ N. and from 115°45’ to 116°15’ W. from April 29, 1969, through 1970 [Epicenters are plotted in fig. 51. See Allen and Nordquist, table 2, this vol- ume, for an explanation of symbols. From preliminary lists of shocks in southern California during 1969 and 1970 supplied by C. R. Allen, California Inst. Technology] ‘3 Year 5 _§ '3 Latitude Longitude E Quadrangle 8 a: a .= § 3: E o m i m 2 07 3155.67 33°10’71” 116°02’58" Borrego Mountain. 10 58 13.92 32°59’57" 115°47’37" Plaster City. 11 19 25.66 33°10’15" 115°48’97” Kane Spring. 11 22 30.38 33°01’12” 115°48’31” Do. 13 59 37.62 33°12’76" 115°57'14” Do. 22 40 06.46 33°00'37” 116°12’44” Borrego Mountain, 15 29 57.24 33°01’17” 116°13’50" Do. 15 39 24.89 33°00’12” 116°13’02” Do. 20 01 37.84 33°00’68” 116°14'50" Do. 14 38 17.85 33°00’31” 115°50'46” 23 17 30.17 33°10’58” 116°03’83” 16 25 11.70 33°11’94” 116°07’36" 01 43 19.46 33°09’36" 116°07’19” 22 18 41.99 32°55’27” 116°09’38” 15 15 35.01 32"56’28" 116°10’48” 14 59 39.87 32°58'52" 115°53’51” 12 04 36.65 33°10’72” 116°03’59" 00 06 22.56 33°14'43” 115°57’63” 10 23 57.15 32°50’45" 115°51'77” 16 04 07.09 33°05’53” 116°08’97” 03 49 26.41 33°01’62" 115°56’50" Kane Spring. _ Borrego Mountain. Do. Do. Carrizo Mountain. Kane Spring. Plaster City. Borrego Mountain. Kane Spring. Plaster City. Borrego Mountain. Kane Spring. wwnwo>womoo>wwmwowno§a Quality 9999999999999wew9wwww Nwa——¢wn~mq¢oboooohwh The observed creep conceivably is related to after- shocks of M<3, for which there is no reliable cover- age during the period considered here, or it may reflect a delayed response to the initial or subsequent displacement in underlying rocks, as discussed in this volume by Burford and by Allen, Wyss, Brune, Grantz, and Wallace. THE UNUSUAL STATUS OF THE SOUTH BREAK The history of seismicity and displacement along the south break has only a few characteristics in common with that of the central and north breaks, in addition to the rupturing of all three in 1968. A major similarity of the south and central breaks is that both experienced postearthquake creep, although at different times and in different amounts. The major common characteristic of the south and north breaks probably is their proximity to epicenters of large earthquakes. A magnitude 6.5 earthquake oc- curred on or near the southeast end of the south break in October 1942. (See Allen and Nordquist; Clark and others, this volume.) No surface investi- gation has been reported for the epicentral area of the 1942 earthquake, but if the location is correct, 82 . 2. _ , Base from U. 3. Geological Survey |=24,000 Harpers Well, l958 FIGURE 49. — Special features of the south break (light lines). a—d (dashed lines), relict collapse fissures; e, fractures that probably formed after January 1969 and extended after September 1969; f, area of renewed movement after January 1969 and collapse of new and reopened fractures after the rains of September 1969 and later; g, fractures that developed after November 1969 (parts of relict fissure at c also reopened after January 1970); h, old collapse features exposed in gully. Base and kilometer grid from plate 1. surface ruptures may have formed at that time near or on the south break and, perhaps, on the Supersti— tion Hills or other nearby faults. Indeed, the older collapse fissures evident near the south break (fig. 49, d) may be remnants of breaks formed during the large earthquake of 1942. In strong contrast to the south break, however, the north break displays no comparable history of displacement immediately be- fore the 1968 earthquake and, so far, no significant postearthquake creep. Lack of evidence of preearth- quake movement on the north break could be simply the result of poor preservation of the evidence. Nevertheless, lack of postearthquake creep is an established and important difference, particularly if it is a typical difference after other earthquakes. Contrasts in displacement history and geometry between the south break and the central and north THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 FIGURE 50.—October 1969. Sharp-edged fracture (fig. 49, e) that formed after the channel that it crosses carried water in September 1969. Pencil is parallel to the direction of net movement. breaks seem to be more important than similarities. In addition to the completely different history of creep, the south break showed much smaller displace- ment in 1968 and was shorter. It trends roughly N. 50° W., Whereas the central and north breaks each trend about N. 40° W. The south break connects to the central break at an anomalous north-south segment of the 1968 rupture, whereas the central and north breaks join by an en echelon overlap that repeats the pattern shown throughout the 1968 rup- ture for smaller fractures. These differences suggest that the south break is fundamentally distinct from the central and north breaks and that it will continue to behave differently from the remainder of the 1968 rupture. The south break is possibly more a transitional element between the Coyote Creek fault and the THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 83 , ‘ .xrossnlhedsm/ J 2 / Dismmd TEN /, ,1 / V K SALTON SEA TESJ » ,:'-'.fi"-\/ \-. surface ‘ ‘ m | '0;- > ‘2 3 O ‘ x“ D Z ._. O 3 IO .................. :3 o .J [J J <0 , calfJ “7 p a o \r‘i‘ .+:;, ~\.._ \ \Jf "l \ \ ...... k 1) Belr\.§()n Uri/1+ \ " QCD‘ \__ .. ‘ Eh . 7 v“ \ \\q \ka; ( > (9‘ ,. , V, - lm‘ \ match 5 Ram h_ ~-~ \ \ ‘ x ‘ ‘ »\\_ \ . o ,5 .. ~ " a v- C"? J . Sunk“? \ ~ , /{1 {‘.,ll.arp(:r§..!4 ., i' 5/ \\ Q / ‘ l #7 Prehfiloric animal tr , 7'7»! '\ (:{é‘fw‘zsh L \ HHHHH -. ROCKY \ 4-7:“ I «0°» l) ‘2, ’ FEHBzFFK amp» \-‘ 0:! l OUNTAlrJS \ \2 i 3/ I f" *r 14 s I sumsnnort .\ I ,2, ,_ ~1_1_| l we MOUNYAIN_ _ _. .. _ V r“ . b , .s RlZO P I I ,‘L >7 I EXPLANATION ET AREA ("if '¢ARRIZO! I ’I \ ,, : ' ,« l sr MESA " O 30-3.4 g '._.-JMPACT I v, {-1 8:3 . l '3' l‘ G) 3.5-3.9 :NL.‘ AREA 5*? r“ O 40-4.4 l "‘73 , ii U s NAVAL H ‘ fizo 105‘yk ‘ ‘j (D 4.5-49 \gijf “7, \\ i } “J"J GUNNERY RAP / \l‘xd A 5.0-59 l we um: El >60 "‘77)" """"" ”l” I Earthquake epicenters. showing magnitude (1. I §———g,—§M'LES r '- ~ N AIN KILOMETERS L u I l ‘ r“ 0 Base from US. Geological Survey Salton Sea, “6 l959-67, Santa Ana, I959, Son Diego, [958-64, and El Centro, I958 April 9-June 30, l968 FIGURE 51.—Locati0n of aftershocks and other earthquakes of M23 and areas of creep (heavy line) during three periods after April 8, 1968. Locations and dates of seismic events from table 11 and from figure 9. Locations and dates of creep from figure 39 and tab1e10. Light line shows 1968 surface rupture. 84 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 Fossxl Midi, »» a“ \ \ lh'n's‘m; 1)er 41!“: “‘x [any NI‘H’VFI \ \:g»_L!xut)re ILL-wt! R/e‘c ‘ t 'vl/ , \. ‘ § \1 . \‘\ 9' I i \) . N ( / “ \ H t , 'x { j'.\\ K { /_/ / / “4" g)(/ .5‘ ~ . ~ @"flpQ / , A o / 0A0 r 674 \N / \\\ M J/— \.\ I 330‘ éHSHMEix O Mouwmms i" ‘4 S I SUPERSTH’ION 3 . (.w MOUNIAIN” V 1 ( r J —— I ~° x i , 1 CARR'ZO «15’ me/w I I v - '«V \\Y.,\:l - IMP/Aw AREA ( 5 (f / 1 WEST MESA r" ...I \ f - -JMPAN : [.1 ~~~~~ A 11 . .s U s NAVAL w. . ‘ 1 ‘ GUNNERY RAF "1\ fir ”:1 ‘ 5 ; §\ ' 'EFT'/“”’I¥‘}(.w.wk x“ g ‘ / ,/"’ / I 2 ‘T ~ V ..... NW A ()unmhn’a } /" 3 : \ _ Springs/1m: I // ,‘““'~ J. 2 ('nHA " ‘ g } ”wa4J’; / i .‘COI/ 'Ba‘-r‘1(fl§§m Sweerey Pas: I 5 [ Flange! $H’mz \ I ' \ / \\ 1 r4 ”6° July l—December 3|, I968 FIGURE 51. — Continued. THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 85 1 i 1 :' . . fir; l \ WWW I \ \ / “f 5‘ fl\ SALTON SEA T531 ( i L ‘ surface M ‘J/ h 1) _ 43 \ J f / G gowv ,A Lu 9 ~? 0 ,/ _I I i i t i i ! __1_L & § .7?” :9" 1 AN gmco INDERIAL COUNTY :5 .1 I -- c~ ,, \ 391;; 9:; L’T“ Sunl -\\\“‘ 'lIanpqrs M' u 7’ ’ ’ , ' FISH CREEK > flOUNTAIf‘lS 3" I T 1,4 3 § SUPERSTITION .\ x 76!.) MOUNIAIN_ . ( ,._ 311* CARRIZO J~ I " I 4 STATE PARK ' ~ ‘ “ " ’1 E i IMPAng AREA 6" QARR‘ZO' _ .f f \ Q .._, / <. 2 ,4 SI MESA r L kg f 3: ,-4_jMPACT 1 .1 Q ‘ I :7" 31: ,l I 2. 5:5. \L \:. 1 L 3%»: >17. U 5 NAVAL < l 5 \L ,, _ GUNNERY RAP H‘( N . fiflmntuin " Naval. Springs Ana January I, |969 Through I970 FIGURE 51. — Continued. 476-246 0 - 72 - 7 86 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 Superstition Hills and Imperial faults to the south- east than it is an integral part of the Coyote Creek fault. Its history of displacement and seismicity has intriguing similarities to that of the Superstition Hills and the Imperial faults. Except for the lack of comparably large historic earthquakes near it, the Superstition Hills fault has several characteristics in common with the south break. Creep since the 1968 earthquake and the in- ferred creep during the preceding 20 years or so along the south break resemble the intermittent small displacements recorded along the Superstition Hills fault that are evidently related to or triggered by local or distant earthquakes. (See Allen and others, 1965; Allen and others, this volume.) Since 1934, the region common to these two faults has also experienced several earthquakes of magnitude 5—5.6 (see Sharp, “Tectonic Setting of the Salton Trough,” this volume, fig. 4), at least one of which was appar- ently associated with small displacements on the Superstition Hills fault. Although neither the south break nor the Superstition Hills fault appeared to move during the latest nearby moderate earthquake (M253 in Superstition Hills in September 1971; see Allen and others, this volume), it seems reason- able to suppose that the south break may have moved during one or more of the earlier moderate earth- quakes, some of which were closer than that of 1971. Historic behavior of the Imperial fault perhaps comes closer to that observed and deduced for the south break. In 1940, there was a swarm of major earthquakes along the Imperial fault (Ulrich, 1941; Trifunac and Brune, 1970) and associated large dis- placements, followed by later displacements, includ- ing delayed creep, that were apparently associated with small or distant earthquakes (Brune and Allen, 1967; Allen and others, this volume). Although both faults combine major earthquakes with displacement at other times, their behavior certainly is not iden- tical. The earthquake of September 1971, for ex- ample, was roughly equidistant between the faults, but displacement then occurred only on the Imperial fault. REFERENCES CITED Allen, C. R., St. Amand, Pierre, Richter, C. E., and Nord— quist, J. M., 1965, Relationship between seismicity and geologic structure in the southern California region: Seismol. Soc. America Bull., v. 55, no. 4, p. 753—797. Allen, C. R., Grantz, Arthur, Brune, J. N., Clark, M. M., Sharp, R. V., Theodore, T. G., Wolfe, E. W., and Wyss, Max, 1968, The Borrego Mountain California earthquake of 9 April 1968—a preliminary report: Seismol. Soc. America Bull., v. 58, no. 3, p. 1183—1186. Brown, R. D., Jr., Vedder, J. G., Wallace, R. E., Roth, E. R., Yerkes, R. F., Castle, R. 0., Waananen, A. 0., Page, R. W., and Eaton, J. P., 1967, The Parkfield-Cholame California, earthquakes of June-August 1966—surface geologic effects, water resources aspects, and preliminary seismic data: U.S. Geol. Survey Prof. Paper 579, 66 p. Brown, R. D., Jr., and Wolfe, E. W., 1970, Map showing recently active breaks along the San Andreas fault between Point Delgada and Bolinas Bay, California: U.S. Geol. Survey open-file map. Brune, J. R., and Allen, C. R., 1967, A low stress-drop, low- magnitude earthquake with surface faulting—The Im- perial California earthquake of March 4, 1966: Seismol. Soc. America Bull., v. 57, no. 3, p. 501—514. Chinnery, M. A., 1966, Secondary faulting; I and II: Cana- dian Jour. Earth Sciences, v. 3, p. 163—190. Clark, M. M., 1971, Map showing recently active breaks along the Garlock and associated faults, California: U.S. Geol. Survey open-file map. Dibblee, T. W., Jr., 1954, Geology of the Imperial Valley region, in Jahns, R. H., ed., Geology of southern Califor- nia; chap. 2, Geology of the natural provinces: California Div. Mines and Geol. Bull. 170, p. 21—28. Environmental Data Service, 1968—71, Climatological data, California: Natl. Oceanog. and Atmospheric Adm., v. 72—75. Hamilton, R. M., 1970, Time-term analysis of explosion data from the vicinity of the Borrego Mountain, California, earthquake of 9 April 1968: Seismol. Soc. America Bull., v. 60, no. 2, p. 367—381. Hely, A. G., and Peck, E. L., 1964, Precipitation, runofl’, and water loss in the lower Colorado River—Salton Sea area: U.S. Geol. Survey Prof. Paper 486—B, 16 p. Kearney, T. H., Peebles, R. H., and others, 1960, Arizona flora: Berkeley, Univ. California Press, 1085 p. Ross, D. C., 1969, Map showing recently active breaks along the San Andreas fault between Tejon Pass and Cajon Pass, southern California: U.S. Geol. Survey Misc. Geol. Inv. Map I—553. Sharp, R. V., 1970, Map showing recently active breaks along the San Jacinto fault zone between San Bernardino and Borrego Valley, California: U.S. Geol. Survey open- file map. Smith, S. W., and Wyss, Max, 1968, Displacement on the San Andreas fault subsequent to the 1966 Parkfield earthquake: Seismol. Soc. America Bull., v. 58, p. 1955— 1973. Tchalenko, J. S., 1970, Similarities between shear zones of different magnitudes: Geol. Soc. America Bull., v. 81, p. 1625—1640. Tchalenko, J. S., and Ambraseys, N. N., 1970, Structural analysis of the Dasht-e Bayaz (Iran) earthquake frac- tures: Geol. Soc. America Bull., v. 81, p. 41—60. Trifunac, M. 0., and Brune, J. N., 1970, Complexity of energy release during the Imperial Valley, California earth- quake of 1940: Seismol. Soc. America Bull., v. 60, p. 137—160. Ulrich, F. P., 1941,. The Imperial Valley earthquakes of 1940: Seismol. Soc. America Bull., v. 31, p. 13—31. Vedder, J. G., and Wallace, R. E., 1970, Map showing recent- ly active breaks along the San Andreas and related faults between Cholame Valley and Tejon Pass, Cali- fornia: U.S. Geol. Survey Misc. Geol. Inv. Map I—574. Youd, T. L., and Castle, R. 0., 1970, Borrego Mountain earth- quake of April 8, 1968: Am. Soc. Civil Engineers Proc., Jour. Soil Mechanics and Found. Div., v. 96, no. SM4, p. 1201—1219. DISPLACEMENTS ON THE IMPERIAL, SUPERSTITION HILLS, AND SAN ANDREAS FAULTS TRIGGERED BY THE BORREGO MOUNTAIN EARTHQUAKE1 By CLARENCE R. ALLEN, SEISMOLOGICAL LABORATORY, CALIFORNIA INSTITUTE OF TECHNOLOGY, MAx WYSS, LAMONT—DOIIERTY GEOLOGICAL OBSERVATORY OF COLUMBIA UNIVERSITY, . jAMES N. BRUNE, INSTITUTE OF GEOPHYSICS AND PLANETARY PHYSICS, UNIVERSITY OF CALIFORNIA, SAN DIEGO, AND ARTHUR GRANTZ and ROBERT E. WALLACE, U.S. GEOLOGICAL SURVEY ABSTRACT The Borrego Mountain earthquake of April 9, 1968, trig- gered small but consistent surface displacements on three faults far outside the source area and zone of aftershock activity. Right—lateral displacement of 1—21/2 cm occurred along 22, 23, and 30 km of the Imperial, Superstition Hills, and San Andreas (Banning—Mission Creek) faults, respec- tively, at distances of 70, 45, and 50 km from the epicenter. Although these displacements were not noticed until 4 days after the earthquake, their association with the earthquake is suggested by the freshness of the resultant en echelon cracks at that time and by the absence of creep along most of these faults during the year before or the year after the event. Dynamic strain associated with the shaking is a more likely cause of the distant displacements than is the static strain associated with the faulting at Borrego Moun- tain because (1) the dynamic strain was much larger and (2) the static strain at the San Andreas fault was in the wrong sense for the observed displacement. The principal surface displacements on the Imperial fault took place with- in 4 days of the earthquake and may have occurred simul- taneously with the passage of the seismic waves, but the possibility of delayed propagation to the surface is indicated by a 1971 event on the Imperial fault in which the surface displacement followed the triggering earthquake by 3—6 days. All three of the distant faults are “active” in that they show evidence of repeated Quaternary movement, and surface dis- placements occurred only along those segments where the fault trace is well delineated in surface exposures, at least in uncultivated areas. This is the first documented example of fault displacement triggered by seismic shaking far from the source area, although such displacement has probably gone undetected many previous times here and in similar tectonic environmments. This phenomenon forces us to be much more conservative in estimating the probabilities of damage from surface displacements along active faults in seismic regions. 1Contribution 1833, California Institute of Technology, Division of Geologi- cal and Planetary Sciences, Pasadena, Calif. INTRODUCTION The Borrego Mountain earthquake of April 9, 1968 (magnitude 6.4) was associated not only with a con- spicuous surface break in its source region along the Coyote Creek fault (Clark, “Surface Rupture Along the Coyote Creek Fault,” this volume), but also with displacements far outside the epicentral region along three major faults in the Imperial Valley region to the east and southeast of the epicenter (fig. 52). The Imperial, Superstition Hills, and San Andreas2 faults broke along segments at least 22, 23, and 30 km long, respectively, at distances of 70, 45, and 50 km from the epicenter. Remeasurements of several small-scale geodetic networks that had been established before the earthquake, as well as observations of en echelon cracking, showed that right-lateral displace- ments of 1—2% cm had occurred on these three dis- tant faults. This is the first documented case of an earthquake apparently causing fault displacements well outside the epicentral region. A similar phe- nomenon may have occurred along a segment of the Garlock fault as a result of the 1952 Kern County earthquake on the White Wolf fault (Buwalda and St. Amand, 1955, p. 53), but the Garlock fault is relatively close to the White Wolf fault and was almost Within the zone of aftershock activity. This paper presents evidence that the displace- ments on the Imperial, Superstition Hills, and San 2The branch of the San Andreas fault system northeast of the Salton Sea has sometimes been called the Banning—Mission Creek fault because it repre- sents the combined Banning and Mission Creek faults southeast of their point of coalescence near Indio and because of this fault’s debatable continuity with the San Andreas fault farther north. The name San Andreas is used in this report for the sake of brevity and in keeping with usage by Dibblee (1954) and Hope (1969). 87 88 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 "5‘45 Ind-30' ”5‘15- I Il6‘l5' Area of 33°30- 33°|5'- Area of figure 54 El Centro UN |TED_ _ _s_1;A_T_E_S_ ————————————— _____________ MEXlCO l | l l l FIGURE 52.—-Index map showing relation of the Coyote Creek fault (locus of the 1968 Borrego Mountain earth- quake) to the three distant faults, the San Andreas, Super- stition Hills, and Imperial, upon which triggered movements occurred. Heavy lines represent approximate segments which broke at that time. Andreas faults were in fact triggered by the seismic shaking of the distant Borrego Mountain earthquake, that these displacements were not associated with normal aftershocks, and that they were not caused by the change in the regional static strain field that resulted from the fault displacements of the Borrego Mountain earthquake. ACKNOWLEDGMENTS Participation in this study by the California Insti- tute of Technology was supported by National Sci- ence Foundation Grants GA—1087 and GA—12863. Mr. James Hileman assisted in the interpretation of the data from the small geodetic networks, Dr. Wal- ter Arabasz studied the microearthquakes that followed the main shock, and Dr. R. D. Nason first observed the cracks at Highway 80 that stimulated the subsequent investigations. The authors appre- ciate the critical comments of R. V. Sharp. OBSERVATIONS IMPERIAL FAULT Although a number of auxiliary faults near Bor- rego Mountain were examined for possible surface displacements on the day after the April 9 earth- quake, the Imperial fault, 70 km distant, was not visited until April 13. At that time, Wyss and R. D. Nason noticed fresh en echelon cracks along the trace of the fault at Highway 803 (fig. 53) suggest- ing at least one-third centimeter of right-lateral displacement. It was this discovery that then stimu- lated the careful examination of other distant faults and led to the subsequent documentation of surface displacements on the Superstition Hills and San Andreas faults, as well as at other localities along the Imperial fault. Because of the unusual fault displacement along the Imperial fault in March 1966 (Brune and Allen, 1967a) and the suspicion that creep might be occur- ring along this and related faults, Brune and Allen in 1966 and 1967 established a series of small geo- . W” FIGURE 53.—Fresh cracks in unconsolidated material (be- neath ruler) at north edge of Highway 80. Older tar-filled cracks in pavement resulted from earlier movements at this locality. (See Brune and Allen, 1967a.) Photograph by R. D. Nason, April 13, 1968. 3Since completion of the adjacent freeway in 1971, the former Highway 80 is now known as Evan Hewes Highway, Imperial County route SOS. THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 89 detic networks that straddled the Imperial, Supersti- tion Hills, and San Andreas faults. These networks, each comprising a single theodolite station and 5—10 markers within a few hundred meters on both sides of the fault trace, were patterned on the similar net- works that had been established earlier across the San Andreas fault near Parkfield (Smith and Wyss, 1968). The locations of these stations are shown in figure 54 and on plate 2, and their coordinates are given in table 12. TABLE 12.—Caltech fault-crossing geodetic networks in the Imperial Valley area _ Number of Station Fault Lat (N.) Long (w.) obsfgzfiion ggfifsr‘t’g‘ 1—1—71 All-American Imperial ............ 32°40.55' 115°21.4S’ 5—10—67 9 Canal. Baileys Well ..... Coyote Creek... 33°06.20’ 116°03.63’ 3—31—69 3 Bertram ............. San Andreas ..... 33°24.6S’ 115°47.54’ 5— 5-68 6 Harris Road... mperial... 32°53.02’ 115°32.34' 1- 5—68 10 Meloland ................ do ........ 32°48.21' 115°28.01’ 6—19—66 12 North Shore ..... San Andre 33°31.71’ 115°56.21’ 6—20—69 3 0cotillo Wells..Coyote Creek... 33°09.6S’ 116°07.85' 4— 9- 68 12 Red Canyon ..... San Andreas... 33°37.62' 116°02.9l' 5—11-67 8 Superstition Superstition 32°55.60’ 115°41.84’ 5— 7—67 9 Hills. Hills. Wfirthington Imperial ............ 32°50.85’ 115°30.69’ 5—10-67 11 oad. Figure 550 shows that about 2 cm of right-lateral displacement took place along the Imperial fault at Highway 80 between January 5, 1968, and April 19, 1968, and evidence is presented in a later section to indicate that this displacement took place at about the time of the Borrego Mountain earthquake and not as creep distributed throughout the 3-month interval. Three additional geodetic networks had been established across the Imperial fault prior to the earthquake (figs. 54, 55). The southernmost net- work, at the All-American Canal, showed no obvious displacement during the 3-year period shown in figure 55. The northernmost station, at Harris Road (fig. 55A), showed a clear displacement of about 0.8 cm at about the time of the earthquake. However, the intervening network at Worthington Road (fig. 558) showed a displacement of about 1.1 cm between May 1 and December 29, 1967 (the year before the earthquake) and no displacement at the time of the earthquake. We feel that these findings are consis- tent with our argument that the Borrego Mountain earthquake triggered the release of elastic strain along most of the fault; if the strain had already been relieved shortly prior to that time, as it was near Worthington Road, then no further displace- ment took place. The geodetic observations were supported by field evidence of intermittent surface faulting along more than 22 km of the Imperial fault, extending distinctly farther both to the north and to the south than the 10-km segment broken during the 1966 Imperial earthquake (Brune and Allen, 1967a). Clear en eche- lon cracks were observed at several localities from Harris Road on the north to Heber Road on the south (figs. 54, 56—58), although they could not be followed continuously throughout the intervening area because of extensive and continuing cultivation; cracks were not present near Worthington Road, where geodetic data indicated no movement. Fresh cracks in soil were generally observed only on the shoulders of roads between cultivated fields. En eche- lon cracks showed up particularly well in many asphalt roads that were crossed by the fault (fig. 56), although in some places it was difficult to dis- tinguish cracks still existing from the 1966 earth- quake from further cracking caused by the 1968 event (fig. 57). The absence of obvious cracking at some localities suggests that not everywhere did the fracture reach the surface as a discrete plane, which is not surprising in view of the small displacement and the known discontinuous nature of the much larger 1940 displacement in some localities (fig. 54). It is significant however, that wherever fresh cracks were observed, they followed precisely the trace of the 1940 break, which in some areas is well docu- mented to within 1 m (J. P. Buwalda, unpub. data). Although the full extent of the 1968 trace is not known because of agricultural developments at the north end and sand dunes at the south end, careful examination revealed no fresh cracks where the Imperial fault crosses Keystone Road on the north and Highway 98 on the south (fig. 54), thus giving a total length of between 22 and 30 km. For more than 3 years after the Borrego Mountain earthquake, no further slip on the Imperial fault at Highway 80 was observed visually or indicated by surveying records. On September 30, 1971, however, a further displacement was observed after an earth- quake of magnitude 5.3 in the Superstition Hills, 37 km northwest of the Highway 80 locality; this earthquake represents the heaviest shaking the region had experienced since the Borrego Mountain event of 1968. Six days after the shock, right-lateral displacements of about 1% cm were clearly visible in the pavement of Highway 80, and a resurvey of the geodetic network on October 13 showed a dis- placement at this time of 1.4 cm (calculated on the same basis as that used for the earlier measure— ments shown in fig. 550). As discussed in a later section, however, there is some evidence that the observed displacement did not take place exactly at the time of the earthquake, but between 3 and 6 days later. The Imperial fault was first noticed and named at the time of the 1940 El Centro earthquake, but a history of repeated Quaternary displacements along the fault is indicated by (1) a conspicuous scarp as THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 Baselrom U S Geological Survey 1 62 500 Brawley, Holmlle. and Calexlco, 1957 2 32°40‘ 5 MILES \ ' \ I 8 KILOMETERS\\ \ FIGURE 54. — Map of Imperial fault trace based primarily on aerial photographs taken shortly after the 1940 El Centro earthquake and on the unpublished 1940 field notes of J. P. Buwalda. Triangles show locations of Caltech fault-cross- ing geodetic networks that were surveyed before and after the Borrego Mountain earthquake of April 9, 1968. Solid circles show localities where fresh cracks were observed in loose soil on April 19—28, 1968, presumably resulting from triggered fault displacement on April 9. Open circles are localities where freshness or significance of cracks was questionable. Trace is dashed where projected across areas where surficial evidence of faulting (scarps, ground-water barriers, or lineaments on aerial photographs) is indistinct or absent. See figure 52 for location. THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 91 U! i l l l l l t it l O N RIght~loteral dISplocement. cm l T T 1 l @ Imperial fault at Harris Rd A a 1' l l Imperlal fault at Worthington Rd. © Imperial fault of Highway 80 l l 1 L l l l t l—o—— r—o——a i e ? , l l 1 CE) Superstition HIHS fault at Imler Rd, @San Andreas fault at Red Canyon I | -67 l-l-68 Borrego Mtn, earthquake FIGURE 55.—Measurements of fault-crossing geodetic net- works between May 1, 1967, and May 1, 1970, showing displacements at time of Borrego Mountain earthquake. Localities are shown in figure 54 and on plate 2. Circles represent averages of measurements of 2—4 independent lines crossing the fault at each locality, and error bars in- dicate twice the standard deviation of repeated measure- much as 7 m high that delineates the northern 11 km of the fault trace, (2) a marked ground-water barrier that is sometimes visible from the air even in cultivated fields, and (3) a pronounced gravity anomaly along the fault trace (Kovach and others, 1962). However, the total extent of the Imperial fault beyond the 60—km segment broken in 1940 is not known; there is as yet no good geological or geophysical evidence of the fault north of Keystone Road, where the 1940 trace died out, or south of a point near Tortuosa Check (Mexico), at the south end of the 1940 break. Possibly, the Imperial fault is a transform fault whose active segments indeed do not extend much beyond its presently known length (Lomnitz and others, 1970). -69 l-I-70 ments of individual lines in the surveys following the earthquake. Survey lines average 90 m long. Triangles in diagram E show readings of taut-wire creepmeter located 31/2 km north of Imler Road, showing lack of further dis- placement there after installation of creepmeter on May 7, 1968. SUPERSTITION HILLS FAULT On May 11, 1967, Allen and Brune had established a small geodetic network across the Superstition Hills fault where it crosses Imler Road (pl. 2) ; reoccupation of this station on April 19, 1968, 10 days after the Borrego Mountain earthquake, re- vealed about 21/; cm of right-lateral displacement (fig. 55E). At the same time, fresh en echelon cracks showing as much as 1.5 cm of right-lateral displace- ment were discovered along the Quaternary faulty, trace in the same vicinity (figs. 59, 60). On April 25, these cracks were followed northwest for about 8 km along the fault trace, and on May 13—15, Grantz and Wyss mapped the entire broken zone, which extended for 23 km (pl. 2). 92 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 FIGURE 56,—Fresh en echelon cracks crossing Ross Road. View southeast. Photograph taken April 28, 1968. The cracks, as mapped on May 13—15, varied from a single, narrow well—defined break several centime— ters to a meter wide, the most common type, to zones of en echelon cracks as much as 7—10 m wide in which the individual cracks stepped to the left in the manner of ground cracking along right-lateral strike-slip faults. Their dip, as recorded at four near-vertical exposures, were 80°—90°. The cracks followed very closely the trace of the Superstition Hills fault across the strongly folded Pleistocene sedimentary rocks of the Superstition Hills and vi— cinity. Commonly, they lay within the lfli—l QA-m-wide gouge zone of the Quaternary Superstition Hills fault, but in a number of places, the cracks occurred in the Pleistocene sediments as much as 7 In from the Quaternary fault. Where they departed from the Quaternary fault gouge zone, the cracks usually, but not always, lay northeast of it. Where the cracks formed a narrow well-defined zone, they commonly coincided exactly with a single FIGURE 57. —En echelon cracks on Meloland Road first ob- served following the Imperial earthquake of 1966 (Brune and Allen, 1967a) but slightly widened during the Borrego Mountain earthquake. Fault displacement of about one- half m occurred at this same locality in 1940. Photograph taken April 28, 1968. fault-controlled line of desert shrubs (fig. 61), small drainage sumps and collapse pits or trenches (fig. 62), local ground-water barriers, small fault scarps in soft sediments, or, at a few places, right-laterally offset small ridges and gullies. This alinement and the character of the alined features indicate that the fault cracks of April 1968 follow a well-established line of latest Holocene faulting as well as a Quater- nary bedrock fault. The displacements measured on May 13—15 along the partly wind—eroded and sand-filled fault cracks indicated dominantly right-lateral strike slip, with only local vertical displacement and no instances of left-lateral slip. The right slip was as much as 1.8 cm but averaged about 0.8 or 0.9 cm. Of only possible significance, because of the erosion and filling of the THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 93 FIGURE 58. —Fresh cracks across Heber Road. Displacement of about 1% m took place here in 1940, but no further displacements are known to have occurred here until 1968. Photograph taken April 28, 1968. cracks, is the fact that most of the largest right-lat- eral displacements were recorded near, although not at, the ends of the fault break of April 1968. Most displacements of 11/4, cm and greater occurred be- tween 1.7 and 5.8 km of the southeast end and between 2.8 and 4.7 km of the northwest end of the 23-km-long fault break. Displacements of between 0.4 and 1.0 cm were common along the central part of the break. Vertical displacements, along three short (@75 m) segments of the fault, were as much as 21/43 cm. Two, at the northwest end of the break, were up on the northeast; one, near the middle of the break, was up on the southwest. In all three places, the uplifted side is also the topographically higher side, and in two of them the uplifted side holds up a low scarp cut in soft Pleistocene sediments. These relations indicate that the local vertical components FIGURE 59. — Right-lateral displacement of about 2 cm on the Superstition Hills fault in sec. 23, R. 12 E., T. 14 S. Diameter of coin is 1.7 cm. Photograph taken April 22, 1968; note that crack has filled in since time of probable formation on April 9. of slip in April 1968 acted in the same sense as the latest Holocene displacements at those places. When the cracks were first observed on April 19, they were relatively fresh appearing, but by the time mapping was completed on May 15, windblown sand had already obscured much of the fault trace. We conclude that the cracks could not have come into existence long before April 19, and their origin in association with the Borrego Mountain earthquake on April 9 seems highly probable. Because of suspi- cion that creep might be taking place on this fault following the earthquake, a creepmeter was installed across the fault on May 7 at a locality midway along its length (pl. 2). A recorder registered the displace- ment as measured by a taut 10-m Invar wire, similar to an instrument previously used at Parkfield (Smith and Wyss, 1968). Seventeen subsequent readings of 94 FIGURE 60.—En echelon cracks indicative of right-lateral displacement along Superstition Hills fault in sec. 9, R. 12 E., T. 14 S. Photograph taken April 27, 1968. the creepmeter, through November 1971, revealed no evidence of further significant movement on this segment of the Superstition Hills fault, nor have any further cracks appeared in the nearby paved road where it is crossed by the fault. These observations further support the inference that the displacement observed on April 19, 1968, occurred relatively sud- denly in association with the Borrego Mountain earthquake. Minor displacements on the Superstition Hills fault similar to those observed in 1968 have occurred at least three other times in recent years. In 1951, Joseph Ernst (written commun.) noted fresh en echelon cracks along about 3 km of the Superstition Hills fault, approximately centered within the seg- ment that broke in 1968 and along exactly the same trace. A magnitude 5.6 earthquake had been centered in this area about 2 weeks earlier, and Ernst con- cluded that the rate at which windblown sand was THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 FIGURE 61.——New en echelon cracks of right-lateral habit following the line of desert plants that marks the Super- stition Hills fault in NW% sec. 26, T. 13 S., R. 11 E. The larger plants are about 1 foot high. Photograph taken May 13, 1968. filling the cracks demanded their origin within about this period. Similarly, in late December of 1965, Brune noticed fresh en echelon cracks along the Superstition Hills fault at Imler Road, and he and Allen subsequently followed these for about 1 km north and south of the road; these cracks may have been triggered by a nearby magnitude 4.0 shock on November 30, 1965. The cracks were quickly ob- scured by blowing sand, and it is clear from repeated subsequent visits to the area that significant creep is not occurring continuously. Similarly, fresh cracks were noted along the fault south of Imler Road dur- ing a visit to the area on December 20, 1969, al- though they could not be correlated with a specific local earthquake. The movement did not extend far enough north to affect the creepmeter. The move- ment in April 1968 is evidently only one in a series of small episodic surface displacements that have char- acterized this fault in recent years; the same be- havior may well be true of the segments of the THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 95 FIGURE 62. —- Small collapse pits, alined vegetation, and varia- tion in abundance of vegetation along the Superstition Hills fault in SW14 sec. 9, T. 14 S., R. 12 E. New hairline cracks follow the alined features but are largely obscured by windblown sand and silt. Photograph taken May 14, 1968. Imperial fault and San Andreas fault that broke at about the same time. Like the Imperial fault, little is known about the possible extent, if any, of the Superstition Hills fault beyond the segment broken in 1968. On the south- east, the 1968 fractures ended about 1 km north of Edgar Road, at very nearly the same point that the Quaternary trace disappears as observed on aerial photographs and in the field. Likewise, the fractures continued northwest only about as far as the mapped trace of the Quaternary fault (Dibblee, 1954; unpub. data). The Superstition Hills fault, together with the nearby Superstition Mountain fault, appear to be branches of the San Jacinto fault zone, and if pro- jected still farther northwest, they would join the Coyote Creek fault—also a branch of the San Ja- cinto zone— on which the Borrego Mountain earth- quake occurred. SAN ANDREAS FAULT In February 1967, A. G. Sylvester (oral commun.) found fresh cracks along the trace of the San An- dreas fault in the Mecca Hills north of the Salton Sea; they were particularly evident in the 4-km segment between Painted Canyon and the unnamed canyon (locally called Red Canyon) in sec. 28, T. 6 S., R. 9 E. (Thermal Canyon quadrangle). We were not convinced at that time that the cracks necessarily reflected tectonic movements, because among other reasons there has been very little historic seismicity (Allen and others, 1965) or microearthquake activity (Brune and Allen, 1967b) along this segment of the San Andreas fault zone. However, to check this possibility, we established a small geodetic network across the fault in the unnamed canyon noted (sta- tion “Red Canyon” in table 12; pl. 2) on May 11, 1967. Resurvey of this network on April 24, 1968, indicated 1.3 cm of right-lateral displacement, and since that time there has been no significant addi- tional change (fig. 55D). A second fault-crossing geodetic network was established across the fault near Bertram (station “Bertram” in table 12; pl. 2) on May 5, 1968, and it likewise has shown no signif- icant change in six subsequent surveys through August 7, 1970. Despite the low seismicity along this segment of the San Andreas fault, the fault trace is so clearly marked in the field that displacements along it must have occurred in very recent years. Small scarplets as much as 50 cm high and only slightly eroded are abundant along the fault (fig. 63). Zones of en eche- lon fractures, eroded and distinctly older‘ than the recent fractures (fig. 64) , appear to be too transient to have persisted for much more than a decade; conceivably, all the erosion along the older fractures could have been accomplished hb'y‘one heavy...rain- storm. Similarly, piping of rainwater into the frac- ture zone in places appears very recent. From this evidence, it would appear that more than one episode of recent fault displacement has occurred in this area despite the scarcity of historic seismic activity. Whether such previous displacements likewise ac- companied distant earthquakes is unknown. When this area was first visited after the Borrego Mountain earthquake on April 24, fresh en echelon cracks were observed at the base of the scarplets (fig. 65) and at several other localities along the fault trace. Wallace and Wyss subsequently mapped the fresh break for more than 30 km from near Bertram on the south to Thermal Canyon on the north (pl. 2), although it is significant that surface fracturing was by no means continuous throughout the 30-km segment. The average right-lateral dis- 96 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 FIGURE 63.—En echelon cracks (in shadow, foreground) at base of scarplet along San Andreas fault in sec. 28, R. 9 E., T. 6 S. View northwest. Geodetic data at nearby station “Red Canyon” indicated about 1.3 cm right-lateral displacement (fig. 55D). Photograph taken April 28, 1968. placement was estimated to be between 0.5 and 1.0 cm. One of the puzzling features of the San Andreas fault in this region is its apparent termination oppo- site the south end of the Salton Sea, particularly because this segment of the fault may have as much as 260 km of post-early-Miocene strike-slip displace- ment (Crowell, 1962). The apparent termination might be explained simply by concealment of the fault trace southeast of this point by Quaternary sediments, including those of Lake Cahuilla that inundated all this region for 1,300 years, ending perhaps as recently as 300 years ago (Lake LeConte in Hubbs and others, 1960). Likewise, there is some geophysical evidence suggesting continuity of the fault farther southeast (Kovach and others, 1962; FIGURE 64.—New fracture in old fracture zone along San Andreas fault in sec. 28, R. 9 E., T. 6 S. New en echelon fractures are subparallel to old en echelon fractures. Note that erosion has progressed along old fractures. Photograph taken May 9, 1965. S. Biehler, unpub. data). On the other hand, the San Andreas fault may be a transform fault whose active segment does indeed terminate near the Salton Sea. Adding weight to such a hypothesis is the presence of the northeast-trending Salton volcanic domes that might be a manifestation of a ridge segment in a transform fault system (Lomnitz and others, 1970). Nonetheless, it is significant that the 1968 fractures on the San Andreas fault extended southeast to almost exactly the same point where the continuous Quaternary trace disappears (Hope, 1969). DESCRIPTION OF THE SURFACE DISPLACEMENTS En echelon cracking oriented so as to clearly in- dicate right-lateral displacement was the typical style of surface rupture along the Imperial, Super- stition Hills, and San Andreas faults (figs. 65—67). Individual cracks rarely gaped more than 1 or 2 mm when fresh. Individual en echelon breaks were typi- THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 97 FIGURE 65. — En echelon cracks along the San Andreas fault near Salt Creek, in the center of sec. 28, T. 8 S., R. 11 E. View northwest. The fault here brings rocks of the Palm Spring Formation (right) into contact with those of the Borrego Formation (left), and the contact is visible in line with the cracks in the middle distance. Photograph taken April 28, 1968. cally less than 1 m long where the zone of cracks was several centimeters to less than 1 m wide, but they ranged in length from 2 to 30 m in those less common places where the fractured zone was 1—7 m or more wide. The fractures showed up not only in undisturbed soil but also in asphalt roads that were crossed by each of the three faults. All the broken sections of the Superstition Hills and San Andreas faults were traversed in their entirety by one or more of the authors. This was not possible along the Imperial fault because of intensive cultivation of most of the area; instead, each road and canal cross- ing was checked. The newly formed fractures along the three dis- tant faults were such minor features that they would easily have escaped detection if we had not specifical- ly looked for them and if we had not known from FIGURE 66.——En echelon fractures along the San Andreas fault approximately 2 km south of Salt Creek, in SW14 sec. 34, T. 8 S., R. 11 E. Photograph taken May 10, 1968. FIGURE 67.———En echelon fractures, sec. 28, T. 6 S., R. 9 E. Scale is 19 cm long. View southeast. Photograph taken May 9, 1968. other geologic evidence the exact locations of the active fault traces to within a very few meters. The precise trace of the Imperial fault, despite its loca- tion within the heavily cultivated floor of the Impe- rial Valley, was known from J. P. Buwalda’s very detailed notes on the much larger 1940 earthquake displacements, as well as from the subsequent dis- placements along part of the trace in 1966 (Brune and Allen, 1967a). The detailed trace of the Super- stition Hills fault is clear on large-scale aerial photo- 98 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 graphs and on the ground because of the abrupt truncation of well-exposed and highly deformed Pleistocene sedimentary rocks at the fault. Similarly, segments of the San Andreas fault are well deline- ated on detailed aerial photographs because of ground-water damming and the truncation of young sedimentary rocks (figs. 68, 69) (Hope, 1969). Both the Superstition Hills and San Andreas faults traverse desert areas that, for the most part, have never been cultivated or otherwise culturally modified. Considerable stretches of both faults are mantled with a veneer of sediments dating from the last major filling of the Salton depression by Lake Cahuilla between about 300 and 1,600 years ago. Seldom have such well-defined faults been exam- ined in such great detail after a major nearby earthquake, and although this may be the first docu- mentation of fault displacements caused by seismic shaking, the same phenomenon may have happened many times before, not only here, but also on other active faults in similar tectonic environments. It is particularly significant that, with minor ex— ceptions, surface fracturing occurred in 1968 only along those segments of the Superstition Hills and San Andreas faults where examination of aerial photographs and field exposures could clearly delin- eate preexisting active breaks. Several examples have already been cited in the preceding section, but perhaps the most intriguing localities are along the San Andreas fault northeast of the Salton Sea. In this area, some segments of the fault are much more clearly delineated by features of recent displacement than others, and it is only on the clearly delineated parts that fresh fractures were found after the Bor- rego Mountain earthquake, despite careful searches in some of the intervening “inactive” segments (for example, Box Canyon Wash, Salton Sea State Park headquarters area, Highway 111 northeast of Bom- bay Beach). This distribution is well illustrated by Hope’s (1969) map, which shows both the 1968 fractures and the Quaternary breaks that were vis- ible on 1:14,000-scale aerial photographs flown in 1966. Either (1) contemporary movements are tak- ing place along a discrete fault plane at the surface in some segments and throughout a distributed zone in other segments or (2) contemporary displace- ments are limited to certain weak segments of the fault in contrast to other stronger segments that are temporarily locked, to be broken through during a larger earthquake at some time in the future. The second argument is analogous to that sometimes used for the San Andreas fault as a Whole (for example, Allen, 1968), but whether this kind of reasoning is applicable on such a small scale is unknown. It should also be noted that the geodetically mea- sured displacements (fig. 55) were consistently larger than those estimated from field observations of the faulting. Inasmuch as the geodetic lines used for the calculations of figure 55 were typically about 100 m long, this variation suggests that some shear deformation was taking place that was not expressed as discrete visible fractures at the surface. Indeed, in attempting to interpret the geodetic data at locali- ties where many stations were surveyed, one could not escape the conclusion that, at least at some localities, deformation took place in a distributed fashion rather than entirely along a single fault plane. The accumulation of such distributed deforma- tion during repeated fault displacements is the mech- anism that produces the drag so commonly observed along faults in layered rocks. At the Coyote Creek fault rupture of April 9, 1968, this mechanism pro— duced pronounced drag in late Holocene sedimentary rocks and in places produced more than half the total late Holocene deformation. (See Clark and others, this volume.) OTHER FAULTS At the same time that fresh displacements were being discovered on the Imperial, Superstition Hills, and San Andreas faults, a number of other faults in the region were carefully checked in the field and found to show no evidence of surface displacements. These include the Superstition Mountain fault, Elsi— nore fault, Laguna Salada fault, Earthquake Valley fault, San Felipe fault, and branches of the San J acinto fault system north of Borrego Valley. One feature that distinguishes these faults is that they are predominantly in crystalline rocks, whereas the parts of the three faults that moved are all in deep alluvium or late Cenozoic sedimentary deposits. The estimated minimum distance to crystalline basement based on seismic work (Kovach and others, 1962; Biehler and others, 1964) is 3,500 m along the Superstition Hills fault, 6,000 In along the Imperial fault, and perhaps 2,000 m, but locally only 400 m (Babcock, 1970), along the San Andreas fault. It is also probably significant that the only three faults in southeastern California for which we had some evidence of slippage before the earthquake (and which had therefore been straddled with small geo- detic networks) were the same three faults that moved during the Borrego Mountain earthquake. MECHANISM OF DISPLACEMENT Three lines of evidence lead us to believe that the observed fault displacements took place on or about April 9, 1968: (1) The geodetic measurements that were made during the year before and the year after THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 99 FIGURE 68. — Linear valley eroded along San Andreas fault, Mecca Hills. New fractures followed this valley but could not be found in the dry alluvium of Painted Canyon (far background). the earthquake indicate little or no creep, except at Worthington Road. It thus seems unlikely that creep should have characterized only the short interval that included the earthquake, unless triggered by it. (2) On March 8, 1968, 1 month before the earth- quake, a heavy and unusual rainstorm brought ap- proximately 5 cm of precipitation to the entire Imperial Valley—Coachella Valley area, causing con- siderable runoff and local flooding. All the fresh cracks observed after the earthquake must have originated after this rainstorm. (3) Blowing dust and sand are characteristic of the entire region, and everyone who studied the displacements on these three faults, as well as the main break near Ocotillo Wells, was impressed with the rate at which fresh features disappeared. Within 2 weeks of the earth- quake, many of the cracks along the main break had become barely recognizable because of windblown 100 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9. 1968 FIGURE 69.—-View southeast along San Andreas fault. New fractures closely followed this lineament to point shown by arrow in distance. Bat Caves Buttes on left; Salton Sea in distance; Salt Creek in foreground. sand. It is our judgment, based on field experience in this area, that the fractures first observed be- tween April 13 and April 24 must have come into existence during the first 2 weeks in April. Particu- larly along the Imperial fault, the fresh cracks in powdery alluvium that were first observed on April 13 must have originated within the preceding few days. It seems to us to be a reasonable, indeed a highly likely, conclusion that all the fractures came into existence at the approximate time of the Bor- rego Mountain earthquake on April 9. STATIC VERSUS DYNAMIC STRAIN If the hypothesis is accepted that the breaks on the Imperial, Superstition Hills, and San Andreas faults were caused by the Borrego Mountain earth- quake, the important question still remains as to whether the displacements were caused by the dy- THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 namic strain associated with the shaking or by the static strain associated with the main fault break. The static strain is the permanent strain field caused by the 33-km-long break on the Coyote Creek fault; the dynamic strain is the transitory strain associated with the seismic waves generated by the earthquake. For an estimate of the static strain, we use the diagrams of Press (1965). The length of the surface break is taken as 33 km, and the hypocentral depth now assigned by the Seismological Laboratory, Cali- fornia Institute of Technology, is 11 km. For an upper limit of the static strain at distance, we there- fore use Press’s case in which L:D; the far-field strains were calculated by scaling down Press’s figures to correspond to an average fault displace- ment of 30 cm. The resulting static strains at distances of 45, 70, and 50 km in the directions of the Superstition Hills, Imperial, and San Andreas faults are 4X10‘7, 1X10—7, and ~3><10‘7 respec- tively, assuming these faults to be parallel to the Coyote Creek fault. The minus sign for the San Andreas fault indicates that the residual static strain induced by the Coyote Creek fault displacement was left lateral (Dr. John McGinley, California Inst. Technology, oral commun., 1968). The dynamic strain caused by S waves with a period of approximately 4.3 sec, recorded at El C‘en- tro near the Imperial fault, was about 1.1 ><10‘5 (corresponding to a trace amplitude of 4.9 cm at a period of 4.3 sec on the strong-motion Wood-Ander- son instruments). This dynamic strain is two orders of magnitude larger than the static strain at this distance. This fact, in addition to the persuasive argument that the static strain would have led to the opposite sense of displacement on the San An- dreas fault, strongly indicates that the dynamic (vibratory) strains rather than the static strain induced the observed ruptures on the distant faults. SUDDEN DISPLACEMENT VERSUS CREEP Even granting that the dynamic strains caused the displacements, the question remains as to whether these displacements took place suddenly or during a period of creep lasting several minutes, hours, or days. After the distant displacements were first noticed, geodetic measurements were repeated at closely spaced intervals to determine if creep was perhaps still taking place (fig. 55) ; it appears that within the accuracy of the measurements, no creep was occurring at this time on any of the three dis- tant faults except for one possible increment at Superstition Hills (fig. 55E). The sensitive creep- meter installed subsequently at Superstition Hills further substantiates this conclusion. The observa- tions at Highway 80 indicate that if displacements 476—246 0 - 72 - 8 101 occurred during a period of creep after the earth- quake, this period must have been shorter than 4 days. If any of the three distant breaks had occurred as sudden ruptures, however, seismic waves would have been radiated. Because the seismographic records of most southern California stations were off scale for several minutes following the Borrego Mountain earthquake, we cannot state with assurance that earthquakes did not occur at the three distant locali- ties immediately following the main event, although it seems unlikely that the magnitudes of such events could have exceeded 4.5 without being detected. However, three lines of evidence suggest that such sudden displacements, if they did occur, were not in any sense “normal” earthquakes: 1. The fault lengths of 22, 23, and 30 km are much longer than could typically be associated with earthquakes of magnitude less than 4.5 '(Wyss and Brune, 1968). 2. Only a very few possible aftershocks could be associated with the three distant faults, in sharp contrast to the usual high aftershock activity accompanying fault breaks of this length. Despite a careful search of seismic rec- ords from several stations close to the distant faults, including temporary stations within the Imperial Valley at Obsidian Butte and near Westmoreland, only six small shocks could be found that might possibly have been associated with the Superstitution Hills fault within a month after the earthquake, one small shock that might have been associated with the Im- perial fault, and none in the area of the San Andreas fault. A particular search was made of the Hayfield4 records for events with short S-P intervals, because the San Andreas fault is much closer to this station than to the Coy- ote Creek fault, near which the principal aftershock activity occurred. The absence of aftershocks in the vicinities of the distant faults is substantiated by microearthquake surveys in two of these areas by Walter Ara- basz on April 20—21. Using a backpack instru- ment that recorded on smoked paper and operated at amagnification of 100,000 at 20 cps, he found no nearby microearthquakes during 13 hours of continuous recording at the Im- perial fault near the south end of the fractured segment. Similar but shorter periods of record- ing farther north along the Imperial fault and in a granite quarry at Superstition Mountain (5 km from the Superstition Hills fault) like- ‘For the location of the Hayfield station, see fig. 6. 102 wise revealed little or no microearthquake activity. 3. Another obvious peculiarity of these displace- ments is the unusually low ratio of average offset to length of rupture. For most earth- quakes, when the faulting length is about 20— 30 km, the average offset is about 10—100 cm, whereas the average offset observed here is only 1—21/2 cm. In addition, it appears that the breaks may not have been continuous on the San Andreas and Imperial faults. After the Borrego Mountain earthquake, we con— cluded (1) that the displacements on the three distant faults occurred rapidly, but with a mecha- nism of strain release different from that of typical earthquakes associated with fault breaks of these lengths, (2) that this relatively rapid motion com- menced with the arrival of the first intense seismic energy from the Borrego Mountain earthquake, and (3) that it probably lasted at most only as long as the strong shaking persisted. The displacements were visualized to have been triggered by strong seismic shaking, which is a previously undocumented mechanism of strain release on active faults. These three conclusions must be tempered, how- ever, by subsequent events along the Imperial fault. As was mentioned in an earlier section, there was no further movement along the Imperial fault after the Borrego Mountain earthquake until September 30, 1971, when a magnitude 5.3 shock in the Super— stition Hills, 37 km away, was followed by about 1.4 cm of right-lateral displacement on the Imperial fault at the Highway 80 locality. Fresh cracks were observed along a segment of the fault at least 10 km long extending from south of Ross Road to north of Robinson Road. But evidence suggests that the fault- ing did not occur at the same time as the earthquake: 3 days after the earthquake, Brian Tucker (oral commun., 1971) examined the Highway 80 locality and noticed no fresh cracks despite his familiarity with the locality and its history. However, 3 days later, on October 6, 1971, fresh cracks and lateral offsets were obvious to several visiting parties. Re- peated subsequent visits to the site revealed no increase in displacement, so we must conclude that virtually all the surface faulting took place between 3 and 6 days after the earthquake. One might argue, of course, that the near coincidence of the two events was fortuitous, but this seems unlikely in view of the two earlier movement episodes, both of which (1966, 1968) also occurred in close association with large local shocks. As for the Borrego Mountain earthquake, we are forced to conclude that this later faulting was triggered by the dynamic waves of the THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 nearby shock, but the mechanism of local strain release and its apparent delay remain problematical. Perhaps the Imperial fault was triggered only at depth at the time of the September 30 earthquake, and this dislocation then extended surfaceward only gradually, something like the situation visualized by Smith and Wyss (1968) for the Parkfield dis- placements and by Burford (this volume) for the Coyote Creek fault displacements. We must now recognize that the same sort of delay may have characterized the 1968 displacement as well; the ac- tual triggering may have occurred at a depth of perhaps 4 km in the elastically strained sedimentary rocks, with the displacement then propagating to the surface within the following 4 days. Only contin- uously recording creepmeters will resolve this prob- lem in future events. Many additional aspects of the mechanics of dis- placements on the three distant faults remain unex- plained. We assume that elastic strain was released by the displacements, but the depth at which this elastic strain had accumulated is problematical. An attractive but unproved hypothesis is that creep is taking place continually in the crystalline basement rocks along these three faults, partly reflecting the unusual semioceanic crust and complex fault pattern of the region (Allen, 1968). Elastic strain is Visual- ized to accumulate only in the overlying thick section of indurated sedimentary rocks, to be relieved inter- mittently either by episodic creep, by very shallow small earthquakes such as the Imperial earthquake of 1966 (Brune and Allen, 1967a), or by occasional intermediate-sized shocks such as the El Centro earthquake of 1940 on the Imperial fault. Such events may in turn be triggered by externally caused shaking. What determined the amount of displacement on the three distant faults is another unanswered ques- tion. Scholz, Wyss, and Smith (1969) argued that the amount of episodic creep displacement is not a function of the nature of the instigating event, but instead is related to the stress-drop between a con- stant rupture stress and a constant frictional stress on the fault. Had the tectonic stress accumulated to a critical value, the creep presumably would have started even without instigation by the earthquake, as presumably happened at Worthington Road before the earthquake (fig. 553). On the other hand, field evidence may suggest that the total displacements on the three distant faults might have been at least partly a function of the strength of shaking; the larger displacement on the Superstition Hills fault as compared to the Imperial fault may be an indica- tion not of higher stress accumulation but, at least in part, of stronger shaking closer to the source. THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 STRESS-DROPS In order to estimate the stress-drops associated with the displacements on the distant faults, a fault depth must be assumed. We arbitrarily assume a depth of 4 km, equal to abOut half the thickness of the sedimentary section in the center of the Imperial Valley (Biéhler and others, 1964) and corresponding to the depth of transition between stable sliding and stick-slip in the Parkfield model of Scholz, Wyss, and Smith (1969). If the average displacement is assumed to be 1.5 cm, the stress—drop is 0.5 bar, close to (that is one-half of) the value obtained by Brune and Allen (1967a) for the Imperial earthquake of 1966. TRIGGERING We have chosen to use the word “trigger” in con— nection with the displacements on the Imperial, Superstition Hills, and San Andreas faults in recog- nition of the fact that one event initiated other events. However, it must be recognized in these examples that the displacements were initiated by a much larger event than the displacements them- selves. As was indicated in a previous section, the maximum dynamic strain at El Centro was about 1.5x10‘5, and assuming a fault depth of 4 km and an average displacement of 1.5 cm on the nearby Im- perial fault, the calculated strain associated with the displacement was only 2.5X10'6. In contrast, a different type of seismic triggering might be the multiple ruptures of the Alaska earthquake, where small events apparently triggered larger events (Wyss and Brune, 1967). ENGINEERING IMPLICATIONS In the planning of engineering structures adjacent to or across active faults, it has usually been assumed in the past that significant fault displacements would occur only infrequently — perhaps about once every few hundred years on even the most active faults. The documentation of semicontinuous creep along parts of the San Andreas fault has tended to modify this kind of thinking (for example, Wallace, 1970) , and the present study emphasizes still further the difficult problems of estimating probabilities of fault displacements. A few years ago, for example, one might have estimated that the San Andreas fault opposite the Salton Sea would break perhaps once every hundred years, the inferred recurrence time for large earthquakes on this segment of the fault. However, if one now assumes, on the basis of the Borrego Mountain experience, that an earthquake of magnitude 6 or greater anywhere in the Imperial Valley area might cause a small displacement on parts of the San Andreas fault, then the estimated chances of fault displacement are considerably higher. The recurrence curve for the Imperial Valley 103 region based on 1934—63 records (Allen and others, 1965) suggests that an earthquake of magnitude 6 or greater should occur about once every 61/2 years. Perhaps the reason that some of the scarplets along the faults look so fresh (fig. 63) is that displace- ments indeed take place with about this frequency. Although it does not necessarily follow that active faults in all parts of California would behave in the same manner as the Imperial, Superstition Hills, and San Andreas faults when heavily shaken, it seems clear, nevertheless, displacements occur more frequently than previously recognized along the myriad of active faults that underlie California and other tectonically similar regions. Although many types of engineering structures have sufl‘lcient flexi— bility to withstand fault displacements of a centi- meter or two without significant damage, it should be remembered that the disastrous failure of the Baldwin Hills Reservoir in 1963 was caused by a fault displacement of a comparably small amount (Hudson and Scott, 1965). Lest the impression be left that all faults in earthquake-prone regions should now be suspect of possible small but frequent displacements, we em- phasize once again that the three distant faults that moved because of shaking generated by the Borrego Mountain earthquake were all “active” faults in that they showed abundant evidence of repeated Quarter- nary displacements. Furthermore, the only segments on which displacements took place were clearly de- lineated by surface exposures (at least in unculti- vated areas), and each of the three faults had histories that suggested similar movements within the previous few years. These findings point out the need for thorough geologic studies prior to engineer- ing developments in faulted areas but also give some confidence that even in highly faulted regions danger- ous areas can reasonably be differentiated from safe areas. REFERENCES CITED Allen, C. R., 1968, The tectonic environments of seismically active and inactive areas along the San Andreas fault system: Stanford Univ. Pub. Geol. Sci., v. 11, p. 70—82. Allen, C. R., St. Amand, Pierre, Richter, C. F., and Nord- quist, J. M., 1965, Relationship between seismicity and geologic structure in the southern California region: Seismol. Soc. America Bull., v. 55, p. 753—797. Babcock, E. A., 1970, Basement structure and faulting along the northeast margin of the Salton Sea [abs.]: Geol. Soc. America Abs. with Programs, v. 2, p. 68. Biehler, Shawn, Kovach, R. L., and Allen, C. R., 1964, Geophysical framework of northern end of Gulf of California structural province: Am. Assoc. Petroleum Geologists Mem. 3, p. 126—143. Brune, J. N., and Allen, C. R., 1967a, A low stress-drop, low- magnitude earthquake with surface faulting: The Im- perial, California, earthquake of March 4, 1966: Seismol. Soc. America Bull., v. 57, p. 501—514. 104 1967b, A micro-earthquake survey of the San An- dreas fault system in southern California: Seismol. Soc. America Bull., v. 57, p. 277—296. Buwalda, J. P., and St. Amand, Pierre, 1955, Geologic effects of the Arvin-Tehachapi earthquake: California Div. Mines Bull. 171, p. 41—56. Crowell, J. C., 1962, Displacement along San Andreas fault, California: Geol. Soc. America Spec. Paper 71, 61 p. Dibblee, T. W., Jr., 1954, Geology of the Imperial Valley region, in Jahns, R. H., ed., Geology of southern Cali- fornia; chap. 2, Geology of the natural provinces: Cali- fornia Div. Mines Bull. 170, p. 21—28. Hope, R. A., 1969, Map showing recently active breaks along the San Andreas and related faults between Cajon Pass and Salton Sea, California: US. Geol. Survey open-file rept. Hubbs, C. L., Bien, G. S., and Suess, H. E., 1960, La Jolla natural radiocarbon measurements I: Am. Jour. Science, Radiocarbon Supplement, v. 2, p. 197—223. Hudson, D. E., and Scott, R. F., 1965, Fault motions at the Baldwin Hills Reservoir site: Seismol. Soc. America Bull., v. 55, p. 165—180. THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 Kovach, R. L., Allen, C. R., and Press, F., 1962, Geophysical investigations in the Colorado delta region: Jour. Geophys. Research, v. 67, p. 2845—2871. Lomnitz, C., Mooser, E, Allen, C. R., Brune, J. N., and Thatcher, W., 1970, Seismicity and tectonics of the north- ern Gulf of California region, Mexico—Preliminary re- sults: Geofisica Internacional, v. 10, p. 37—48. Press, F., 1965, Displacements, strains, and tilts at teleseismic distances: Jour. Geophys. Research, v. 70, p. 2395—2412. Scholz, C. H., Wyss, Max, and Smith, S. W., 1969, Seismic and aseismic slip on the San Andreas fault: Jour. Geophys. Research, v. 74, p. 2049—2069. Smith, S. W., and Wyss, Max, 1968, Displacement on the San Andreas fault initiated by the 1966 Parkfield earth- quake: Seismol. Soc. America Bull., v. 58, p. 1955—1974. Wallace, R. E., 1970, Earthquake recurrence intervals on the San Andreas fault: Geol. Soc. America Bull., v. 81, p. 2875—2889. Wyss, Max, and Brune, J. N., 1967, The Alaska earthquake of 28 March 1964: A complex multiple rupture: Seismol. Soc. America Bull., v. 57, p. 1017—1023. 1968, Seismic moment, stress and source dimensions for earthquakes in the California-Nevada region: Jour. Geophys. Research, v. 73, p. 4681—4694. CONTINUED SLIP ON THE COYOTE CREEK FAULT AFTER THE BORREGO MOUNTAIN EARTHQUAKE By R. O. BURFORD U.S. GEOLOGICAL SURVEY ABSTRACT Two alinement arrays designed for the precise determina- tion of continued slip or deformation were established across surface breaks associated with the Borrego Mountain earth- quake within a few days of the main shock. One array was established on the north break near the point of maximum initial offset (about 38 cm), another on the central break about 6 miles southeast of Ocotillo Wells where the initial offset was about 13 cm. Measurements made over a 33- month period show that a complex pattern of relatively small displacement (0.5 cm) developed at the north site, compared with a simple pattern of larger continued slip (18 cm) at the central site. Apparent logarithmic decrease in’slip rate at the central site, a similar decrease in frequency of aftershock occurrence, and the presence of a thick sedi- mentary cover (3 km) over basement rocks along the central trace in Lower Borrego Valley indicate that continued post- earthquake slip on the central break is largely due to delay in upward propagation of sudden basement-rock displace- ment through the sediments. Studies of continued slip on the San Andreas fault within Cholame Valley after the 1966 Parkfield earthquakes led other investigators to similar conclusions. INTRODUCTION A few days after the Borrego Mountain earth- quake, two alinement arrays were installed across the surface fractures. One was placed across the north break at Borrego Mountain, and the other across the central break in Lower Borrego Valley about 6 miles southeast of Ocotillo Wells (fig. 70). The arrays were designed for long-term periodic monitoring of continuing slip on fracture surfaces (termed “afterslip” by Nason, 1969, 1971) and of shallow elastic strain changes in the Vicinity of the surface breaks. Initial observations were com- pleted at the Borrego Mountain site on April 19 and at the Lower Borrego Valley site on April 25, 1968. A third array, installed slightly beyond the south- east end of the south break near the Fish Creek Mountains on December 10, 1969, was resurveyed on January 25, 1971. All three alinement arrays will be remeasured at approximately yearly intervals. A typical alinement-array configuration is shown in figure 71. Experience with similar installations in central California has demonstrated that monuments installed this way are stable enough under usual surface conditions, but the stability of individual monuments has not been tested. A general indication of monument stability is evident in the final results. The contribution of possible monument instability to the total noise in the alinement-survey data is thought to be negligible compared with the noise level introduced by random errors in instrument positioning, pointing, and scale reading. Angles and alinements were measured with a Wild T—3 theodolite. Typical standard deviations of the changes in mean angle values obtained by com- parison of theodolite pointings on different dates cluster about :2 seconds, as determined according to the following formula: “A = (012+U22)V2! where 0'1 and 02 are the standard deviations of the mean angle values at time 1 and time 2, and 0A is the standard deviation of the change in the mean value of the angle. This error corresponds to an uncertainty of about :1 mm in the value of lateral shift at an end station 100 m from the instrument station. Estimated average positioning errors for theodolite and targets are :0.1 mm; this estimated error adds to the uncertainty in the value of lateral shift by only an insignificant amount. The average error in the determination of alinement changes for stations between the instrument station and the end stations is estimated to be about :02 mm. ACKNOWLEDGMENTS Installation and measurements of the alinement arrays were completed with the help of field crews and other US. Geological Survey personnel. The sites were selected on the basis of preliminary frac- ture maps that were compiled by geologists of the 105 106 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 IO' 5‘ |l6°00' l l l l \ -|\\ \‘.\ [Ix ~'\\ “5'4’0 I60 ‘3“ \39» Epicenter of main shock ‘ ’9’? 1w “’ *FA '1 8 l968 l\ 86 _\\ p“ r \ '\~\ \ 0 4/ \ Magnitude 6.5 \\ /‘ ’01!” \\\\ ' V 5, ,"’ \7Iq4,’\\,<\.\8onego Mountain IO _ I , \\ \) \\0,9 alunement array IO' "\ ) \ 6‘ H ‘V \\ 4' OCOTILLO WELLS HIGHWAY 78 \ \ \ Lower Borrego Valley )(‘\ u alinemem array "' \ \‘74- 5 - _ 5' \ \ ‘Fish Creek Mountains alinement array & \ '\\ 00)? m K‘\ _) ‘w W1 ’7 \ \\ v \\ r\ \F/SH C \“ /‘ \\\\ \o\ (L4_.|_L_1__? KM \\\ ”EEK “\J “\\\ ’96“;\‘ .\ MOON \\I\ 4' X \ MIN \vb 330 l . l \\ S I \ 33. l0 5. "6°00. FIGURE 70. — Main surface breaks associated with the Borrego Mountain earthquake of April 1968, and the locations of three alinement arrays established for afterslip investigation. California Institute of Technology and the US. Arthur Grantz, and James C. Savage and from the Geological Survey. The author profited from discus— critical reviews of the manuscript by Robert 0. sions with Robert M. Hamilton, Malcolm M. Clark, Castle and Robert V. Sharp. This work was sup- THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 reference sIaIIon OrienIcIIon I! O kSurface, fruciures, /or prommem Ierrom scars I I OOOOOO \\\‘_\ AIInemenI 000 O O Oemonumems Instrument ’ sIuIIon Cross-secnon mew of monumem / Standard bronze Ground IobleI suvfoce u Sonqube G-In. dIumeIer IO _ \ ( OnenIotIon reference sIuIIon .5-(1 copper-weld rod, 9/I6 in. dIameIev \ l 0 I 2FEET FIGURE 71. — Typical alinement-array configuration. In the inset is a cross section of the monument construction. ported in part by research funds from the Atomic Energy Commission, Division of Reactor Develop- ment and Technology. RESULTS OF MEASUREMENTS BORREGO MOUNTAIN SITE The initial alinement of the Borrego Mountain array was determined on April 19, the 11th day after the main shock. Observations were repeated on April 24, 1968, and again on February 15, May 5, and December 11, 1969. Comparison of each set of repeat measurements with the initial set yielded the results shown in figure 72. Abrupt steplike offsets between stations are inter- preted‘ as being due to fault slip, whereas lesser offsets comprising a pattern of gradual bending across the terrain are thought to be due either to changes in elastic shear or perhaps to permanent deformation. No attempt has been made to establish definite displacement limits to distinguish between the possible cases. The general displacement pattern developed during the course of the repeated measurements shows that slight right-lateral slip continued to accumulate across the zone of surface fractures 107 STATION NUMBER 2 3 456789|O|||2|3l4 I6 3— 4 ~24~s 2- H4 | I- 1| III.6mm O 4-I9-sa *"l I I \____ I I' ‘_'. A if a g 2— I-o-I F b fl 929mm 2 4-I9-sa I I g 0 I I -I m I- I:I|.3mm 2 2— I‘*\ _I ‘\\ 2-) 3; ‘~o E 31’ 5-5-69 ._I <1 .— z o N E o I 2- . ,!\| I I' ‘V’ I\ I 0 4-I9-se 4'I\\ \ :l' ‘ I— I \I 3.3 mm ' l‘N-‘IL —————— * 2;» I l . r 4~ ‘ 3 \\\\ Ml : \ ‘II~ 2' 69 1r---1 " : | 3.0 mm 449-68 0 I I ‘~ -I " 1l|,l.3 mm 2- 0 50 METERS r+ _______ ‘ L_L_s___I—l 3. SW NE FXGURE 72. — Plan view of alinement changes at the Borrego Mountain array. Each profile shows the change between April 19, 1968, and the date of the repeat survey shown. Approximate slip values are listed to the right of prom- inent offsets of the monument line; these offsets are indicated by slip arrows. Dashed lines indicate un- certainty about actual distribution of displacement between points. generated during the main shock. The slip rate was higher, and slip was concentrated across a relatively narrow zone occupied by faint surface fractures between stations 7 and 9 during the period of the first three sets of observations. Deformation after the third set of observations on February 15, 1969, was much more complex. The slip rate decreased, and the width of the active slip zone increased to at least 70 m by the time of the fourth set of observations on May 5, 1969. Right and left slips with frequent reversals through time are evident on a number of possible fractures distributed across the array (fig. 73). The cumulative effect of 108 STATION NUMBER 2 3 4 5 6 7 8 9 IO ll I2 l3 l4 l5 I6 2" I l- WIUK (3—4-24-68 III t3 mm A ' r E 2:: : Ir 6'69 \ I 6 E 0— ,4 A‘flb I III III I7 I JI- 245-739 a " THIN I'm“ II‘NI' j 2; 2 2 2N1 ll 4 w 3T 5 2_ L L?) mm | I J la-IL-ss Vi/l < I- 222 H—HI —2 LG I I +£02.22 .. 12.2 2/ E |_ I I | I“ a: (I) 22' 2* l | I V/ I- ’2-/, I I" ‘69 /i4; I L II _ I- "\+——o- I ‘3 2M? I VI$HIII0_9mm I ' | I 2- SW 0 50 METERS NE FIGURE 73.—Incremental alinement changes at the Borrego Mountain array. Approximate slip values are listed to the right of prominent ofl’sets of the monument line; these offsets are indicated by slip arrows. Dashed lines indicate uncertainty about actual distribution of displacement be- tween points. this period of readjustment across the laterally ex— panding slip zone was a smoothing—out of the dis- placement profile and an apparent slight decrease in net right—lateral shift across the entire array (figs. 72, 73). The results suggest that .overfling oc- curred during the initial offset and early afterslip; it was followed by a readjustment that was pre- sumably produced by partial relaxation of elastic compression in the southwest block and elastic ex- pansion in the northeast block imposed by gradual decay of right-lateral slip along the fault between the array and the northwest end of the surface break. We began monitoring level changes along the Borrego Mountain alinement array on February 14, 1969. The terrain profile across the fracture zone that is shown in figure 74 is based on the measured heights of the monuments (1—3 cm above the ground surface) relative to station 9. Approximate positions of the surface fractures that were associated with the main shock and the additional lines of afterslip that were detected by alinement changes are in- dicated. The most obvious surface break associated with the main shock was only slightly active during the period of measurements, while a faint surface fracture noted between stations 7 and 8 on April THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 $52°w<—>N 52°E ICMI STATION NUMBER g 2 3 456789I0|I|2l3l4 I5 I6 : g Position: of mo" Positions of B + 40 _ pronounced aherslip slight nnersllp Lu 3 _ 4‘ - ._ 20 4 , p,— U) Lu 0 u 2 . ’2 d -20. Position at lllgm Potifion 0' main m nanny", Apr" I953 surface break , April ISSB ..2 , r" .. MM 0 244-69 _ _ w ”I '/ HI W f4 \,69 W - ,\ 2 ‘1, III? ?III 0 5O METERS VERTICAL DISPLACEMENT (M M) n V FIGURE 74.—Terrain profile along the Borrego Mountain array, and profiles of relative elevation changes between sets of level readings. Dashed lines indicate uncertainty about actual distribution of displacement between points. 19, 1968, developed into a prominent line of activity (figs. 72—74). Results of comparisons of successive sets of level readings are shown in the lower two profiles of figure 74. The maximum elevation changes shown are barely significant; most of the changes are within the estimated average error of :025 mm. Pressure ridges or scarps with crests near stations 6 and 15 may have been accentuated by slight up- ward movement during the period between February 14 and May 5, 1969. LOWER BORREGO VALLEY SITE Initial alinement observations were completed at the Lower Borrego Valley site on April 25, the 17th day after the main shock. Measurements were re- peated on February 14, May 4, and December 11, 1969. Results of comparisons of repeated alinement measurements with the initial measurements are shown in figure 75. In contrast to the results ob- tained for the Borrego Mountain array, afterslip on the central break accumulated at a high rate. The slip activity was confined to the prominent surface fractures formed during the main shock. Surficial material in the northeast block shows very little distortion, except for a possible slight left-lateral slip between stations 9 and 10 which developed late in the course of the measurements. An increase in displacement near the fractures on THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 109 STATION NUMBER 2 3 4 5 6 7 8 9 I0 ll |2 2r ,4 I 0 4-25-55 I ' I- | 2- l 3- 1I63°m 4— l 5- I 6r 'W‘W z} A 5—4-69 "‘1 2 I.F_._—.—/1”|B em 3 c 4-25-68 l ' l-— I- I 5 2- I 2 m 3- | 2 4, 1'169 cm 5 5‘ ' 5 s- | . "W _I 7 i If 5} <75 3' l E 2W4 20 cm 2 l' 1|l | | I 0 4-25-65 i i |p 2» ' I - I 3 I I I‘IIScm 4- / 5' | 6 o SOMETERSI . 7 w |-.. J:[__ 8 "I 9 SW NE FIGURE 75.—Plan view of alinement changes at Lower Borrego Valley array. Each profile shows the change between April 25, 1968, and the date of the repeat survey shown. Approximate slip values are listed to the right of prominent offsets of the monument line; these offsets are indicated by slip arrows. Dashed lines indicate uncertainty about actual distribution of displacement between points. the southwest side is consistent with elastic rebound within surface material, a conclusion that is based on dislocation models developed by Chinnery (1961). Lack of evidence for elastic rebound in the north- east block and the resulting asymmetry in the dis- placement pattern across the fracture zone remain unexplained; the pattern may be associated with a shallower depth to crystalline basement rocks on the southwest side of the fault. Cumulative right-lateral slip across the Lower Borrego Valley array is plotted with respect to time elapsed since the main shock in figure 76. The E .. o .- V vi 0 . m _ I— i ‘9 g B 2 m a, — GI Lu W - o: . —_ 2 w: . u, _ Lu ‘1’- ! m ' a o . ‘ ‘ '5’ <1 '0 >‘ " g > _, an: 5 v E 3 0r -— 3 I: a ”2 '3': 5 s a a '3‘ << E z o 1 _l E. II I I I m 40 ,_ < _l I g I— 30 I ‘3 m Lu 20 2 Z " I0 I) 2 D 0 lMaIn shock O I I I I 1 I 1 I I I I 0 200 400 600 800 IOOO ELAPSED TIME (DAYS) FIGURE 76. —- Cumulative right-lateral slip at the Lower Bor- rego Valley alinement array plotted with respect to time elapsed since main shock. 0) e 8 - (p Q m "‘ 3 ‘° ‘0‘: :3 _* -. 9 ‘ Em " g A _ .. _ 5 E In >. 0 o a, N 3v» '9 t‘ V ~— = e E g P a a so“ 8 s E < 4 LL: 0 3 E | I II | I , Lu — I O ’x < _1 30'— Q. ‘2 _ O _J _ < I: .— Lu I— < _ _| I I— 20— r :1: 12cm. ,/ <2 - I, D: /” m ‘ / 2 x’ '— "/ < / _I D _ 2 8 lo‘Main shock I I IIIIIIII IIIIIIIII IIIIIIIII J I IO IOO IOOO ELAPSED T|ME (DAYS) FIGURE 77. — Cumulative right-lateral afterslip at the Lower Borrego Valley alinement array plotted with respect to logarithmic time scale. initial offset is based on measured slip a day or so after the main shock (Clark, “Surface Rupture along the Coyote Creek Fault,” this volume). It is assumed that offsets measured on the fractures at that time took place during a brief period of main— shock slippage, although early afterslip at a rate of about 2 cm/day probably had already begun (fig. 77). The shapes of the time-slip curves shown N60°E-> STATION NUMBER 23456789|0lll2 E 180— z 9 I4OE\"\ |— ‘=| l cm <>[ 3 2 Com l00’ L” ofterrsnlip (right lateral) d - 7. 5cm oflerslip (right lolerol) 60- g ' \ f." 20- *306m 0.4 cm c — -60' Z _ AUXIIIOFy b'reok Moin cracks —' A ril “368 A ril I968 3; -I00- 9 p l2-lO—69 3. 2 A — 5‘4~69 2 I- 5-4-69 2 ' H v 0 2-14-69 / I— _ l ./ l2-lO-69 Z ,. “J H 2 Lu o < _J CL <1) 5 _l I <1 ~. B l v / l l; Ill "#4 lll? LIJ L__._ . > " ' 0 so METERS I .J FIGURE 78. — Terrain profile along the Lower Borrego Valley alinement array, and profiles of relative elevation changes between sets of level readings. Error bars at station 3 indicate the estimated average error of $0.25 mm ap- plicable to the measured vertical displacement at each station. Dashed lines indicate uncertainty about actual distribution of displacement between points. in figures 76 and 77 indicate that the rate decreased logarithmically after the main shock in a manner similar to that demonstrated for slip on the San Andreas fault in Cholame Valley after the 1966 Parkfield earthquakes (Wallace and Roth, 1967; Smith and Wyss, 1968). A data point based on a fifth set of measurements, completed on January 25, 1971, was added to both figures. THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 STATION NUMBER 5 6 B 9 IO ll I2 N / 0 50 METERS |_L_|._L_._.I——J E E V |~I/Io ,4 E4 l,’]$heor i _:i ’ I g 3 I Lu 6’,\\ 2 2 48mm lli4-5 "'m J lll , 25 I l ‘ l D 0 l L. - 4 / ', Iz-Io—ss ‘ IS I i, 4. i Z O I: D: O I FIGURE 79. —Plan view of alinement changes at Fish Creek Mountains array from December 10, 1969, to January 25, 1971. Approximate slip values are listed to the right of possible offsets of the monument line; these possible offsets are indicated by slip arrows. A terrain profile across the fractures at the array site and elevation changes during the period of February 14 to December 10, 1969, are shown in figure 78. A nearly uniform surface slope to the northeast along the array is interrupted by north- east-facing scarps at the lines of prominent surface fractures. Scarp heights compare with those at the Borrego Mountain site. A third minor scarplet may lie between stations 3 and 4, but its existence was unconfirmed by noticeable surface fractures or by afterslip. A slight broad swell with a crest near station 10 suggests the development of a pressure fold. Significant vertical slip components accompanied the right-lateral slip recorded during 1969 (fig. 78). Relative vertical movement was generally down on the northeast sides of fractures, a continuation of the prominent Holocene trend developed along most of the Coyote Creek fault. (See pl. 1.) FISH CREEK MOUNTAINS SITE The Fish Creek Mountains array was installed, and initial measurements were completed on De- cember 10, 1969, after Malcolm Clark (oral com— mun., 1969) observed that fresh surface cracks along the south break were enlarged by continued slip after the rainy season of the previous spring. The array was positioned slightly beyond the southeast end of the obvious surface fractures, in anticipation of possible southeastward propagation of afterslip along the same trend. Results of the first remeasure- ment, completed on January 25, 1971, suggest right- lateral slip of about 4 mm during 1970, accompanied by possible elastic rebound of about 1x10—* within a sliver of material bounded by apparent slip sur- faces. The results, shown in figure 79, remain ques— THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 tionable; additional measurements are needed to confirm or clarify the pattern. TECTONIC IMPLICATIONS The marked difference in the level of afterslip activity at the north and central alinement array sites is representative of the general difference in behavior of the two separate breaks. The north break is apparently characterized by a general lack of significant afterslip; this fact confirms the re- sults of early postearthquake measurements by Allen, Grantz, Brune, Clark, Sharp, Theodore, Wolfe, and Wyss, (1968) and by Nason (1969) at a number of sites. In contrast, there is abundant evidence of significant afterslip at a number of localities along the central break (Clark, “Surface Rupture along the Coyote Creek Fault,” this vol- ume). The difference in afterslip behavior correlates with differences in thickness of poorly consolidated sediment within the two areas. The available evi- dence indicates that a blanket of sediment about 3 km thick covers basement rocks along the central break within Lower Borrego Valley. Basement rocks are exposed immediately southwest of the north break, and the maximum thickness of sediment in the basin on the northeast side of the north break is probably about 1 km (Hamilton, 1970). The slip-date decay on the central break cor- responds in a general way to the decay in after- shock activity (Hamilton, this volume). A similar history was demonstrated for afterslip and after- shocks of the 1966 Parkfield earthquakes (Eaton, 1967; Wallace and Roth, 1967; Smith and Wyss, 1968). Thus, afterslip on the central break probably was due chiefly to delay in propagation of slip from the granitic basement upward through the layer of sediments. CONCLUSIONS During a 33-month period following the Borrego Mountain earthquake, the initial slip of about 13 cm on the central break was increased to perhaps as much as 31 cm through afterslip activity. During this same period, only about 0.5 cm of additional slip accumulated across the north break adjacent to Borrego Mountain, where the initial slip amounted 111 to about 38 cm. The pattern of afterslip distribution at the Bor- rego Mountain site was particularly complex com- pared to the simple pattern of continued slip across well-defined lines of surface fractures at the Lower Borrego Valley site on the central break. The afterslip rate at the array site on the central break apparently decreased logarithmically during the period of repeated measurements; this decrease corresponded to a similar decrease in frequency of aftershock occurrence. The presence of a thick blanket of sedimentary cover over basement rocks along the central break in Lower Borrego Valley and an afterslip and after- shock history similar to that established for the Parkfield earthquakes of 1966 indicate that after- slip on the central break probably was due largely to a delay in propagation of sudden basement-rock displacement through the sediments. REFERENCES CITED Allen, C. R., Grantz, Arthur, Brune, J. N., Clark, M. M., Sharp, R. V., Theodore, T. G., Wolfe, E. W., and Wyss, Max, 1968, The Borrego Mountain, California, earth- quake of 9 April 1968—a preliminary report: Seismol. Soc. America Bull., v. 58, no. 3, p. 1183—1186. Chinnery, M. A., 1961, The deformation of the ground around surface faults: Seismol. Soc. America Bull., v. 51, no. 3, p. 355—372. Eaton, J. P., 1967, Instrumental seismic studies, in The Parkfield-Cholame, California, earthquakes of J une-Aug- ust 1966: U.S. Geol. Survey Prof. Paper 579, p. 57—65. Hamilton, R. M., 1970, Time-term analysis of explosion data from the vicinity of the Borrego Mountain, California, earthquake of 9 April 1968: Seismol. Soc. America Bull., v. 60, no. 2, p. 367—381. Nason, R. D., 1969, Continuing fault movement after earth- quakes: EOS (Am. Geophys. Union Trans.), v. 50, no. 4, p. 252. 1971, Instrumental monitoring of postearthquake fault movements (afterslip), in The San Fernando, California, earthquake of February 9, 1971: U.S. Geol. Survey Prof. Paper 733, p. 89-90. Smith, S. W., and Wyss, Max, 1968, Displacement on the San Andreas fault subsequent to the 1966 Parkfield earthquake: Seismol. Soc. America Bull., v. 58, no. 6, p. 1955—1973. Wallace, R. E., and Roth, E. F., 1967, Rates and patterns of progressive deformation, in The Parkfield-Cholame, California, earthquakes of June-August 1966: U.S. Geol. Survey Prof. Paper 579, p. 23—40. HOLOCENE ACTIVITY OF THE COYOTE CREEK FAULT AS RECORDED IN SEDiMENTS OF LAKE CAHUILLA By MALCOLM M. CLARK, ARTHUR GRANTZ, and MEYER RUBIN U.S. GEOLOGICAL SURVEY ABSTRACT Much of the late Holocene history of the southern part of the Coyote Creek fault, including the part that ruptured during the magnitude 6.4 Borrego Mountain earthquake of 1968, can be inferred from progressive vertical deformation of the flat-lying sediments deposited in Holocene Lake Ca— huilla, which covered parts of the fault until at least 800 years ago. The recurrence interval for tectonic events like that of 1968 along the Coyote Creek fault was determined by comparing vertical components of displacement of 1968 with earlier movement recorded in offset sediments of Lake Cahuilla. Pre-1968 displacements were measured from two structural profiles across the central break of 1968 and from a trench across a branch of this break. The offset strata were dated by C14. Fault offsets measured in the two profiles suggest that during the last 860 years, 1968—size events occurred every 160 or 190 years. Total vertical deformation (fault offset plus adjacent bending) measured across the branching break suggests that during the last 3,080 years, 1968-size events occurred every 205 years. Vertical fault offsets alone across this same branching break suggest that during the last 1,230 years, 1968-size events occurred every 195 years. Thus, three sets of CH-controlled structural measurements suggest a recurrence interval of roughly 200 years for tectonic events like that of 1968. This recurrence interval is compatible with those intervals estimated from historic seismicity of the San Jacinto fault zone and from long-term slip rates and historic faulting of the San Andreas fault north of the Transverse Ranges. Rates of late Holocene strike-slip displacement along the Coyote Creek fault were derived in two different ways from measured rates of vertical displacement. Both methods sug- gest that the strike-slip rate is about 3 mm/yr, a value compatible with Sharp’s (1967) estimated minimum rate of 2.5 mm/yr for the San Jacinto fault zone during the Pleistocene. Structural profiles and trenches also show that total vertical deformation is a combination of slip at the fault and inelastic bending of equal or greater amount within a zone 2—50 m wide or more on either side of the fault. Thus, at least in poorly consolidated deposits, estimates of seismic moment or stress-drop at faults during earthquakes may be in error by a factor of 2 or more if they are based solely on fault offsets. The trenches and natural exposures reveal that most late Holocene faulting occurred repeatedly within zones 1 m or 112 less in width along the same traces. Two exposures show recurrent fault displacement along a single trace during the last 2,400—3,000 years. In one location at least three and possibly 17 episodes of faulting have taken place during the last 3,000 years along a planar zone only a few tens of milli- meters wide. A few new fractures, however, do appear in the fault zone. A conspicuous new fracture that branches from a trace showing repeated late Holocene movement broke 1,650-year-old sediments for the first time in 1968. INTRODUCTION Tectonic deformation of Holocene strata amen- able to C“ dating has yielded valuable information about timing, position, and rates of Holocene activ— ity along those parts of the Coyote Creek fault that ruptured during the Borrego Mountain earthquake of April 9, 1968 (fig. 80, 81; pl. 1). Most of the information was obtained from three trenches ex- cavated across the central break and its branches (see Clark, “Surface Rupture Along the Coyote Creek Fault,” this volume) in November 1969. These trenches showed progressively greater vertical offset of older strata. The oldest dated layer (about 3,000 yr B.P.) is offset nearly three times as much as the base of the youngest (860 yr B.P.). Natural exposures of Holocene strata elsewhere along the 1968 rupture show evidence of pre-1968 faulting in some places and evidence of lack of pre-1968 move- ment in others. Displacement rates and recurrence intervals de- scribed here are based on local vertical components of movement along a predominantly strike-slip fault. This approach was taken for several reasons: (1) We could not find offset geologic features from which pre-1968 horizontal displacements could be dated; (2) vertical displacements are locally large along the Coyote Creek fault (Clark, “Surface Rupture Along the Coyote Creek Fault,” this volume), and the vertical component of displacement in 1968, although variable, was everywhere in the same sense as in the recent past; (3) shelly sediments deposited in THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 Holocene Lake Cahuilla provide a means of dating small, very recent progressive vertical offsets along the central and south breaks of the Coyote Creek fault. The use of vertical offset as a proportional mea- sure of horizontal or net displacement rates and as a basis for determining recurrence intervals for slip on the Coyote Creek fault rests on the assumption that the pattern of slip in 1968 was typical of prior late Holocene events. This assumption can, at present, only be evaluated by the internal consistency of our data and by their agreement with other kinds of information on displacement rates on the Coyote Creek and related faults. Although we feel that this assumption is probably valid for the Coyote Creek fault, we recognize that it cannot be rigorously test- ed until other sorts of data become available. (For example, the results of long-term geodetic monitor- ing of vertical and horizontal movement across the fault.) We also tacitly assume that all displacements at the fault result from sudden slip accompanied by earthquakes and by associated preearthquake and postearthquake tectonic creep similar to that as- sociated with the earthquake of April 1968. Regard- less of the validity of these assumptions, our data provide useful information on the history of vertical deformation along the Coyote Creek fault. The shelly strata in which recent vertical dis- placements along the 1968 breaks could be measured were deposited in Holocene Lake Cahuilla. This fresh-water lake occupied the Salton Basin and covered part of Lower Borrego Valley from at least 1,900 years ago until perhaps as recently as 220:100 years ago, according to C“ dates of charcoal, pelec- ypods, and calcareous tufa associated with shore- lines at high stands of the lake (reported in Hubbs and others, 1960, 1963, and 1965 and Hubbs and Bien, 1967). Prominent well-preserved beaches and waterlines of high stands lie 12.8—13.7 m above sea level (fig. 81; pl. 3; Stanley, 1963). This highest shoreline roughly follows the 40-foot contour on plate 1. Shorelines of an older lake, considered to be late Pleistocene by Stanley (1963, 1966), have been detected south of the area of the 1968 rupture at about 49 m above sea level. We have not identified the deposits and shorelines of this older lake in the area of the 1968 rupture, nor recognized its deposits in the trenches. The highest shoreline of Holocene Lake Cahuilla is not obviously displaced where it crosses the 1968 rupture near location 19 on plate 1. However, the uncertainty of the horizontal position of the shore- line at this place is on the order of 10—15 m, so any horizontal displacement of the shoreline less than about 15 m would be difficult to recognize. 113 The youngest deposit of Lake Cahuilla (860 yr B.P.) in Lower Borrego Valley, called layer A in this report, is an important datum for measuring and estimating the latest Holocene activity of the Coyote Creek fault. This deposit is a prominent whitish shelly layer as much as 0.5 m thick that is conspicuous both in the field and on aerial photo- graphs (fig. 80). In spite of its youthfulness, much of it has been reworked or entirely removed by erosion or buried by waterborne and windborne mud, silt, and sand. Where layer A is exposed, pelecypod and gastropod shells are abundant on the surface. Erosion of layer A from the top of the locally upwarped sediments on the northeast side of the central break of 1968 at location 22.5 on plate 1 indicates uplift of this small area between the time that Lake Cahuilla withdrew, less than 860 years ago, and 1968. Similarly, prominent scarps cutting layer A at the central break of 1968 clearly reveal vertical displacement along the fault between 860 years ago and 1968. The displacement rates and recurrence intervals derived in this report are based on C14 ages with analytical uncertainties that range from :200 to :600 years, or from :10 percent to :23 percent of the nominal ages. These uncertainties are a mea- sure of analytical precision. They cannot tell us how close the stated ages are to actual ages, nor do they necessarily encompass the actual ages. For these reasons, only the nominal C14 ages, without un- certainty ranges, were used in the calculations described here. ACKNOWLEDGMENTS Many people contributed to this investigation and report. R. V. Sharp helped sketch the walls of all trenches, and J. E. Kahle, of the California Division of Mines and Geology, helped collect shells for C14 dating and assisted in surveying the profiles. Tom J. Allen of Borrego Springs kindly loaned us a sur- veyor’s level for the profiles. The US. Bureau of Land Management, the Southern Pacific Land Co., and its lessee, the Landmark Corp. of San Diego, were very cooperative and readily gave permission to dig trenches on their lands. We greatly appreciate the ideas and comments of our Geological Survey colleagues, K. R. Lajoie, R. V. Sharp, J. H. Healy, R. D. Brown, Jr., and R. E. Wallace. HISTORY OF FAULT DISPLACEMENTS AND RECURRENCE INTERVALS CENTRAL BREAK Displacements of the Coyote Creek fault during the last 860 years are recorded by offset and warping of layer A at the central break of the 1968 earth- quake. (See fig. 82.) For a distance of about 4 km 114 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 FIGURE 80.—Vertical aerial photographs of central and branching breaks of the 1968 earthquake at Old Kane Spring Road showing a shallow graben between the breaks and the location of trenches 1, 2, and 4 and profiles 1 and 2. (These photographs were taken in April 1969, about 1 year after the earthquake. Fig. 37 shows the same area a few weeks after the earthquake and gives a general description of this scene. Numbers along the bottom of the photographs refer to the coordinate system of pl. 1.) Heavy rains of July 1968 and January 1969 enlarged tectonic fractures of the branching break to gaping fissures (dark lines). The sediments in trench 1 record repeated late Holocene ofl‘sets along the branching break, whereas the THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 115 1968 fracture crossed by trench 4 broke the upper 3 m of sediments for the first time. Dendritic light-toned strips northeast of the branching break in the bottom part of the right-hand photograph are topographically high remnants of the youngest deposit (layer A) of Lake Cahuilla. This whitish shelly layer stands as much as 0.5 m above the 100 METERS adjacent surface, which has been lowered by wind erosion. These strips may represent topography inverted from channels beneath which the whitish lake beds became slightly indurated. Aerial photographs by Light Photo- graphic Squadron 63, U.S. Naval Air Station, Miramar, Calif. 116 J’ Holocene . . (‘ - . shoreline !\ Soulh break ‘\ ’\ / 0 IO 20 30 KM I 1 1 J FIGURE 81.—Index map showing the three principal breaks of the 1968 rupture of the Coyote Creek fault. Dashed line shows the highest shoreline of Holocene Lake Cahuilla (modified from Strand, 1962; Rogers, 1965; and Jennings, 1967). s _ SW NE \\\ EXPLANATION ’ PROHLE I F ____ \‘ 51 Base of layer A 5 \+ E . . ————— \+ Posmon of base of layer A PROFILE 2 \‘\ estimated from profile of ground 4_ +\, surface \g; . ...... ‘ ' m 70714 L T: E2 Proper: posmon 5 rEcroN/cg ’iF/fl/L r 0 05° r—3— OFFSEr 3 RT of layer A I; _ — x25 mm/yr, because these beds are probably no more than 2 my. old. The long-term slip rate for the San Andreas fault is about eight times as great, which is in gen— eral agreement with the 7: 1 ratio between the recurrence intervals of magnitude 6.4 earthquakes along the two faults. Adjustments in these ratios to accommodate possibly larger post-Irvingtonian dis- placement on the San J acinto and higher displace- ment rates on the San Andreas would tend to increase both ratios. Comparison of ratios involving the main San J acinto fault zone on one hand (long- term slip) and the Coyote Creek fault on the other (late Holocene recurrence intervals) may be justi- fied because the Coyote Creek fault bears clear evidence of late Holocene activity, whereas definite evidence of such activity has not been discovered on other strands of the San J acinto fault zone east of Borrego Badlands. The fact that the long-term slip rate and the recurrence interval for magnitude 6.4 events on the San Andreas fault are, respective— ly, several times larger and smaller than on the San J acinto—Coyote Creek fault lends some support to our estimated recurrence intervals for the Coyote Creek fault. The large ratios also follow from the THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 fact that the slip rates and recurrence intervals for the San Andreas north of the Transverse Ranges probably reflect most of the relative motion between the Pacific and North American plates, whereas the Coyote Creek fault probably shares this motion with at least three other active fault systems south of the Transverse Ranges (the Banning—Mission Creek segment of the San Andreas, the Elsinore—Agua Caliente, and the Newport-Inglewood fault zones). Our 200-year recurrence interval for the 1968 break is also compatible with recurrence intervals estimated from historic seismicity of the entire San Jacinto fault zone. Seven earthquakes of magnitude 6.0 or greater have occurred in the zone since 1899, at an average interval of 11.4 years. (See fig. 86; table 16.) Assuming (1) a length of 215 km for the San J acinto fault zone, (2) a constant crustal strain rate along the entire zone, and (3) that ground ruptures are associated only with earthquakes of magnitude 6.0 or greater and release all the strain accumulated along the ruptured length of the zone, a simple relation, derived from the concepts of Wallace (1970), exists: _Rt>60 ....... Apr. 21, 1918 ........ 33%° 117° 6.8 18.3 July 23, 1923 ........ 34° 1171/4" 6%. 5.3 Mar. 25, 1937 ...... 33°28' 116°35’ 6.0 13.7 Oct. 21, 1942 ........ 32°58’ 116°00’ 6.5 5.6 Mar. 19, 1954 33°17’ 116°11’ 6.2 11.4 Apr. 9, 1968 ........ 33°12’ 116°08' 6.4 14.0 1Near San J acinto. ”Estimated. where R,:recurrence interval at a point in the zone for earthquakes that produce ruptures, R,:recurrence interval for all earthquakes along the zone that produce ruptures, L,:total length of the fault zone, and Lzzlength of rupture for a given earthquake. Equation 1 simply states that the number of events, Lt/Lz, required to break the entire length of the zone multiplied by the average time interval between these events, Rt, equals the interval between succes- sive events at the same point in the zone, R, For a point in the San Jacinto fault zone (R,=11.4 years, L,:215 km), 2450 Because L. is not known for the six earthquakes that happened before 1968,1 equation 2 is plotted as curve A in figure 87 to show better the range of possible values of L, and R,. The recurrence. interval of 200 years reported here and the length of the 1968 rupture of 31 km (pl. 1) are within a factor of about three of values derived from the historic record in equation 2 (fig. 87, curve A). A reasonable value for average length of ground rupture, L,“ on curve A of figure 87, namely 12 km, corresponds to our recurrence inter- val, R1, of 200 years, and this value also falls Within the range (roughly 5—70 km for magnitude 6-7 earthquakes) given by Wallace’s (1970, fig. 2) least- squares curve for data showing the relation between rupture length and magnitude of earthquakes in western North America. The 1968 rupture is con— siderably longer and gives a value for R, of about 80 1N0 systematic, detailed investigations of the San J acinto fault zone were reported after the earlier earthquakes. Although Danes (1907) described a 3-km surface rupture resulting from the 1899 earthquake, Sharp (1972) indi- cated that the known surficial effects probably are entirely of landslide origin. After the 1918 earthquake, Townley (1918) and Rolfe and Strong (1918) found no evidence for surface ruptures at the places they checked. Laughlin, Arnold, and Kew (1923) evidently did not inspect traces of the San Jacinto fault after the 1923 earthquake, and we are not aware of any field investiga- tions of the effects of the earthquakes of 1937, 1942, or 1954. 123 I04- - MEASURED R. I.— 200 YEARS RECURRENCE INTERVAL (YRS) l0 ........ . 4....... a I I0 l968=32km |oo RUPTURE LENGTH (KM) FIGURE 87.—Recurrence intervals, Rx, at a point in the San Jacinto fault zone for different values of rupture length, L... Curve A, all historic earthquakes of magnitude 6 and greater; curve B, earthquakes of 1899, 1918, 1942, and 1968 only (magnitude 6.4 and greater). Based on equation 1 and data in table 16, with a total length of fault zone, L., of 215 km. For curve A, average interval between earthquakes, R, is 11.4 years. For curve B, R. is 22.8 years. IOOO instead of 200 years on curve A of figure 87. Considering the broad nature of the assumptions underlying equation 1, these values represent fair agreement between the two independent estimates of recurrence intervals. Two additional assumptions about the historic earthquakes bring much closer agreement between the recurrence intervals estimated by the two dif- ferent methods. Of the historic earthquakes in table 16, the ones most likely to have been associated with significant ground ruptures are those larger than the .1968 earthquake. Assuming that the earthquake of 1899 was about equal to that of 1918 (magnitude 6.8, fig. 86) and that the earthquakes of 1923, 1937, and 1954 (magnitude less than 6.4) were not as- sociated with significant faulting, the remaining four earthquakes (magnitude 6.4 or greater) have an average interva1,-'Rt, of 22.8 years. (Notice that the intervals between these four earthquakes, 18.3, 24.6, and 25.4 years, differ less than the intervals between all seven earthquakes given in table 16.) By using the value of R, (22.8 yr) for these four largest earthquakes, equation 1 then becomes R _4900 ..-L—z, (3) 124 which is plotted as curve B in figure 87. With curve B, our recurrence interval of 200 years corresponds to a rupture length of 25 km; conversely, the 1968 rupture length, 31 km, corresponds to a recurrence interval of about 160 years. These values represent very close agreement between the two estimates of recurrence intervals, particularly if the ends of the 1968 rupture, which had very small displacements (pl. 1), overlap by several kilometers into adjoining segments of the fault broken by other earthquakes. Such overlap would shorten the effective length of the rupture used in this analysis. In View of the uncertainties, however, both in the recurrence inter- vals estimated from field data and in the assump- tions underlying equation 1, a conclusion that sig- nificant faulting and regularity of occurrence attend only those earthquakes of roughly magnitude 6.4 and greater along the San Jacinto fault is highly speculative. The record of historic seismicity, how— ever, definitely does not conflict with a recurrence interval of 200 years; indeed, historic seismicity ap- pears to support it. RATE OF LATE HOLOCENE STRIKE-SLIP DISPLACEMENT The rate of late Holocene right-lateral strike-slip along the Coyote Creek fault can be estimated from our measurements in two ways: by applying the 200- year recurrence interval for 1968-size events to the right-lateral strike-slip displacements observed in 1968 or by multiplying the vertical deformation rates by the ratio of the horizontal to the vertical components of slip measured along the relevant part of the central break after the 1968 event. Both estimates rest on a number of assumptions and other estimates, and as with the recurrence intervals, they can be tested only by comparison with other kinds of displacement-rate data. The assumptions are that ( 1) late Holocene vertical displacement rates across the fault at the structural profiles and trench 1 are proportional to horizontal displacement. rates on the central break; (2) the late Holocene recurrence in- tervals we derive from the rate of vertical deforma- tion at three points on the central and branching breaks can be extrapolated to the entire Coyote Creek fault; and (3) the total fault slip in 1968, both sudden offset at the time of the earthquake and subsequent creep, is represented in the fault off— sets recorded on plate 1. The chief estimates are of (1) the horizontal and vertical components of slip in 1968 (and the ratio between these components) along the part of the central break that was most strongly uplifted on the southwest side in 1968 and THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 for which we estimated recurrence intervals and (2) the ratio of the horizontal component of drag to strike slip along the central break. Data sum- marized in the next section suggest that drag at least commonly accompanies slip along faults where they traverse unconsolidated or poorly consolidated sediments near the earth’s surface. Each of the two estimates of late Holocene strike slip rests on the stratigraphically based CM-dated vertical-deformation rates derived from structural data in profiles 1 and 2. The first estimate, inasmuch as it uses the ZOO-year recurrence interval, is in addition based on the C14-dated vertical deformation rates derived from structural data in trench 1. This estimate applies the 200-year recurrence interval to the horizontal-slip components of the 1968 event. The estimate relies heavily on the extent to which the vertical components of displacement near the profiles and at the trench are representative of both the 1968 and earlier events. Comparison with other data in the preceding section suggests that they are, in general, representative. The results of the first approach are presented in table 17. The horizontal deformation rates are de- TABLE 17.—Rate of late Holocene strike-slip deformation on the Coyote Creek fault based on 200-year recurrence interval for 1968-size events Maximum 1968 Total horizontal deformation rates based horizontal on various assumed ratios of horizontal Break disp‘ t drag to fault slip (mm/yr) (mm) 0: 1 1:1 2:1 North ................ 380 1.9 3.8 5.7 Central ............. 300. 1.5 3.0 4.5 South ................. 140 .70 1.4 2.1 rived by first converting the maximum horizontal fault displacements (slip) observed in 1968 (re- corded on pl. 1) to total horizontal deformation by multiplying them by the factors 1, 2, and 3, corre- sponding to three alternatives: no horizontal drag, drag equal to fault slip, and drag twice as large as fault slip. The maximum 1968 slip was chosen be- cause we believe it probably most closely approaches “deep” or bedrock displacement. Horizontal drag is probably a significant factor. Allen, Wyss, Brune, Grantz, and Wallace (this volume) reported that such deformation accompanies displacement along the nearby Superstition Hills, Imperial, and San Andreas (Banning—Mission Creek) faults, which also traverse areas of late Cenozoic sedimentary deposits. (See also the following section of this report.) However, they did not report the propor- tion of horizontal drag to fault slip, so we arbi- trarily use the ratios of vertical drag to fault slip THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 found in trench 1 and profiles 1 and 2 (1:1 and 2:1) and also the ratio for no drag (0:1). The ranges in rates so obtained for the more strongly displaced northern and central breaks are, respec- tively, 1.9—5.7 mm/yr and 1.5—4.5 mm/yr. Within these ranges, the rate-s corresponding to a drag-to- slip ratio of 1: 1, namely 3.8 and 3.0 mm/yr, may possibly be best, because this ratio characterized vertical drag in two of the three structural sections measured in profiles 1 and 2 and trench 1. A similar analysis, presented in table 18, esti- mates vertical deformation rates along representa- tive parts of the north, central, and south breaks of 1968. The vertical drag coefficients used, 1: 1 and 2:1, are based on those found in trench 1 and profiles 1 and 2. Such estimates may be useful, for example, in geomorphic, structural, or engineering studies of the hills or scarps that are being tectoni- cally uplifted now in the region of the Coyote Creek fault. TABLE 18.—Rate of late Holocene vertical deformation at selected points along the Coyote Creek fault based on 200- year recurrence interval for 1968-Size events Total vertical deformation Sample 1968 rates based on assumed ' ' ratios of vertical drag 3m“ {$3515le (:22???) to fault slip (mm) (mm/yr) 1 :1 2:1 North ................ 200 8.1 2 3 Central ............. 100 19.1 1 1.5 South ................. 40 30.2 .4 .6 The results of the second approach to estimating the rate of late Holocene strike slip along the fault are presented in table 19. They were derived by multiplying (1) the stratigraphically based late Holocene vertical deformation rates in profiles 1 and 2 and (2) the ratio of the horizontal to the vertical component of slip along the part of the central break showing the greatest and most uniform uplift in 1968. This ratio is the cotangent of the angle between the net slip and the horizontal com- ponent of slip if the Coyote Creek fault is indeed vertical, as suggested by field mapping. TABLE 19.—Rate of late Holocene strike- slip deformation on the Coyote Creek fault based on vertical deformation rates and ratio of horizontal to vertical displace- ment Rate adjusted for 1968 horizontal to vertical displacement ratio of 2.7: 1 and various assumed ratios of horizontal drag to fault slip (mm/yr) 0 : 1 1 : 1 2: 1 0.5 1.4 2.7 4.0 Displacement rate (vertical fault slip only) in profiles 1 and .9 (nun/m) 125 The deformation rates in table 19 are based on an average ratio of horizontal to vertical slip derived from ratios measured at eight places showing large right slip along the most uplifted part of the central break of 1968. These places lie between localities 16.1 and 21.9 on plate 1 and encompass profiles 1 and 2. The ratios at the eight places range from 1.3 to 4.6 and average 2.7. Although only measurements with more than 100 mm of right slip in 1968 were used, the average value of the ratio would not differ greatly if all the ratios recorded in plate 1 from the most uplifted area were used. The product of the vertical deformation rate and the average ratio was modified by factors for no horizontal drag, drag equal to, and drag twice as large as horizontal slip. The strike-slip displacement rates that result from this analysis, 1.4—4.0 mm/yr, are similar to those obtained by the first. This approach avoids using the recurrence intervals, which depend heavily on the representativeness of the 1968 deformatiOn at trench 1 and near profiles 1 and 2, and it has the advantage of using the average of eight ratios of horizontal to vertical slip from points evenly spaced along 5.8 km of the most uplifted part of the central break. The considerable variation in this ratio along even this most uplifted part of the central break, however, causes a large uncertainty in the calculated strike-slip deformation rate. The rates of late Holocene strike slip estimated for the Coyote Creek fault in tables 17 and 19 are comparable to Sharp’s (1967) estimate of the rate of Pleistocene slip, >2.5 mm/yr, for the main San J acinto fault. Sharp’s rate falls within the range of strike-slip rates estimated by both approaches and lies closest to the displacement rates estimated for the central break with a drag-to-slip ratio of 1: 1. For‘the north break, estimated only by the first method, Sharp’s minimum rate lies between the values for drag-to—slip ratios of 0: 1 and 1: 1. The fairly close correspondence of Sharp’s estimates and ours is encouraging but, because of differences in the age and character of the offset features used in Sharp’s estimate and ours, does not constitute conclusive verification of our estimates. Neverthe- less, it does support the suggestion, based upon our estimates of vertical displacement rates and re- currence intervals for the central break and the most commonly Observed ratio of vertical drag to fault slip in the profiles and trenches, that the rate of late Holocene strike-slip displacement on the Coyote Creek fault is about 3 mm/yr. 126 SIGNIFICANCE OF DRAG ADJACENT TO THE FAULT Both horizontal and vertical components of fault drag were found adjacent to the fractures of the Coyote Creek fault, even where it crosses late Holo- cene sedimentary deposits. This evidence of drag indicates that displacements across fractures alone, at least where measured at the earth’s surface, may be significantly less than total permanent deforma- tion in this and other fault zones. The vertical com- ponent of drag is clearly demonstrated in the deformation of originally flat-lying Holocene sedi- ments exposed in trench 1 (fig. 83) and in steeply dipping Quaternary sediments adjacent to the fault in Ocotillo Badlands; horizontal drag is shown by a characteristic bending of strike of the steeply dipping strata next to the fault (Sharp and Clark, this volume). Thus, the horizontal and vertical offsets recorded on plate 1, which represent the cumulative slip (including subsequent creep) along the discrete fractures that constitute the 1968 fault break, are probably not a true measure of the total displacement across the fault zone associated With the 1968 earthquake. The fact that figure 83 shows greater vertical drag in successively deeper Holocene beds indicates that the drag is permanent strain. Our data do not reveal whether the ratios between vertical slip on macrofractures and drag (roughly 1: 1 in trench 1 and profile 2 and 1: 2 in profile 1) hold also for horizontal movement at those places or elsewhere along the Coyote Creek fault. Preliminary measurements of horizontal creep from an aline— ment array on the central break established near profile 2 a few days after the 1968 earthquake sug- gest that postearthquake displacements across the smaller of two fractures in the fault zone consist of roughly equal amounts of slip on fractures and dis- tortion in the adjacent sediments, although the larger fracture does not show drag (Burford, this volume, fig. 75). Whether this distortion is perma- nent or is instead elastic and temporary is not yet known. Without repetitive measurements of pre- earthquake lines across the rupture of 1968 on the Coyote Creek fault, the distribution of total strain across the rupture cannot be determined. Repeated regional triangulation across the northern part of the Coyote Creek fault near the Truckhaven road (Whitten, 1956), however, suggests continuing right-lateral strain across the general zone of the Coyote Creek and San J acinto faults. THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 It is of interest here that where preearthquake geodetic 1i es 100 m long had been established across the segme ts of the Imperial, Superstition Hills, and San Andr as (Banning—Mission Creek) faults on which sli was triggered by the 1968 earthquake, drag defo mation was found to have occurred in the horizontal plane. (See Allen and others, this vol- ume.) The horizontal offsets determined by re- measuring these geodetic lines after the earthquake were consistently larger than the slips measured at the fault itself, indicating that distributed horizontal deformation (drag) accompanied the fault slip. It should be emphasized that with the exception of the differentially uplifted and intensely deformed block of Ocotillo Badlands, nonelastic strain (slip plus drag) along the Coyote Creek fault in the area of the 1968 break accumulated only in a narrow zone. This zone is about 4 m wide on the branching break cut by trench 1 and about 100 m wide in profiles 1 and 2 and in the zone of vertical dips exposed next to the central break of 1968 in the Ocotillo Badlands. Deflected fold axes in the Ocotillo Badlands indicate that the active fault zone there may, at times, have been as much as a few hundred meters wide. Even the large possibly diapiric struc- tural knot of Ocotillo Badlands, which lies between the central and north breaks of the Coyote Creek fault, is less than 2 km wide. Compared with the elastically strained zone on both sides of such breaks (the elastic zone of the .Imperial fault, for example, is 120 km wide according to Scholz and Fitch, 1969, p. 6652), these zones of nonelastic deformation have negligible width. For regional studies the slip and drag components, where present, can usually be considered together as displacement at the fault, but for detailed studies along active faults, these two aspects of deformation must be differentiated. Drag has been reported along other active faults. It was a common feature of displacement along the San Andreas fault in 1906. Lawson and others (1908, p. 148) reported that its effects extended, at one place, almost 1 mile from the fault, and they stated that: The recognition of the distribution of the movement on auxiliary cracks, some of which may not have appeared at the surface, and the deformation of the ground along the zone of rupture, justifies the conclusion that, except under peculiar conditions * * * the maximum figures obtained for the displacement by the measurement of offsets at the surface must be a minimum expression for the true extent of the movement in the firm rocks below. Drag also accompanies deformation during earth- quakes at thrust faults, as at San Fernando in 1971 (U.S. Geol. Survey, 1971; Kamb and others, 1971), THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 and at normal faults, as at Hebgen Lake, Montana, in 1959 (Myers and Hamilton, 1964, p. 81—83). It probably occurs, to some extent, in most episodes of faulting. Drag associated with slip events along active faults is generally difficult to detect where there are few manmade features of known predeformation shape such as roads, fences, or survey lines. Our observa- tions serve as a reminder that where such features or lines are not present, slip measured across the surface breaks of active faults may not represent total offset and should not be taken as a quantitative measure of total strain. Thus, for example, conclu- sions drawn from comparisons between local strain at faults and regional strain may be in error, and calculations of seismic moment or stress-drop will be too low, if determined solely from fault slip where drag was a significant component of the deformation. POSITION OF PRE-1968 FRACTURES The trenches and natural exposures of Holocene sediments along the 1968 rupture record important information about the position of past fractures. In some locations fracturing has occurred repeatedly along exactly the same fault plane for as long as 3,000 years, whereas elsewhere sediments as much as 1,650 years old were first ruptured in 1968. Trench 1 (fig. 83) reveals the clearest evidence of repeated displacement along a single narrow trace. Differential displacement of the three dated layers cut by the trench demonstrates at least three episodes of faulting in the last 3,080 years. Furthermore, if the total vertiCal deformation from each episode were the same as that in 1968 (estimated 100 mm), then 17 such events would have been required to produce the total offset (1.7 m, including that of 1968) of the oldest dated layer exposed in that trench. What- ever the actual number of tectonic events, the trench shows that about one-half the total vertical deforma- tion in the last 3,080 years has taken place along a fault plane that is in places no more than a few tens of millimeters wide. Two trenches excavated across the main scarp of the central break also showed that most past fault displacement has been confined to a narrow zone. Trench 2 was at the site of profile 2 (fig. 80), and trench 3 was about 1 km farther southeast along the same scarp. Although no strata could be corre- lated across the fault in either trench (layer A was eroded off the surface at each site), the main rupture zone in each consisted of several discrete fractures a few tens of millimeters wide that occupied a band less than 1 m wide. 127 About 10 km farther southeast, along the south break, natural exposures in the walls of a wash that has developed since Lake Cahuilla withdrew also show a long history of repeated activity along the same fault strands. The wash is 1.5 m deep and cuts across three old en echelon fault strands, two of which moved either at the time of the earthquake or later, as a result of creep (Clark, “Surface Rup- ture Along the Coyote Creek Fault,” this volume, fig. 49h; pl. 1, loc. 30.7). Figure 88 shows two episodes of channel cutting along a relict fracture of one of these breaks, as exposed in the walls of the wash. Channeling is interpreted to have been caused by water flowing in the fracture toward nearby areas of active collapse in the same fracture. Such collapse occurs only in open fractures (Clark, “Collapse Fis- sures Along the Coyote Creek Fault,” this volume) ; hence, these channels were cut soon after the faulting that opened the fractures and during periods when Lake Cahuilla had withdrawn below the elevation of this location (about 5 m below sea level). C“ dates on shells that fill the lower channel show that it may be as much as 2,450 years old (W—2469), whereas the younger channel is post-Lake Cahuilla (860 years) but pre—1968. The earlier date represents a maximum for the older channel, inasmuch as the shells may have been reworked from older deposits. In contrast to fault strands on which movement had occurred repeatedly in the past, some branching ruptures of 1968 broke late Holocene strata for the first time. One fracture broke sediments 1,650:250 years old (W—2454) for the first time in 1968 (fig. 89), yet it joined two fractures—one only 50 m distant— that show evidence of repeated movement during the last 1,650 years. Elsewhere, several rup- tures branching from the central and south breaks appear to have breached the youngest lake bed for the first time. These new fractures outside the main fracture zone, together with the diffuse fractures shown in the other trenches and cuts (figs. 83, 88, '89), contribute to the drag component of displace- ment adjacent to the main fault breaks and decrease the slip component. New breaks like these may reflect a temporary or permanent shift in position of a fracture zone, and such shifts in the past would decrease the number of tectonic events apparent in the main fracture zone. Although fault breaks tend to remain in the same places during successive local earthquakes, new breaks do form and must be recognized as a hazard, even near a fault with a demonstrated history of repeated movement along preexisting traces. 128 DEPTH (METERS) THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 Owerchannd. — \Bottom of wash l l I I 0 I I 2 '3 4 HOP/ZOIV TAL D/ST/l/VC‘E (METERS) EXPLANATION Sin and eand Eol/‘an and flu V/‘0//?/ ZZZZE Fill along animal burrow, root, or possible fracture Cons/52‘s 0f mafer/a/ from you/2965f /0A’e bed and Me nexf 0/der bed ‘ Fill in upper channel Cone/ere 0f raid/ed p/eces 0f younges/ /0/re bed /’/7 mbfr/X of mafer/‘d/ from nexf Mo o/der beds Layer A Wh/‘fe 5/7e//y ca/careous s/b‘. Younges/ depos/‘f of Lake Cebu/d0 Crossbedded to thinly bedded silt to coarse sand Con/a/ns some she/L9; probdb/y lacusfr/ne Fracture Dashed where Mferred W— 2469 [:1 (24501250) Radiometrically dated sample (See table l4) Numbers /’/7 parenfbeses lhd/C‘O/é’ 6/4 age, m years before presenf DEPTH (METERS? THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 l l I l O | 2 3 4 HOP/ZONZ'AL D/STA/VCE (METERS) EXPLANATION Clay and rewor e calcareous silt F /aw'a/ and eo/fan/W Layer A Wh/‘Ie she/0x ca/careous sf/t Younges/ deposff of La/re Ca/uW/a Light-groy-green lime-rich silt containing abundant mollusks Crossbedded to massive clay, silt, and sand containing scattered mollusks Fr 0 c t u r e Dashed where mfe rrea’ |’_‘l W-2454 (/6501250/ Radiometrically dated sample (see table l4) Numbers /'n parenf/Ieses ind/cafe 6‘" age, #1 years before presenl 129 REFERENCES CITED Allen, C. R., St. Amand, Pierre, Richter, C. F., and Nordquist, J. M., 1965, Relationship between seismicity and geologic structure in the southern California region: Seismol. Soc. America Bu11., v. 55, no. 4, p. 753—797. Bolton, H. E., 1930, Anza’s California expeditions, v. 1, An outpost of empire: Berkeley, Univ. California Press, 529 p. Danes, J. V., 1907, Das Erdbeben von San Jacinto am 25. December, 1899: Geogr. Gesell. Wien Mitt., nos. 6—7, p. 339—347. Dibblee, T. W., Jr., 1954, Geology of the Imperial Valley region, in Jahns, R. H., ed., Geology of southern Califor- nia; chap. 2, Geology of the natural provinces: Califor- nia Div. Mines and Geology Bull. 170, p. 21—28, pl. 2. Frick, Childs, 1921, Extinct vertebrate faunas of the bad- lands of Bautista Creek and San Timoteo Canyon, southern California: California Univ. Pub., Dept. Geol. Sci. Bull., v. 12, p. 277—424. Hubbs, C. L., and Bien, G. S., 1967, La Jolla natural radio- carbon measurements V: Am. Jour. Sci. Radiocarbon Supp., v. 9, p. 261—294. Hubbs, C. L., Bien, G. S., and Suess, H. E., 1960, La Jolla natural radiocarbon measurements I: Am. Jour. Sci. Radiocarbon Supp, v. 2, p. 197—223. 1963, La Jolla natural radiocarbon measurements III: Am. Jour. Sci. Radiocarbon Supp., v. 5, p. 254—272. —._1965, La Jolla natural radiocarbon measurements IV: Am. Jour. Sci. Radiocarbon Supp, v. 7, p. 66—117. Jennings, C. W., compiler, 1967, Geologic map of California, Olaf P. Jenkins edition, Salton Sea sheet: California Div. Mines and Geology, scale 1 : 250,000. Kamb, Barclay, Silver, L. T., Abrams, M. J., Carter, B. A., Jordan, T. H., and Minster, J. B., 1971, Pattern of fault- ing and nature of fault movement in the San Fernando earthquake, in The San Fernando, California, earth- quake of February 9, 1971: U.S. Geol. Survey Prof. Paper 733, p. 41—54. FIGURE 89.—South wall of trench 4 showing a surface fracture associated with the 1968 earthquake that developed where there had never been a break before. Displacement across this fracture was almost entirely pull apart (30 mm max.), with virtually no lateral or vertical displacement. None of the older fractures shown is as large as that of 1968. The second and third beds from the surface correlate, respectively, with the first and third beds from the surface in trench 1 (fig. 83). Location of trench 4 is shown in figure 80. (From sketches by M. M. Clark and R. V. Sharp.) FIGURE 88. — Pre-1968 fractures and fissure collapse along the Coyote Creek fault as exposed in a natural cut at location 30.7. Although this fracture did not rupture during or after the 1968 earthquake, two nearby strands did. This section shows two pre-1968 episodes of erosion and filling presumably related to collapse along the fracture zone. In both episodes the channels were evidently eroded by water running in or along the frac- tures toward deeper openings. The younger episode is post-Lake Cahuilla, as shown by rotated blocks of the youngest lake bed (layer A) in the fill of the upper channel. The older episode is represented by a shallower channel cut into the lowest deposit evidently during a period when Lake Cahuilla had withdrawn from this area. The fill of the lower channel contains a slump block of the lowest deposit. Eolian and fiuvial(?) mud, silt, and sand have covered the youngest lake deposit at this location since Lake Cahuilla withdrew. 130 Larson, R. L., Menard, H. W., and Smith, S. M., 1968, Gulf of California—A result of ocean floor spreading and transform faulting: Science, v. 161, no. 3843, p. 781—784. Laughlin, Homer, Arnold, Ralph, and Kew, W. S. W., 1923, Southern California earthquake of July 22, 1923: Seismol. Soc. America Bull., v. 13, p. 105—106. Lawson, A. C., and others, 1908, The California earthquake of April 18, 1906, Report of the State Earthquake In- vestigation Commission: Carnegie Inst. Washington Pub. 87, v. 1, 451 p. Mendenhall, W. C., 1909, Ground waters of the Indio region, California, with a sketch of the Colorado Desert: U.S. Geol. Survey Water-Supply Paper 225, 53 p. Myers, W. B., and Hamilton, Warren, 1964, Deformation accompanying the Hebgen Lake earthquake of August 17, 1959: US. Geol. Survey Prof. Paper 435-I, p. 55—98. Rogers, T. H., compiler, 1965, Geologic map of California, Olaf P. Jenkins edition, Santa Ana sheet: California Div. Mines and Geology, scale 1:250,000. Rolfe, F., and Strong, A. M., 1918, The earthquake of April 21, 1918, in the San Jacinto Mountains: Seismol. Soc. America Bull., v. 25, p. 63—67. . Scholz, C. H., and Fitch, T. J., 1969, Strain accumulation along the San Andreas fault: Jour. Geophys. Research, v. 74, no. 27, p. 6649—6666. THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 Sharp, R. V., 1967, San Jacinto fault zone in the Peninsular Ranges of southern California: Geol. Soc. America Bull., v. 78, p. 705—730. - 1972, Map showing recently active breaks along the San J acinto fault zone between the San Bernardino area and Borrego Valley, California: US. Geol. Survey Misc. Geol. Inv. Map I—675. Stanley, G. M., 1963, Prehistoric lakes in Salton Sea basin [abs]: Geol. Soc. America Spec. Paper 73, p. 249—250. 1966, Deformation of Pleistocene Lake Cahuilla shoreline, Salton Sea basin, California [abs]: Geol. Soc. America Spec. Paper 87, p. 165. Strand, R. G., compiler, 1962, Geologic map of California, Olaf P. Jenkins edition, San Diego—E1 Centro sheet: California Div. Mines and Geology, scale 1:250,000. Townley, S. D., 1918, The San Jacinto earthquake of April 21, 1918: Seismol. Soc. America Bull., v. 8, p. 45—62. US. Geological Survey, 1971, Surface faulting, in The San Fernando, California, earthquake of February 9, 1971: US. Geol. Survey Prof. Paper 733, p. 55—76. Wallace, R. E., 1970, Earthquake recurrence intervals on the San Andreas fault: Geol. Soc. America Bull., V. 81, p. 2875—2890. Whitten, C. A., 1956, Crustal movement in California and Nevada: Am. Geophys. Union Trans., v. 37, no. 4, p. 393—398. GEOLOGIC EVIDENCE OF PREVIOUS FAULTING NEAR THE 1968 RUPTURE ON THE COYOTE CREEK FAULT By ROBERT V. SHARP and MALCOLM M. CLARK U.S. GEOLOGICAL SURVEY ABSTRACT Detailed structural mapping along the segment of 1968 rupture on the Coyote Creek fault reveals a complex zone of prior faulting and folding. Past vertical tectonism shown by small upwarps in alluvium and by uplift along the principal strands is consistently in the same sense as that shown by the 1968 movement. Faults of east-west, north- south, and northwest-southeast trend diverge from the main strand of the Coyote Creek fault near Borrego Mountain, but none apparently moved in 1968. En echelon discontinuity of the Coyote Creek fault trace at Ocotillo Badlands and compressional folding evident there may be surface mani- festations of a continuous fault at depth. Structural com- plexity along the 1968 trace apparently increases with the age of the exposed rocks. Beyond the ends of the 1968 break, the Coyote Creek fault probably connects northwestward through concealed branch faults with the main strand near Coyote Mountain, and to the southeast it possibly extends along the southwest margin of Superstition Mountain or perhaps to fresh surficial fractures east of Plaster City. The great width of the tectonically disturbed zone indicates an engineering hazard for structures near this or similar faults. Studies and measurements of regional horizontal strain must take into account the complex horizontal and vertical movements that apparently characterize past tecton- ism in this area. IN TRODUCTION This paper describes the relation of 1968 ruptures to older fault strands in the region of the Coyote Creek fault.1 (See fig. 90.) At several locations along or near the 1968 rupture, excellent exposures reveal clear evidence of earlier movements on the same or V other fault strands Within the zone and show that complex folding in young sediments is closely asso- ciated With some of the faults. Some fault strands can be seen at the surface, but others, in areas where exposures are poorer, are inferred from stratigraphic and structural relations. Our detailed mapping shows that the Coyote Creek fault along the segment of the 1968 rupture is at many places a complex zone con- sisting of multiple strands that are either widely 1Physiographic evidence of earlier faulting is discussed in greater detail by Clark in “Surface Rupture Along the Coyote Creek Fault,” (this volume). divergent from or nearly parallel to the regional trend. Several strands have not been previously rec- ognized. Holocene deformation along the Coyote Creek fault is expressed at the surface by folding and doming of Quaternary strata, as well as by strike-slip fault- ing with local uplift shown by scarps. This deforma- tion is locally distributed across a relatively broad zone, and the width of the disturbed belt implies an engineering hazard for structures built across this or similar fault zones. Although there was no move- ment on many strands near the main Coyote Creek fault during the 1968 earthquake, they should not be considered inactive. NORTH BREAK2 The northern segment of the 1968 break along the Coyote Creek fault marks the structural boundary between the southern Borrego Badlands to the north- east and the Borrego Mountain area to the south- west (pl. 3). The southern Borrego Badlands are eroded in Pleistocene terrestrial deposits that are downdropped and displaced in a right-lateral sense relative to the granitic basement rocks exposed across the fault at Borrego Mountain. The depth to granitic rock on the northeast side of the fault probably ranges from very shallow opposite East Butte to a few kilometers beyond the north end of the 1968 break. (See pl. 3.) South of State Highway 78, the sense of net vertical throw on the fault is reversed from that to the north, as shown by the structural and topographic elevation of Pleistocene and Holocene strata in Ocotillo Badlands northeast of the break. Geomorphic and geologic evidence for the exact position of the trace along the north break prior to 1968 is sparse because of a widespread cover of very recent alluvium, particularly along the 10-km 2The designation of fault strands in this discussion is the same as that used by Clark (“Surface Rupture Along the Coyote Creek Fault," this volume). They are termed the north, central, and south breaks. 131 132 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 VALLEY BORREGO COY 35' OTE BADLANDS " MTNifi BORREGO Borrego Springs . ” :jVALLECITO MTS FIGURE 90. —Regional map of the Coyote Creek fault showing its relation to faults outside the 1968 rupture zone. segment where the fault is followed by the wash of San Felipe Creek. As indicated by the position of the 1968 break northwest of the north end of East Butte of Borrego Mountain, the Coyote Creek fault traverses the nearly featureless bottom of San Felipe Wash for about 2 km and then follows the west side of a low, flat bench formed by Holocene terrace gravels for about 1 km. In the next kilometer northwest, the fault again crosses alluvium of San Felipe Creek and projects directly into a small hill at the edge of Borrego Badlands halfway between Big Wash and Third Wash (pl. 3). This is one of the very few places along the entire length of the 1968 break where Cenozoic deposits were apparently newly broken. A few additional minor cracks occurred along the southwest-facing scarp marking a bound- ary fault at the edge of the badlands for about 2 km north-northwest of this hill. The 1968 rupture at the hill between Big Wash and Third Wash apparently did not coincide with the traces of older faulting at that location. Detailed mapping shows that a fault continues for probably at least 2 km north-northwest into the Borrego Bad- lands, nearly on the projected trend of the 1968 break south of Third Wash. (See pl. 3; fig. 91.) The fault that borders the southwest side of Borrego Badlands converges east-southeastward with this fault near the hill just mentioned. Apparently, the 1968 rupture pre-Cenozoic at the hill is a new break that approximately bisects the angle between the previously established branch faults. No evidence of earlier movement on projec- tion from the 1968 rupture was found either on the hill or farther north. The positions of ‘similar fold axes on opposing sides of the fault extending into Borrego Badlands suggest a small right-lateral offset (fig. 91). Al- though the gross structure throughout the southern Borrego Badlands is a north-dipping homocline, small superimposed folds define unique domains of anomalous southward dip on each side of the fault in southeastern sec. 13, T. 11 S., R. 7 E. The traces of bedding in the fault surface as projected from each side of the fault are nearly identical, supporting the interpretation that the folds are correlative and older than the offset. The amount of horizontal shift is about one-third kilometer; the amount and sense of vertical separation are unknown. Exactly how far this fault extends northwestward into the Borrego Badlands is uncertain, but it does not simply cross the badlands and join one of the main fractures of the San J acinto zone near Coyote Mountain (fig. 90), as might be expected from the distribution of aftershocks of the 1968 earthquake in this area. (See Hamilton, this volume.) Exposures of sheared sediments are visible where the fault intersects Big Wash, but the fault is covered by ter- race deposits for at least the next kilometer to the THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 133 HARPERS . fiSUPERSTITION :5»; WELL ‘ \<\O“ 4M ..... 6093??“ V I": ..: _________ SUPERSTITION Area of plate 3 MTN , IMPERIAL VALLEY SOUTH BREAK ....... $ Dixieland |969 rupture~> O 5 l0 KM -:-:-::: Unstippled areas show Holocene alluvium; stippled areas show Pleistocene and older Cenozoic deposits and crystalline rocks. northwest. Continuous badland exposures of, the Palm Spring Formation (Sharp, “Tectonic Setting of the Salton Trough,” this volume) even farther northwest appear to be unfaulted on large-scale aerial photographs, implying either that there is no displacement or possibly that the trace has swung to a slightly more westerly trend and transects the section parallel to the strike of bedding. Although the second possibility has not been investigated fully in the field, such an extension of the fault could at best tie irregularly into the main Coyote Creek fault strand south of Coyote Mountain (fig. 90). More- over, because intervening parts of the Borrego Bad- lands are definitely unfaulted, this fault does not connect northward at the surface with the fault bounding the east side of Coyote Mountain. Although the pre-1968 branch lying within the Borrego Badlands may be- an active fault, several lines of evidence suggest that the most active strand of the Coyote Creek fault is now the branch fault lying along the southwest boundary of the badlands. This segment of the fault ruptured at three places in 1968, and for 1 km west of the branching point, relatively fresh-looking southwest-facing scarps lie along this strand. In contrast, no geomorphic fea- tures of faulting are preserved along the branch extending into the badlands, although a crush zone in the Pliocene and Pleistocene Palm Spring Forma- tion is exposed at Big Wash. Furthermore, the 476-246 0 - ’72 - 10 branch along the badlands’ boundary lies nearly along the most direct route between the 1968 break and. the strand of the Coyote Creek fault that shows evidence of Holocene movement west of Borrego Badlands. (See fig. 90, and Sharp, 1972.) The Coyote Creek fault along the west side of Borrego Badlands is expressed at the surface only by two rounded northeast-facing linear scarps in alluvium about 2.5 km north of Borrego Sink Wash (pl. 3). Northwest from these scarps as far as Coy- ote Mountain (fig. 90), the trace is not visible, but the fault is marked by a very steep gravity gradient reflecting a concealed step in the basement rocks (Biehler and others, 1964, chart 1). Although its trace is concealed by apparently unfaulted alluvium southeast from the same scarps, the fault defines a structural boundary between nearly homoclinally dipping sediments in Borrego Badlands and more intricately folded but otherwise similar beds lying southwest of the fault. Because this strand of the Coyote Creek .fault apparently does not directly cross the wash of San Felipe Creek, continuity of faulting to the southeast probably is achieved by an eastward bend or branch of the fault that connects with the northernmost 1968 ruptures. Southeast from the north end of East Butte to the settlement of Ocotillo Wells, the precise location of the currently active strand of the Coyote Creek fault is known only from the trace of rupturing in 134 EXPLANATION . tr >_ t LL] 3 1s) t—E ° 8 ~ g2 3: Alluwum 0 CC)" x m 0% ° = C] >2 m 2 mu: 5) e . SE \ § ,2 Palm Sprung E4 '\ W ormation LuD m a F ._o m Fault Zigzag segmenis bro/re in L968 Arrows Mei/cafe rev/alive horizon/(fl movement —t—>-¥* Anticline Syncline Fold axes, showing plunge + Generalized strike and dip of beds k\\\V Domain in which dips are in southeast quadrant In other areas dips are in nor/hens! or northwes/ quadran/s FIGURE 91.—Simplified geologic map of the area at the north end of 1968 Coyote Creek fault ruptures. Area in this figure is shown on plate 3. 1968. Extensive alluvial cover and the absence of surficial features of Holocene faulting in the allu- vium prevent close delineation of the fault by geo- logic or physiographic means. Geologic evidence at Borrego Mountain and in the hills east and northeast of Benson Dry Lake demon- strates that several widely divergent branch faults exist along the north break of the Coyote Creek fault. Granitic rocks and overlying beds referred to the Palm Spring Formation (Sharp, “Tectonic Setting THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 of the Salton Trough,” fig. 2, this volume) exposed at East Butte and West Butte of Borrego Mountain (Dibblee, 1954, pl. 2) are broken by several south- trending faults, some of which display small vertical components of offset. There may have been lateral components of displacement on these fractures, but no clear evidence supports strike slip. In contrast, on the east side of the Coyote Creek fault, locally tightly folded and faulted strata of the Imperial Formation, Palm Spring Formation, and Ocotillo Conglomerate of Dibblee (1954) (Sharp, “Tectonic Setting of the Salton Trough,” fig. 2, this volume) define a large compressional zone with east-west structural grain. At least two major east-trending faults, one bounding Squaw Peak on the south and the other underlying San Felipe Wash where it turns abruptly eastward, are required to explain convinc- ingly the distribution of the basement rocks and distinctive Cenozoic strata] units. (See fig. 92.) The presence of a fault at Squaw Peak is indicated by the abutment of south-dipping beds of the Palm Spring Formation against the crystalline basement along the south edge of the hills; this relation could be explained by a depositional contact on the south limb of an anticline centered on the axis of the bills, but uniformity of gneissic structure in the basement rocks indicates that such anticlinal arching has not occurred. The southern east-west fracture is shown by stratigraphic relations—in particular, by the omission of the Ocotillo Conglomerate from the sec— tion exposed on the north side of San Felipe Creek. The post-Pleistocene compression indicated by the folding subparallel to the traces of both these faults suggests that they may be reverse or thrust faults. The regional gravity map of Biehler, Kovach, and Allen (1964) depicts east-west alinement of gradi- ents in this area and establishes that the structural trends evident at the surface are reflected in the configuration of the basement at depth. OCOTILLO BADLANDS SECTION OF THE COYOTE CREEK FAULT The north and central breaks of the 1968 rupture are disposed in a left-stepping en echelon arrange- ment along the margins of Ocotillo Badlands (pl. 1; fig. 93). Tarbet and Holman’s (1944) Borrego For- mation and Dibblee’s (1954) Ocotillo Conglomerate exposed in the badlands almost certainly were folded complexly and bulged upward in response to the same tectonic forces that produced the boundary breaks, and the deformation along the faults and on the folds dispersed between them is considered in this report to be interrelated. THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 135 o of» J KILOMETER 200/ <.»— _ _ . ”(M-2’ I v?" I: fi%/j/Z¢W .. . V . ///"x//"l SQ. -— > wtl’ltlltl._, .' I I _ / ....---.---v’° ”'5'...“ n0, "\°,.° lo .- Inc's... vs. \_ I‘- ,_ \:/——::S\a-1;F£lépe—_CCL _ T, “5., .1; Qq I46 . _..-—.’3"J - . 3—... [58 T. IZS L . . sq ...... //x,, . 4% /////,. ////////, . , v . z. ~ . M /’///////// a, //.- . Ba e from US. Geological ‘Survey l'-24,000 Geology by R. V. Sharp, I969 Shell Reef, Borrego Mountain, I960 EXPLANATION \ Qa Alluvium Contact 7/” Older alluvium Fault Dashed where approximate/y /ocofea’,- (Io/fed where concea/ed Pleistocene or V QUATER N A RY 0f Dibblee U954) ’ Anlicline Syncline P/iacene P/e/‘sfocene Ho/acene Ho/ocene ‘Ocotillo Conglomerate >. g Fold axes, showing plunge ,: Doffed where concea/ed 0: Imperial Formation ‘,-'_-' _l"_ \\\\\§ Sf ‘k d d' fbeds m n e an up a Gneissic crystalline rocks PRE-TER- TIARY FIGURE 92.—Geologic map of the area near Squaw Peak. (See pl. 3 for location of this map.) 136 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 a; I w; vi /// il ' ‘ Little , i ism Bul'egn :- ir ,7 l' ' u ' [Ii/I” '1 _.; ‘1 - 1 .14 1 / #1/ ., in. ., _ in” ‘tos Pom //'- l.O KILOMETERS 50 fl ,’ ,7 , View“ 1,1 a. ., i - ‘I ro "Se \ . 4..., , :13 £3. ’3'“ 73 ref 2* , 5\11 26 ,1, 41 \/4/ so W'l/I . . *’ iwnx If" - - "II/l" ' . ~ 4/8 ; \ ‘~~ 4 7‘37“ U;- "I // . 4. ‘ ' . .fi‘: ' Base from US. Geological Survey |:24,000 Borrego Mln. S.E., 1958‘, Shell Reef, I959 filo/acme f/a/acehe P/e/s/ocehe or P/e/s/a cede Alluvium \V \ Older olluvium 0°C 1' Ocofillo Conglomerate of Dibblee (I954) Borrego Formation of Torbel and Holmon (I944) EX PLANATION QUATERNARY FIGURE 93. — Geologic map of the Ocotillo Badlands. Geology by MM. Clark and RV. Sharp, l969 C o n to of Dashed where araa’ah’ana/ ...._7.._?--... . Foull Dashed where approxxmafe/y /oca/ed,~ auer/ed where therred; def/ed where concea/ed <__I____ ..... +_§____...... Anticline Syncline Fold oxes, showing plunge Dashed where appraX/Ma/e/y /acafed; defied where concealed <—-Pr-—- Overturned syncline, showing direction of dip of limbs 20 Strike and dip of beds (See pl. 3 for location of this map.) THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 Whether the faults bounding the Ocotillo Badlands project beyond those segments that broke at the time of the 1968 earthquake (the north and central breaks) is unknown. Although the north break ter- minated as a coherent line of displacement near the northwest corner of the badlands, the linear form of the west margin of the hills suggests that the fault may continue southeastward and may intersect the central break. Abundant geologic evidence demon- strates the continuity of the central break along nearly all the northeast boundary of the hills. The possibility that the central break extends beyond State Highway 78 at least to Benson Lake is sug- gested by the alinement of alluvium-filled passes between low hills covered with terrace gravels, but no other evidence for a continuation of the fault is known. Geologic structure in the Pleistocene Borrego For- mation and Ocotillo Conglomerate exposed in the Ocotillo Badlands signifies compressional tectonics (fig. 93). Because bedding in these formations and in older alluvium dips away from the hills around the periphery of the badlands, the gross geologic structure is a closed anticlinal upwarp. However, the detailed structure is extremely complex, involv- ing a large number of tightly folded anticlines and synclines that have essentially an east-west orienta- tion. The folds are relatively open and widely spaced near the west boundary of the hills, where they are exposed in the Ocotillo Conglomerate, but they be- come tightly appressed and closely spaced eastward in the underlying Borrego Formation. In the central part of the badlands, fold axes are generally straight, but many folds are abruptly terminated either by convergence and cancellation of anticlines by syn- clines, by offset along minor faults, or by a combi- nation of both mechanisms. In addition, extreme attenuation in the limbs of nearly isoclinal folds locally has destroyed bedding and reduced originally competent sand beds to scattered phacoidal bodies in clay or silt matrix. Near their eastern extremities, , the fold axes curve and become parallel with the trace of the central break, suggesting right-lateral drag along the fault. In addition, dips increase to nearly vertical within a few tens of meters of the break, perhaps as a result of drag from the vertical component of displacement. The geometry of the folds on the southwest side of the badlands is more varied than to the cast, but the nonuniformity probably is partly related to the age of the exposed beds. Fold axes in the Borrego For— mation trend southwest before dying out or turning northwest near the southwest margin of the bad- 137 lands. Although the folds are not nearly as tight or abundant on the southwest as they are on the north- east, where present their relation to the probable southwestern boundary fault is similar to that of the northeastern folds to the northeastern boundary fault. In contrast, fold axes in the Ocotillo Conglom- erate overlying the Borrego Formation at the south end of the badlands bend southwestward and die out in southwest-dipping homoclinal beds. Moreover, at some places in the badlands, bedding in the Ocotillo Conglomerate is strongly discordant to structure in the underlying Borrego Formation. (See fig. 93.) Thus, the structural simplicity apparent along parts of the southwest edge of the badlands may belie more pronounced but concealed folding at depth that may closely resemble the complex structural effects near the northeastern boundary fault. The tightness of folding probably reflects principally the age of the beds involved, but it is also possible that depth of burial and lithology affected the structural develop- ment to some degree. Another structural complexity at the central break in Ocotillo Badlands, in addition to the apparent drag effects, is the presence of several small closed anti- clines and synclines elongate parallel to the fault. These closed structures are less than several hundred meters long and lie next to the fault. The fact that they represent both doming and subsidence of strata implies that they resulted from differential vertical movements of material below the surface adjacent to the fault. Restriction of the closed anticlines and synclines to the region of apparent drag effects sug- gests that they were created by flowage of buried strata of clay and silt adjacent to the fault in reac- tion to local drag effects. As already noted, tight folds exposed in the center of Ocotillo Badlands reveal widespread flowage of such strata.3 Deformation in the Ocotillo Badlands is almost certainly a manifestation of continuity of distributed right-lateral displacement transferred from the north to the central break. Surficial material that is deformed in a right-lateral sense across two breaks with these orientations must theoretically be short~ ened in the north-south dimension and extended 3Closed anticlines and synclines are even more striking 25 km to the south- east in the uplifted block adjacent to the Superstition Hills fault. Inspection of aerial photographs and brief field investigation show that the upthrown block (the Superstition Hills) consists of a complex sequence of closed, open, and faulted anticlines and synclines in poorly consolidated sediments similar to those of Ocotillo Badlands. The number and complexity of these folded structures increase toward the fault, although the Superstition Hills fault does not everywhere follow a course that is completely obvious from local structure. Future detailed mapping and analysis of the well-exposed structure of Superstition Hills may lead to answers concerning the origin of such small closed folds and also to a better understanding of their relation to the faults associated with them. 138 either vertically or in the east-west direction, or both. In this area, there is obvious vertical extension. At great depth, however, it is probable that both surficial breaks converge into a single continuous shear zone, perhaps in the manner shown in figure 94. At depth, such a continuous fault surface would not require the type of strong deformation in the adjoining materials that is now present at the sur- face in Ocotillo Badlands. CENTRAL BREAK SOUTHEAST OF OCOTILLO BADLANDS Structural expression of the central break south- east of Ocotillo Badlands is simple and is entirely lacking in the southeasternmost part. (See Clark, “Surface Rupture Along the Coyote Creek Fault,” this volume.) Briefly, a low scarp (southwest side up) extends along the central break for nearly 4 km southeast from Ocotillo Badlands. This part of the central break also acts as a barrier to ground water, limiting abundant growth of mesquite (Prosopis juli- flora) to the southwest (upthrown) side of the fault. North of Old Kane Spring Road, a secondary break diverges eastward from the scarp, creating a graben between the two breaks. Near the southeast end of this scarp, the sense of vertical uplift is locally reversed, and domed strata are exposed for about Position of »Ocoiil|o Badlands FIGURE 94.—Three-dimensional view of possible configura- tion for Coyote Creek fault beneath Ocotillo Badlands. Distal ends of volume of strongly compressed sediments (stippled) shown by dashed and dotted lines. THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 100 m along the northeast side of the break. This small closed anticline resembles those in Ocotillo Badlands described previously but is unusual in its isolated setting and small size. Southeast of the low scarp the central break has no structural expression at the surface. The scarp, graben, and localized doming resulted from a consistent but small vertical component of displacement associated with the dominant strike slip. These surficial manifestations of faulting are visible today only because the rate of deposition on this alluvial plain (and former lakebed) is less than the rate of tectonic uplift. AREA EAST OF THE CENTRAL BREAK Several of the 1968 breaks extend along isolated low hills east of the central break. The largest of these hills are as much as 1 km long and consist of upwarped and eroded sediments, commonly in the form of closed anticlines with moderately dipping strata that are easily recognized on vertical aerial photographs; others are covered by unusually large mesquite-stabilized dunes. None is more than 10 m high. Although all such hills east of the central break as far as a closed anticline 2 km south of Harpers Well (fig. 90) were checked in June 1968, fractures were associated with only two: the large upwarp 1 km east of the central break and immediately south of Old Kane Spring Road, and a prominent hill 5 km farther southeast and about 3 km east of the main break. The fracture along the first hill was clearly associated with renewed uplift in 1968, as shown by the sense of vertical displacement (pl. 1). Unfortu- nately, blown sand obliterated most of the fractures and all evidence of relative displacement along the second hill. However, the surviving fractures lie along the base of the scarp that bounds one side, suggesting renewed uplift of this hill also in 1968. Because the fractures were probably more extensive originally, it is possible that fracturing occurred during the earthquake or later on other hills within a few kilometers to the northwest and southeast. The conclusion seems inescapable that these low hills of upwarped sediments are growing as a result of present tectonic activity in this area and that those crossed or bounded by fresh breaks were given additional uplift in 1968. Furthermore, the hills along Old Kane Spring Road and those to the south- west lie approximately on the projection of the Superstition Hills fault, and they may mark a con- nection between the fracture branching from the central break at Old Kane Spring Road and the Superstition Hills fault. THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 HILLS SOUTHWEST OF OCOTILLO BADLANDS In contrast to the upwarped and gently folded strata that lie to the east of the central break, tightly folded and locally vertical strata are exposed in the linear hills that lie about 1 km southwest of Ocotillo Badlands. Small ruptures were found in the central part of these hills after the earthquake. Those frac- tures that traversed exposed strata followed older structural trends, such as offset strata or nearly vertical bedding. However, no new fractures ex- tended to the southeast through the steeply dipping strata of the southern part of the hills, although this area appears to be an anticline with an axial fault. Strata 0n the flanks of these elongate hills dip gently to moderately away from the axis but become ver— tical at the axis. By analogy with structure along the northeast margin of Ocotillo Badlands where the 1968 rupture followed a narrow zone of vertically dipping beds, this zone of vertical strata is thought to contain a fault, although its exact position is uncertain. SOUTH BREAK In contrast to the north and central breaks, almost all the south break lacks obvious geologic clues at the surface that indicate the presence of the under- lying fault. Vegetation, however, locally defines the position of the active trace, and both the geology and the distribution of vegetation to the southeast sug- gest connections with other traces. The south break lacks significant vertical displacement and occupies an area of active deposition; as a result, no scarps are visible, and recurrent burial by sediment con- ceals structural complexity caused by faulting. A channel, however, that crosses the break about 0.7 km east of R. 9 E. reveals the fault and evidence of repeated activity during the past 2,400 years. (See Clark and others, this volume.) In addition, lines of creosote bushes and shallow sinks along the south break mark several relict tectonic fractures that formed perhaps 20 years before the earthquake. (See Clark, “Surface Rupture Along the Coyote Creek Fault,” this volume.) Like segments of the central break, parts of the south break act as a barrier to ground water, encouraging abundant growth of mesquite to the southwest of the fault. (See Clark, preceding reference.) Alined mesquite- covered dunes, which may mark the course of pre- vious surface fractures, and low hill‘s of upwarped alluvium occur along the southeastward projection of the south break and imply a connection between the 1968 break and the colinear fault that forms the southwest boundary of Superstition Mountain. Part of this projected segment lies exactly over a steep gravity gradient interpreted by Biehler, Kovach, and 139 Allen (1964) to represent a concealed fault. The faults of Superstition Mountain and the low hills between Superstition Mountain and the south break were inspected after the earthquake, but no indica- tion of recent breakage was evident. The south break and its projection along the south- west boundary of Superstition Mountain trend nearly 20° E. of the S. 40° E. average trend of the Coyote Creek fault to the north. According to interpretation of gravity data by Biehler, Kovach, and Allen (1964), faulting along this S. 60° E. trend probably also continues in the opposite direction (northwest- Ward) west of the 1968 rupture toward the front of the Vallecito Mountains. Alternatively, the principal strand of the Coyote Creek fault may not continue southeastward to Superstition Mountain but instead extends S. 40° E. from the south break across the nearly featureless alluvial flats that lie to the west of Superstition Mountain. A reconnaissance of this projected trend in December 1970 revealed no evi- dence of present or past fracturing for a distance of nearly 30 km from the south break; however, an extensive zone of fresh fractures of possible tectonic origin crosses Highway S80 along the projection about 1.5 km west of Dixieland (fig. 90). SUMMARY Geologic mapping along the Coyote Creek fault demonstrates various relations of Quaternary struc- ture to the 1968 surface rupture: 1. Bending of fold axes, increase of dips to nearly vertical, and local doming or downwarping of Quaternary strata in the immediate vicinity of the 1968 break give strong evidence of drag effects. These structural clues to tectonic activ— ity and other geologic evidence revealed addi- tional recently active branches and nearby faults in Quaternary deposits along the 1968 break, many of which were inactive during the Borrego Mountain earthquake. 2. Tectonically upwarped hills of alluvial sediment as much as 3 km from the main 1968 break were apparently uplifted slightly in 1968. 3. The sense of the vertical component of very re- cent offset changes at several locations along the fault yet is always consistent with all evi- dence that shows the sense of earlier move- ment. 4. Structural complexity along the Coyote Creek fault is greatest in the regions of the north and central breaks, where the oldest sediments and rocks are exposed. Young alluvium and lake deposits may mask complex branch fault- ing and folding that affect the underlying but 140 possibly only slightly older sediments at many locations. 5. The principal fracture of the Coyote Creek fault in the area of 1968 breakage is probably con- nected with the unbroken main strand farther northwest by a fracture that is mostly con- cealed beneath the alluvium of San Felipe Creek. Subsidiary faults along the north break diverge from the main fault strand along nearly east-west lines northeast of the fault and along north-south lines on the southwest side. 6. The Coyote Creek fault possibly extends beyond the 1968 breakage east-southeast along isolated uplifted mounds of sediments to the fault that bounds Superstition Mountain on the south- west, although it may also extend southeast to connect with fresh fractures near Highway 880, nearly 30 km distant. 7. Structural features of the strongly folded and uplifted Cenozoic sediments of Ocotillo Bad- lands indicate a zone of compression between the en echelon overlap of the north and central breaks. The en echelon pattern, however, may be the surficial expression of a continuous fault at depth. The pattern of fault strands established by our structural mapping, together with the ruptures formed by the 1968 earthquake, probably only partly reveal the complexity of the Coyote Creek fault. Although many recognizable branch faults diverge from the main zone at small and large angles near Borrego Mountain, other undetected strands and branches of the Coyote Creek fault may also exist, particularly along the south break and parts of the central break that traverse very young Holocene deposits. The definite nature and extent of complex branch faults even in some areas of well-exposed Cenozoic beds may not be revealed until movement occurs again or until extensive exploration is under- taken. The recently detected aseismic ground fractur- ing near Dixieland (mentioned previously) may prove to be widespread and prevalent; its documen- tation will certainly supplement and will perhaps be THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 a critical part of the information about geologic structure in this region. Our investigation emphasizes that the area of potential damage to structures extends far beyond the narrow zone of the main rupture of 1968 and includes outlying low hills of upwarped sediment, splaying faults, and the southeastward and north- westward projection of 1968 ruptures. Any proposed site for a major building in the region of the Coyote Creek fault should be examined carefully below the surface for evidence of Holocene or Pleistocene displacement. The clear evidence of repeated dis— placement in the past along the 1968 rupture (Clark and others, this volume) suggests that movement can occur again on any trace of the Coyote Creek fault with a record of late Pleistocene or Holocene offset. This investigation also demonstrates that studies of strain in this region must consider the complex nature of outlying fractures and the widespread occurrence of significant vertical components of movement in the region of the main fault. Attempts to measure regional strain for a simple strike-slip model could fail if stations move vertically in reac- tion to regional horizontal strain (for example, in Ocotillo Badlands) or if stations undergo complex movement because they are placed in blocks bounded by parallel or splaying faults. ' REFERENCES CITED Biehler, Shawn, Kovach, R. L., and Allen, C. R., 1964, Geophysical framework of northern end of Gulf of California structural province, in van Andel, T. H., and Shor, G. G., Jr., eds., Marine geology of the Gulf of California: Am. Assoc. Petroleum Geologists Mem. 3, p. 126—143. Dibblee, T. W., Jr., 1954, Geology of the Imperial Valley region, California, in Jahns, R. H., ed., Geology of south— ern California; chap. 2, Geology of the natural provinces: California Div. Mines Bull. 170, p. 21—28. Sharp, R. V., 1972, Map showing recently active breaks along the San Jacinto fault zone between the San Bernardino area and Borrego Valley, California: US. Geol. Survey Misc. Geol. Inv. Map I—675. Tarbet, L. A., and Holman, W. H., 1944, Stratigraphy and micropaleontology of the west side of Imperial Valley, California [abs.]: Am. Assoc. Petroleum Geologists Bull., v. 28, p. 1781—1782. INTENSITY DISTRIBUTION AND FIELD EFFECTS, STRONG-MOTION SEISMOGRAPH RECORDS, AND RESPONSE SPECTRA By SEISMOLOGICAL FIELD SURVEY, NATIONAL OCEANIC AND ATMOSPHERIC ADMINISTRATION ABSTRACT The Borrego Mountain earthquake was widely felt through- out southern California and neighboring parts of Arizona and Nevada, but damage was generally light owing to the sparse population and lack of structural development. A maximum intensity of VII was assigned to a small area coinciding with the zone of surface breaks along the Coyote Creek fault in the vicinity of Ocotillo Wells. Strong-motion seismograph records obtained between 41 and 240 miles of the epicenter reveal the following: (1) The peak values for acceleration, 0.14 g, and displacement, 7.25 cm, were recorded at El Centro. (2) Accelerations recorded at sites on unconsolidated materials were higher than at sites on or near bedrock. (3) Displacement for the first 200 miles was approximately defined by the equation log A= 2.32‘0.011 d (A=maximum single displacement; d=epi- central distance). (4) Large seismoscope amplitudes oriented at a considerable angle to the trend of faulting were recorded at El Centro. (5) ‘Seismoscope records at earthfill dams indicated an amplification factor for the response of the dams compared to nearby bedrock. (6) A better distribution of strong-motion seismographs would be desirable, partic— ularly along the active fault zones and at free-field sites in the Los Angeles area. (7) Relative velocity response spectra for the El Centro, San Diego, and San Onofre records show the usual rapid decay with distance from the epicenter. With no accelerometer closer than 40 miles from the epicenter, the vertical components recorded are not as important as the horizontal components. INTRODUCTION The magnitude 6.4 Borrego Mountain earthquake of April 9, 1968 (G.m.t.) was generally felt over an epicentral radius of 150 miles, and although signifi- cant damage was largely restricted to 900 square miles surrounding the region between Ocotillo Wells and Borrego Springs, minor damage occurred in the more populous sections of the Imperial Valley and in the San Diego, Los Angeles, and San Bernardino— Riverside areas. Reports from the Borrego Springs— Ocotillo Wells area show that broken pipes, minor building cracking, and glass breakage comprised the bulk of vibration-related damage effects, apart from the actual faulting that disrupted State Highway 78 and part of the local airstrip. Strong-motion seismograph records were obtained from 114 accelerographs operated by the Seismologi- cal Field Survey1 in California and Nevada, and although relative accelerations were high in only one location, El Centro, the results are particularly valuable for several reasons: (1) The distribution of a large number of records may suggest local and regional geologic influences on strong ground shak- ing. (2) The first significant earthquake motion to be measured at a nuclear generating plant was recorded at the San Onofre reactor site. (3) The geographic pattern of instrument location and opera- tion could provide a basis for future planning of the strong-motion network. (4) A large number of recordings were obtained in the upper stories of buildings in Los Angeles. These reflect the periods of the structures as excited by seismic forces. In addition to the acceleration data, records were also obtained from 17 displacement meters and at least nine seismoscopes. Although more than 130 seismoscopes were located in the felt area, recorded amplitudes were generally too small to definitely as- sign them to the Borrego Mountain earthquake rather than to previous local tremors that had gone unnoticed. ACKNOWLEDGMENTS The authors thank M. Engle, J. Martine, G. Mur- ray, and E. Etheredge, who retrieved and processed the strong-motion records, and also B. J. Morrill, who carried out the field investigation after the earthquake. Thanks are also extended to C. F. Knud- son and V. Perez, who reviewed the section on strong-motion seismograph records. 1Formerly a unit of the US. Coast and Geodetic Survey. 141 142 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 120' 118' 116' 114' 112' 37. \\ I :- A. x ‘1 \ , \2 N E V A D A l, \\ _ .\ I ‘t ' n-LAS “GAS / \ .\\ . . \ \ \ P I3?” 2 5 (Q), \ _ mm... 0‘? ~. < 35 CALIFORNIA .s (9‘10 ° .- ARIZONA p40 / '°/ v . '5. . V'. Z. ' .i' . (\ &" V . Vli°'~*‘[‘) . .. 33-—~‘ APRIL 3,. 1968,18:28:58.9 EST. 0 20 4o 60 80 Too g4__L_I__;L_L_I_I__L_l STATUTE MILES l l l 36' 34' 121' 119' 117' 115" 113' FIGURE 95. —Isoseismal map of the Borrego Mountain earthquake, showing earthquake intensity as measured on the Modified Mercalli scale. INTENSITY DISTRIBUTION AND FIELD EFFECTS By WILLIAM K. CLOUD SEISMOGRAPH STATION, UNIVERSITY OF CALIFORNIA NINA [ED SCOTT SEISMOLOGICAL FIELD SURVEY, NATIONAL OCEANIc AND ATMOSPHERIC ADMINISTRATION A review of reports from an extensive question- naire canvass by the Seismological Field Survey indicates that the generally felt area of the Borrego Mountain earthquake in the United States was ap- proximately 60,000 square miles (fig. 95). Limits of the generally felt area were from Santa Barbara, northeast to Tehachapi, China Lake, Trona, and Tecopa; in Arizona, southeast to Wikieup, and south to Wenden, Horn, and Dateland. Outside the gen- erally felt area, the shock was reported to be felt slightly at Santa Maria, Fresno, Yosemite Valley, and at Las Vegas. Intensity VII was assigned to a small area in northeastern San Diego County, prin- cipally in the Borrego Mountain—Ocotillo Wells area (Coffman and Cloud, 1970). By contrast, Youd and Castle (1970) and Castle and Youd (this volume) report maximum intensities as high as IX, although these were based on ground effects only. Within the intensity VII area, minor right-lateral displacement on the Coyote Creek fault was observed, and Highway 78 was cracked adjacent to Ocotillo Wells. Park rangers reported that rockslides oc- curred in Palm Canyon, at Split Mountain, and at Font’s Point. Huge boulders blocked the Montezuma- Borrego Highway, from the head of Culp Canyon (about 3 miles southwest of Borrego Springs) to near the San Diego County Sheriff’s office. Press reports relating to effects at Ocotillo Wells stated that at the Walter Morton home, walls split over doorways and at room corners, abedroom separated from the rest of the house, dishes and glassware flew out of cupboards, and 3,600 gallons of water gushed out of a storage tank. At the Desert Ironwood Motel, about 3 miles west of Ocotillo Wells, the manager reported loud earth noises, tile block cracked, water pipes in building broke, water from swimming pool flooded the motel, and groundwaves moved toward the north. The manager also reported THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 feeling a light tremor about 45 seconds before the main shock. At the M. A. Smith residence, across Highway 78 from the Desert Ironwood Motel, a schoolbus moved 3 inches, furniture moved 4 inches, and the pumice-block building was slightly damaged. At the Borrego Air Ranch, about 10 miles northwest of Ocotillo Wells and west of Borrego Mountain, well water became dark, a pipeline cracked, and a swim- ming pool lost 11/2 feet of water, but there was no damage to the concrete-block and steel Class “A” building. At the San Felipe Substation, about 3.2 miles south-southeast of Ocotillo Wells, large trans- formers were shifted, shearing anchor bolts and breaking X-bracing. Reported effects of the earthquake at sample loca- tions outside the intensity VII area were: Brea (Orange County). Canned goods fell from shelves. Hanging objects swung violently northwest- southeast. Chula Vista (San Diego County). Strong rolling motion. Power out in 4-square-mile area. Many gro- cery shelves emptied. Desert Center (50 miles northeast of the epicen- ter; eastern Riverside County). “Waves 6 inches high on swimming pool and about 100 gallons of water splashed out on northwest-southeast sides. Movement of earth was a slow roll and lasted about 30 seconds.” Hemet (Riverside County). “Impossible to walk in trailer during shock. Trailer movement was vio- lent. Water in swimming pool was sloshing over the sides 10 minutes after the shock was apparently over.” Coachella (Riverside County). “Most damage was to grocery and liquor store stocks in this area; a great deal of breakage.” Imperial County: Power disruptions occurred in some sections of Imperial Valley. Brawley (8 miles west of Wieman Ranch). “1 have experienced all shocks here in the last 45 years. This one was the most violent of all but its vibra- tions were mostly vertical and therefore it did not topple bookcases, china closets, etc., as in previous shocks. Damage was negligible.” Calexico. (Press) Part of the ceiling fell in Safeway store. El Centro. (Press) Part of a ceiling fell at the J. C. Penney store in Valley Plaza. The Balboa Hotel, which was battered by the 1940 earthquake, was damaged again. Plaster fell from walls and ceilings on the hotel’s second floor. Merchandise fell in stores. Imperial. Damage at Imperial was reported as generally light. About 7,500 books fell from shelves 143 at the public library. About 6 miles west of Imperial, on the Worthington Road grade leading down to New River, there was a crack in the road about 200 feet long and 2 inches wide. “Shock lasted nearly a minute. Sensation of standing in small boat in choppy water. Walking would have been difficult. Feeling of nausea persisted for 15—20 minutes following the shock. Lights were off for 10—15 minutes after the shock.” Westmorland. (Press) Top part of brick wall collapsed at a laundromat. Walls cracked in other buildings. Long Beach (Los Angeles County). Press re- ported the passenger liner Queen Mary, in drydock at the Long Beach Naval Shipyard, rocked back and forth on its keel blocks for 5 minutes following the shock. Few windows cracked. People ran outdoors at the US. Naval Base. Los Angeles. (Press) Los Angeles area residents reported instances of chairs sliding across floors, cracks in plaster, and water sloshing out of swim- ming pools. The following report was received from a structural engineer: I inspected two downtown buildings on April 16, 1968, for an insurance company. One of the buildings is 13 stories, steel frame, brick filler walls. Both are of about the same construction and were constructed prior to any earthquake code design criteria. Interior partitions are mostly plastered hollow tile. Damage in both buildings was limited almost entirely to plaster cracks where old cracks from the 1933 and 1952 earthquakes were reopened or were slightly en- larged. There was a slight “banging” crumbling where one building is built tight against the other to the north. Palm Springs (Riverside County). “Light fixture swayed violently 2 or more feet in north-south direc— tion. Dinner guests became partially ill. Neighbors ran outside, some screaming.” Salton Sea. (Press) Residents on the north shore of the Salton Sea reported a 6-inch tide was mea- sured on concrete retaining wall. San Diego (San Diego County). Press reported hundreds of broken windows and severed power lines in beach communities of San Diego County. In the south San Diego Bay area, high-tension wires swung and arced over. People seated found it difl‘icult to stand erect. West Covina (Los Angeles County). Strong rolling northwest-southeast motion, lasted 1 minute. Diffi- cult to walk. Whittier (Los Angeles County). “Water splashed out of pool and ran into northeast drains, uphill, and back onto deck from northwest drains. Moved filter off of underground filter in northeast direction. Ground visibly wavy.” 144 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 0 1‘) LAKE ISABELLA 100 .A. O BAKERSFIELD 40 (» CACHUMA DAM O. LOS ANGELESI INSET SEE FIGURE 97 EXPLANATION SAN ONOFRE ‘ O TEIGGERED INSTRUMENTS A NON TRIGGERED INSTRUMENTS 50 RADIAL DISTANCE FROM EPICENTER -MILES 0 2 INSTRUMENTS OPERATED, I6 DID NOT 0 20 40 MILES SAN DIEGO I \_ _ 117'00' 1100' LAS VEGAS8 .. \J . HOOVED DAM YE) EPICENTER —""E\ 4- 9- so .——. 3723636 FIGURE 96.———Location of strong-motion seismographs that were triggered by the Borrego Mountain earthquake. Cantil (Eastern Kern County). Felt by all. Ve- hicles rocked. Fillmore (Ventura County). “Water in swimming pools splashed by rocking motion.” Fort Irwin (about 25 miles northeast of Barstow; San Bernardino County). Dizzy sensation experi- enced. Blythe (Lost Lake Resort, about 31.5 mi‘es north of Blythe, Colorado River area of eastern Riverside County). “I was sitting on couch in trailer. The trailer shook violently for a few seconds. The effects were very noticeable.” Coso Junction (southern Inyo County). “I was leaning on desk and thought I was having a dizzy spell. Also felt nauseated.” Tecopa (southeast Inyo County). Telephone poles and lines shook. Arizona: Horn (about 150 miles east of the epicenter and just north of Dateland). “One 1,800-foot well pumped red clay. Production of the well dropped off 900 gallons per minute. Desk shifted 1 inch.” Yuma. “Reports of very few cracks in concrete walks and driveways. Noticeable movement of parked vehicles. Water splashed out of pools.” Quartzsite. Water disturbed. Topock. Small objects and furniture shifted. Wenden. All persons in home felt a sensation of seasickness. STRONG-MOTION SEISMOGRAPH RECORDS By RICHARD P. MALEY SEISMOLOGICAL FIELD SURVEY, NATIONAL OCEANIC AND ATMOSPHERIC ADMINISTRATION GEOGRAPHIC DISTRIBUTION Examination of the geographic distribution of the strong-motion seismographs triggered by the Bor- rego Mountain earthquake (figs. 96, 97) ShOWS that within the first 170 miles, all 91 accelerographs were triggered, including every Los Angeles area station, whereas from 170 to 240 miles, 21 of 28 operated. In addition, two of the 18 installations maintained by the Environmental Research Laboratories Special Projects Party in Las Vegas, Nev., 215 miles from the epicenter, were also actuated. Most of the 16 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 0 SAN ERNANDO l“In“ nan- an, O GLENDALE .IZ ANGEL u! IIIIY VERNON .3: PASADENA clnuAL Lol llfllLll t z 0,, .- Du . ’~ "5‘ 5/ a n q -34. W L Inn Nov: 1405 s FIGUEIOA 5. ml 1"er" 4.0I1 s HILL 5. 250 1,“le 5.616 s OLIVE 7 no. 3' OLIVE 0. 020 5 BRAND Io CAL Teen 9 SANTA ANNA oAI . A W " Q 5 —uuuu-—uu / :3, asvlo wlLsmIE .I$DD WILSNIIE .IIDI AVE, suns .945 TIVEITDN .UCLA ENC ILDCV . I24: omen ATNENAEUI IAILLIKAN LIIRANV JEY PROPULSION LA! 633 E BROADWAY who IAIENIO all! LouA vIsTA us: I. Noovtl 3.07 In six” 3‘70 VILSNIRE 4.67 SUNSET I015 NJIIIIMLAND 7000 NOLLV woon Izo u. nounuou IucI VENTUIA PAccmA DAM TERMINAL ISLAND 2:5 w. IIOAnwAv ZOII laNAL PASADENA sEIs. LAI nu wlLsNIIE u—u-_—uu-uuuuuuuuau-u—»N— 118'30’ / 0 118‘15' l OLONG BEACH 5 MILES FIGURE 97. -— Location of strong—motion seismographs in the Los Angeles area that were triggered by the Borrego Mountain earthquake. 146 that did not operate are in upper levels of taller buildings. It is not known how many of these were in standby condition, since they are generally used to record the effects of nearby nuclear detonations rather than kept in ready status for earthquakes. Table 20 gives the strong-motion stations at increas- TABLE 20.—Strong-motion seismograph records from the Borrego Mountain earthquake Instruments Distance Stations epfizgnmter C & GS AR—240 RFT—ZEO (miles) El Centro.... .............. 1 41 San Diego... 1 67 Perris Dam site... 1 78 San Onofre ........................... 1 83 Colton .................................... 1 93 San Bernardino ............... 1 98 Devil Canyon ........................ 1 102 Cedar Springs Dam site... 1 104 Santa Ana ............................ 109 Puddingstone Dam .............. 1 117 Los Angeles area .............. 10 65 4 125—145 Pearblossom ........................ 1 142 Mohave, Nev ........................ 1 170 Lake Hughes 1 .................... 1 171 Fairmont Dam .................... 1 172 Castaic Dam site.. .. 1 1‘73 Santa Felicia Dam 2 177 Gorman ............ 1 189 Port Hueneme 1 190 Wheeler Rldg 1 210 Las Vegas, Nev.... ....... 2 210 Hoover Dam, Nev ................ 3 212 Lake Isabella Dam ........... 14 220 Santa Barbara ..................... 1 222 Bakersfield .................. 1 225 Taft ........................................ 1 1 237 Cachuma Dam ______________________ 2 240 Totals ...................... 26 78 10 Grand total ............... 1Not simultaneously connected. Four operated, one did not. ing distance from the epicenter, with the type of instrumentation indicated for each site. The closest nonactuated instruments (table 21) TABLE 21.—Strong-motion seismographs not triggered by the Borrego Mountain earthquake but within potential operable range Nominal distance . N b i - Station Instill-nursintts to (expllizfgifer Lake Hughes 4, 9, 12 ................... 3 170 Fort Tejon ................................... 1 190 Tehachapi ............................. 1 195 Lake Isabella ....................... 1 220 Buena Vista ...................... i 225 Total ..................................... 7 were part of the Lake Hughes array, 171 miles from the epicenter. The number 1 station there, a few hundred feet northeast of the San Andreas fault, was triggered, but the three others, 3-9 miles south- west of the fault, were not. Since the maximum accelerations recorded at station 1 were near the instrument’s operating threshold, the fact that the three nearby stations were not actuated may reflect minor variations in trigger sensitivities and subtle THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 differences in geology rather than their greater dis- tance from the San Andreas fault. Beyond the 170- mile epicentral radius, all accelerographs more than 45 miles inland were not actuated, whereas all those within 30 miles of the coastline and out to the 240- mile operable limit were. Preliminary inspection suggested that this pattern was related to the re- gional variation of geologic structure introduced by the San Andreas fault system and the existence of different tectonic provinces. The seven accelero- graphs near the coastline are located well west of the San Andreas fault in the Transverse Ranges, while 11 of the 18 inland accelerographs that oper- ated (excluding those in Nevada) are near to and east of the San Andreas fault in the central Trans- verse Ranges and the southern regions of the San Joaquin Valley and Sierra Nevada. Closer investi- gation suggests the possibility that this pattern of operation was due to local geologic conditions, rather than any regional structure. Six of the eight non- operating instruments are on a basement complex or other highly competent rock, while those beyond 170 miles and west of the San Andreas fault are on considerably poorer foundation materials, including earthfill dams and alluvium. None of the near-coast- line instruments are at basement-rock sites, and only one is on well-consolidated sediment. A review of the regional distribution of strong- motion seismograph records show that only six of the 114 records were obtained within 100 miles of the Borrego Mountain earthquake epicenter. The primary reasons for the lack of close-in stations are twofold: (1) A majority of the accelerographs are owned by various organizations and agencies other than the Seismological Field Survey and therefore are operated at the engineering structures of par— ticular interest to those groups. (2) The largest cluster of instrumentation is located in Los Angeles because of the building code provision requiring accelerographs in most taller buildings. ENGINEERING-TYPE LOCATIONS A breakdown of instrument locations according to “structure type” is shown in table 22. Twenty of the larger buildings had simultaneously connected ac- celerographs at the lowest, intermediate, and highest levels, with identical timing signals and a common starting capability (that is, when any one of three instruments is triggered, the remaining two are immediately actuated, thereby providing a uniform zero time on all records). Of those 20 buildings, 19 have been equipped under the Los Angeles building code that requires three acclerographs in most new structures taller than six stories. Three other build- ings have two accelerographs per structure, again THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 147 ACCELEROMETER MOTION up SENSITIVITY - I15 CM/g DISPLACEMENT MOTION EAST MAGNIFICATION II.0 ACOELEROMETER MOTION SOUT SENSITIVITY I I4.7 CM/g ‘ DISPLACEMENT MOTION SOUT MAGNIFICATION - I.O ACCELEROMETER MOTION wesr SENSITIVITY-l3.5 CM/y s on I-1 sea-1 I M A V ”V V FIGURE 98.—E1 Centro accelerograph record from the Borrego Mountain earthquake. similarly interconnected, and the remaining seven single instruments in the taller buildings are located at the lowest level except the two in Las Vegas that are on the roof. Thirteen accelerographs were in buildings of less than six stories, five of these in the Los Angeles area and the rest in smaller cities throughout central and southern California. Strong- motion seismographs in small buildings or sheds are those related to facilities other than dams in the California Department Of Water Resources Feather River project. The 15 records obtained at existing dams are from eight different structures where the instruments have been installed in various configura- tions of one to five accelerographs per location. The last three stations are at future Department of Water Resources dam sites. A casual inspection of table 22 reveals that most of the accelerographs were in locations where the records to a considerable extent reflected the charac- teristic response of a structure rather than pure ground motion. In fact only 20 of the 114 accelero- grams were obtained at sites with free-field condi- tions, and in Los Angeles just one of these was within 15 miles of the central complex of 20 mul- tiple-instrumented buildings. Among the 16 ground stations outside the Los Angeles area, 14 were near existing or planned hydraulic facilities. ACCELEROGRAPH RECORDS Acceleration records were written by three differ- ent instruments: the Coast and Geodetic Survey standard strong-motion seismograph and the AR—240 and EFT—250 strong-motion recorders; the latter two are manufactured by Earth Science Division of Teledyne Industries, now Teledyne-Geotech. Since the three instruments supply similar data (that is, a continuous three-component record of acceleration throughout the duration of a potentially damaging earthquake), it follows that their func- tional descriptions are closely analogous, although engineering characteristics vary considerably from instrument to instrument. The accelerographs are triggered by a horizontal 1-second pendulum during the early phase of an earthquake, thus actuating the “Operate” cycle at an acceleration level of approxi- mately 1 percent of gravity. Three unifilar torsion- type accelerometers detect motion along mutually perpendicular axes, two horizontal and one vertical, and through a system of mirrors deflect a beam of light focused upon translating photographic paper or film behind a collimating lens. Included on the record are 1-second time marks supplied by an internal clock system and a series of reference lines adjacent to each of the three data traces (figs. 98, 148 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 A Son Onofre ACCELEROMETER NOTIoN N33E SENSITIVITY - 1.9a cu/a ACCELEROMETER MOTION DOWN I SENSITIVITY . 158 CM/y 'MWWWWVWMWM‘ ACCELEROMETER NOTION N57w SENSITIVITY - 7.62 CM/g 'WW 5 SECONDS —-l .I“WWI—AWJI—'_'I—-_H-Hm-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-._-_-_- - B San Diego ACCELEROMETER MOTION UP SENSITIVITY - 20.0 CM/g —MWW~WW AccsLenousTER MOTION EAST SENSITIVITY} Is CM/g - - - -I—5 schNos°I -------------- ._MMWWWWWWWW ACCELEROMETER MOTION SOUTH Few—7 FIGURE 99.—Acce1erograph records. A, San Onofre accelerograph record from the Borrego Mountain earthquake. Note that the sensitivity of this instrument was about one—third of that at San Diego. B, San Diego accelerograph record from the Borrego Mountain earthquake. TABLE 22.—Locati0ns of strong-motion seismographs trig— gered by the Borrego Mountain earthquake Location Number of Free-field ground instruments conditions Upper level of buildings > 6 stories 26 Middle level of buildings > 6 stories 21 Bottom level of buildings > 6 stori 28 Lowest level of buildings < 6 stories 13 than a simple one-story structure. Small one-story building or shed ............................ 9 9 Existing Jam: 16 8 Future dam sites .......................................................... 3 3 Total 1116 20 1The number of accelerographs triggered is one larger than the number of records Obtained since one was lost owing to an instrumental malfunction. 99). Upon cessation of significant shaking, the in- strument automatically shuts off but remains in an operable mode in case there are subsequent earth- quakes. The nominal physical characteristics of the different accelerographs are given in table 23. (For more detailed descriptions see Halverson, 1969.) Displacement meters that are an integral part of many strong-motion seismographs will be discussed later. TABLE 23. — General characteristics of strong-motion seismo- graphs Model Period Sensitivity (13:23:13,351: Record Reggzgiing (sec) (cm/g ) critical) size (cm/sec) U.S.C. & G.S................0.045—0.084 7-20 60 6 or 12 in. 1 or 2 Teledyne AR—240 ........ .052 7.6 60 12 in. 2 .062 115.2 60 Teledyne EFT-250 ...... .048 1.9 60 ‘70 mm 1 .070 13.8 60 70 mm 1 1Only four of these models were in use, all in the Los Angeles area. A maximum acceleration of 0.14 g was recorded in a north-south direction at El Centro, the station nearest the epicenter, which was 41 miles to the southeast. As previously pointed out by Cloud and Scott (1968), this large pulse had an apparent sinelike character and was. superimposed upon a smaller background motion approximately 7 seconds after the recording began (fig. 98). The acceleration THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 levels were no higher than 0.06 9 before or after this major pulse in other directions at El Centro, nor for that matter at any other sites. Two other impor- tant records shown in figure 99 were obtained from San Onofre and San Diego, respectively 67 and 83 miles from the epicenter. The highest accelerations at San Diego, occurring between 9 and 13 seconds after the instrument was triggered, were about 0.03 g on both horizontal components. Vertical mo- tions, considerably smaller, had maximums near 0.01 g. The accelerogram from San Onofre, the first significant strong-motion record from a nuclear gen- erating site, shows two separate trains of relatively high amplitude motion: one at the beginning, and the other 16 seconds later. The peak horizontal acceleration was 0.04 9, larger than at San Diego and Perris, two stations nearer the earthquake. Per- haps the most notable feature of the San Onofre record was the vertical acceleration of slightly more than 0.04 9; this acceleration is higher than that recorded at any other site, including El Centro at one-half the epicentral distance. Table 24 lists the TABLE 24. —Maximum accelerations recorded at stations less than 120 miles from the epicenter Difstance C Maxilmum . rom omponent acce era- Station epicenter direction tion Geology (miles) (cm/sec”) El Centro ................ 41 South ........... 140 Alluvium, 2,000 ft over West... 47 late Tertiary sediments. Vertical ....... 33 Basement at 18.000—20,000 ft. San Diego ............... 67 South ........... 29 Alluvium, 50—100 ft over West... 25 Pliocene sediments. Vertical ....... 10 Granite at 4,000-5,000 ft. Penis ....................... 78 South ........... 12 Alluvium, thin veneer West ......... 18 over granitic rocks Vertica 6 (estimated less than 50 ft thick). San Onofre ............. 83 N. 33° E ..... 40 Slightly cemented Pliocene N. 57° W... 27 sandstone. Estimated Vertical ....... 45 depth to basement rock, 11,000 ft. Colton ...................... 93 23 Alluvium, more than 500 18 ft. Basement estimated Vertical... 19 at 1,000 ft. San Bernardino 98 18 Alluvium, approximately Ha of 4 1,000 ft. Basement Records). 3 estimated at 2,000 ft. Devil Canyon 102 140 Fractured limestone and (San 47 gneiss. 0n Santa Ana. Bernardino). 33 fault (strand of the San Andreas system). Cedar Springs ........ 104 . 6 Granitic rocks. N. 70" E ..... 6 Vertical ...... 7 Santa Ana. .............. 109 14 Alluvium, 350 ft over 14 Pliocene sediments. 5 Basement rock at approximately 10,000 ft. Puddingstone ......... 117 N. 25° E ..... 16 Volcanic agglomerate over N. 65° W.... 17 Miocene sediments. Vertical ...... 4 Basement estimated at 500 ft. maximum accelerations for the three components at all stations up to 120 miles from the epicenter, and briefly describes the geology of each site. (For a complete tabulation of strong-motion seismograph results, see US. Coast and Geodetic Survey and California Institute of Technology, 1968.) 476-246 0 - 72 - 11 149 Figure 100 shows the variation of maximum ac- celeration with epicentral distance for the horizontal components of all strong-motion seismographs that operated during the earthquake. Comparison of the soil conditions at sites less than 120 miles from the epicenter (table 24) indicates that sites on or near bedrock — Perris, Devil Canyon, and Cedar Springs —fall on the lower area of the curve envelope, whereas those sites on alluvium or other less-consoli- dated materials tend to be on the upper parts of the graph. At Perris there is only a thin veneer of allu- vium overlying the basement complex (John Wolfe, oral commun.), and the short seismogram (6 sec) obtained there is more typical of that recorded directly on basement rock (that is, low acceleration level, less than 0.02 9, but relatively high frequency content near 10 Hz) than that recorded on alluvium. Data were also plotted for the maximum vertical accelerations at sites less than 100 miles from the epicenter, thus showing the relatively higher values recorded at San Onofre. The.vertical maximums observed at El Centro, San Diego, and Perris were respectively 75, 22, and 13 percent of that at the San Onofre site. The square area outlined on figure 100 encloses measurements from 35 ground-level instruments in the Los Angeles area at a nominal epicentral dis- tance of 125—145 miles. Peak accelerations ranged from near zero to slightly greater than 0.01 9; there was no apparent relationship of attenuation to dis- tance. Attempts to correlate the same information with the local geology were unsuccessful; however, the amplitudes were generally quite small, and the potential scaling error involved is of itself rather large compared with the amplitudes themselves. .06 \ {\MAXIMUM ACCELERATlON AT EL CENTRE) . NORTH-SOUTH 0 My “FEAST-WEST COMPONENY .05 \ HORlZONTflL COMPONENY' VENTlCAL COMPONENY I \ \ .SAN ONDFRE \ \ .EL CENTRD \\ o a \ ' \ \ c u \ comm . ?.\ o m \ . LOS ANGELES AREA .\ . SAN DlEGO. DEVIL CANYON. ACCELERATION, FRACYION 0F GRAVITY .o . , . cams , SPRINGS. (v o . .SAN BERNARDINO‘J‘X'e ° 00 \“w' ‘—_—.‘ 7"“.L IZO IGO DISYANCE Y0 EPICENYER. MVLES reams. FIGURE 100.—Maximum horizontal accelerations plotted against epicentral distance for the Borrego Mountain earthquake. Vertical accelerations are also included for records obtained at less than 100 miles. 150 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 ————— m—I.—Ir-J-I.I-I.—II-I.-I.I— WWW—Immw—.—.— ........ ACCELEROMETER MOTION SOBW SENSITIVITY I [5.5 CM/g 5 cm ACCELEROMETER NOTION DOWN SENSITIVITY - I5.3 cM/g ACCELEROMETER MOTION s 82 E SENSITIVITY - I5.4 CIA/p ~ WW I s SECONDS —-————1 FIGURE 101. ——Accelerograph record from the Central Engineering building at the Jet Propulsion Laboratory in Pasadena, ninth floor. Possibly, as suggested by Cloud and Hudson (in US. Coast and Geodetic Survey and California Institute of Technology, 1968), “significant studies of the distribution of ground motion can be made” through the use of the entire strong-motion record with com— plete soil profiles and more sophisticated analytical techniques. In addition to the 35 ground-level records, 44 accelerograms were written in the middle and upper level of the taller buildings in Los Angeles and Pasadena. The maximum accelerations and the character of these recordings reflect the structural characteristics of individual buildings, as shown by the record in figure 101 which is from the ninth floor of the Central Engineering building at the Jet Pro- pulsion Laboratory in Pasadena. Maximum accelera- tions on the top floor were 0.03 9, six times those recorded in the basement; the predominant periods were 1.09—1.12 and 1.06—1.14 seconds in north-south and east-west directions, respectively, compared with values of 0.92 and 0.93 second measured during an independent observation of ambient building vibra— tions. The difference may be attributed to the “soft- ening” effect in the structure under higher horizontal force loads. A peak acceleration of slightly over 0.07 g at a period of 0.60 second was recorded on the top floor of one structure in Los Angeles, although in most instances the maximums were 0.04 g or less, frequently considerably less. DISPLACEMENT RECORDS Displacement measurements were obtained from two different instruments, the Coast and Geodetic Survey displacement meter and the Carder displace- ment meter, both optional features installed at 11 accelerograph stations where ground-level records were obtained during the earthquake. The Carder displacement meter, operated as an integral compo- nent of the accelerograph, is an inverted pendulum suspended by a cross hinge and stabilized by a tor- sional restraint wire. The seismogram is produced on photographic paper thrOugh a system of mirrors that transmit a reflected light ray from the pendulum to the recording surface (Carder, 1964). The Coast and Geodetic Survey displacement meter, although triggered and controlled by the accelerograph, is housed separately from it and records independently. It consists of two horizontal pendulums set at right angles to each other and held level by angular sup- port wires. The axes of the pendulums are free to pivot in a needle-in-jewel arrangement. The record is made by an optical system similar to the one in the Carder displacement meter. Both instruments are normally adjusted for a magnification of 1. (See table 25 for typical physical characteristics.) THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 TABLE 25.——General characteristics of displacement meters Damp- Model Period Magni- ins Rffzzrd 535;; (sec) fication $25-$23) fin.) (cm/sec) Carder meter ......... ..12.0— 6.0 1 60 6 or 12 1 or 2 U.S.C. & G.S. me 10.0 1 60 6 .5 1Nominally adjusted to 2.5 or 5.0 see. Figure 102 shows the locations of displacement meters at ground-level stations in California and Nevada between 41 and 240 miles of the earthquake epicenter. The largest displacement occurred at ‘El Centro, where a recorded double amplitude of 13.9 cm (maximum single amplitude of 7.2 cm) was correlated in time of occurrence and duration with the sinelike acceleration pulse of 0.14 9 previously referred to in figure 98. (A double amplitude is the total peak-to-peak measurement of successive trace deflections in opposite directions, whereas a single amplitude is a measurement from the null, or zero, line to the deflection in one direction.) As noted by Cloud and Scott (1968), the displacements at El Centro were large compared to the small accelera- tions, and in fact continuous single amplitudes greater than 4.0 cm were recorded for more than 90 seconds. \ \ \\ \ \ 04 (12° (40‘ L40 H OOVEV 06>\\4 Dar“. 49 \ 4 \ \\ \ .Bakersheld \ x \ \ .Cochuma Dom 'Fai'mon' Port . Hueneme : UC LA . .Colmn ,Los Angeles .Long Beach 'Sunvo Ana Q Earthquake % a Epicenler v9 (9 A 1‘ Vb El Cenlro. N on cayFORNLE ————-"“ I <2, -- " nglco 1» 0 20 40 Miles FIGURE 102.— Location of the ground-level displacement meters actuated by the Borrego Mountain earthquake. 151 TABLE 26.—Maximum amplitudes recorded by ground-level displacement meters Distance to Maximum Station enicsmter mm“ M (mlles) Double Single Carder displacement meters El Centro ........................................ 41 7.25 6.30 Santa. Ana ...................................... 109 1.40 1.15 Long Beach .................................... 127 .60 .70 UCLA (Los Angeles) .................. 148 .45 .40 Fairmont ......................................... 172 .17 .11 Port Hueneme ................................ 190 .20 .40 Hoover Dam ................................... 210 .25 .86 Cachuma Dam ................................ 240 .20 .81 U.S. Cout and Geodetic Survey displacement meters El Centro1 ....................................... 41 South ............... 9.0 4.5 East... 9.0 4.6 Colton ............................................... 93 3.20 1.65 2.66 1.70 Subway terminal (Los Angeles) 138 3.70 2.10 3.30 1.96 Bakersfield ...................................... 225 3.40 1.96 1.30 .66 1Pendulums hit stops. Peak north-south displacements (table 26) were plotted against epicentral distance in figure 103. Amplitudes measured from the six Carder displace- ment meter records at distances less than 200 miles fell on a relatively straight line defined by the equation log A:2.32~0.011d, where A=maximum single displacement, in millime- ters, and d=epicentral distance, in miles. Measurements scaled from three of four Coast and Geodetic Survey displacement meters are related to the curve in the following manner: (1) not appli- cable at El Centro because the pendulums struck the stops at both ends of the swing-arc; (2) reasonably close at Colton, 93 miles from epicenter; and (3) well beyond the arbitrary ZOO-mile equation limit at Bakersfield. The fourth record, obtained in central Los Angeles, showed abnormally large displacements, the only major deviation from the trend of the curve. Note the comparatively large amplitude also recorded at Bakersfield, 225 miles from the epicenter. Displacement records were also made by instru- ments installed on structures at Hoover and Ca- chuma dams and in the top of two Las Vegas buildings, but these are not considered here. SEISMOSCOPE RECORDS After the earthquake, 131 seismoscope plates were removed from instruments in central and southern California. Since amplitudes were generally quite small, there is no certainty that records thus ob- tained were from the Borrego Mountain earthquake rather than from minor local earthquakes, except IUU - _ CARDER DISPLACEMENT METER 0 3‘ “NW0 USC 655 DISPLACEMENY METER m . .. .. LOG A = 2.32-.0IIa g l E has ANGELES El | V BAKERSFIELD I B ._ 5 couou l 1 LIMITING DISYANCE 3 SAN?! ANA FOR EQUATION | ‘3 3. a | 1 I0- I '5 .- ° l in P i I p- _ ‘5 I z I(,om‘. BEACH ‘ l D E ' | I: UCLA | a _ I reeves DAM I 0 son NUENEMEU) I 2 G) CAcuquA emu) ( rumour | l l I l l l l 0 40 so 120 I60 zoo 240 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 DISTANCE FROM EPICENTER (MILES) FIGURE 103.——Variation of maximum displacement (north- south) recorded by ground-level displacement meters with distance from epicenter of Borrego Mountain earthquake. for El Centro and some dam sites, where instru- ments are inspected frequently. The seismoscope consists of a free conical pendu- lum supported by a single pivot suspension wire providing freedom of movement in all horizontal directions (Cloud and Hudson, 1961). The relative motion between the instrumental base and pendulum is etched on a smoked watchglass by a scriber at- tached to the frame. As suggested by Hudson and Cloud (1967), the El Centro results are presented in terms of the relative displacement response spec- trum that is calculated by the following solution from direct measurements off the watchglass g T’- n 511 =4—7TTZ (Pmax E where, ’ Sd =relative displacement response spec- trum, in inches, T =undamped natural period of the seismo- scope, _At (Pmax-‘E, with, A, =single amplitude on the record plate, 5 =ti1t sensitivity of the instrument, approximately 2.15 inches per radian, and n =fraction of critical damping for the am- plitude,A,. Table 27 summarizes the results from four El Centro seismoscopes located at sites 40—42 miles TABLE 27.—Relative maximum displacement spectra from the El Centro seismoscope records Maxi- Al‘race mum mpli- relative . . tude displace.L Direction Distance Damp- Sensi- - mg tivity 322:? (percent (in./ of radi- Location epi- center (miles) critical) an) (m.) 7151“? Central Union High School ........ 40.3 0.75 10.4 2.16 0.34 0.88 N. 13" E. Electric substation 41. .73 11.6 2.14 .34 .87 S. 5" W. Steam plant ............ 41.5 .75 10.0 2.19 .40 1.00 S 4" W. Waterworks ............ 42.0 .74 10.5 2.19 .33 .82 S 21° W. from the epicenter. The calculated displacement re- sponse spectra ranged from 0.82 to 1.00 inch. Superimposed on a map of El Centro at their re- spective locations (fig. 104), these records indicate that the polarized amplitudes are oriented north- south, approximately 45 degrees to the trend of ground rupture on the Coyote Creek fault. An angle this large between recorded amplitudes and the trace of nearby fault movement had been observed at Hollister in 1959 and 1960 (Cloud, 1967) and at Parkfield in 1966 (Hudson and Cloud, 1967). Un- fortunately, there were no other seismoscopes less than 78 miles from the epicenter, so the angle can- not be measured elsewhere. At many dams in southern California there are two seismoscopes, one on the crest and the other nearby on rock typical of that underlying the dam. Table 28 gives the unreduced maximum amplitudes scaled directly from the records obtained at five earthfill dams at epicentral distances of 117—166 miles. Although there is no absolute certainty that these records were from the Borrego Mountain earthquake, record size and frequency of inspec- tions suggest that at least those recorded at the Puddingstone, San Gabriel, and San Fernando Dams are from that earthquake. Table 28 also gives records from an earthquake in Borrego Valley on April 28, 1969, and from a shock of unknown origin. The largest amplitudes from the crest records were essentially normal to the axis of the dam, and when these values were compared with the peak measurements obtained from nearby bedrock re- gardless of direction, maximum crest to abutment THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 32° 50' 00" covor: N CREEK FAULT EL CENTRO O HIGHWAY 56 HIGHWA" az°dfso" HIGHWAY 86 u5°3s'oo” us" 32 so 32°4roo” FIGURE 104.——E1 Centro seismoscope records superimposed on a map of the city. amplitude ratios ranged from 2:1 to 16:1. While four of the sites showed ratios between 2:1 and 6:1, it is interesting that San Gabriel Dam with a crest height of 320 feet, more than twice that of the next largest structure on the list, had ratios of approxi- mately 16 :1 for two different earthquakes. The rock underlying San Gabriel Dam is a crystalline base- ment complex, whereas at the other sites the rocks are volcanic agglomerates, Tertiary shale, and al- luvium (table 28) ; consequently, on strictly geologic grounds, a larger range in relative response be- tween the crest and abutment seismoscope may be 153 anticipated at San Gabriel Dam. One should exercise considerable caution in in- terpreting these results for two reasons. (1) The values in table 28 are raw measurements scaled directly off the seismoscope plates. (2) The ampli- tudes represent only one point on the relative dis- placement response spectrum curve corresponding to the physical characteristics of the instrument. In calculating the relative displacement, it would be necessary to consider the damping factor for very small records (as at the abutments), and since calibration of the seismoscope does not permit suf- ficiently accurate determination of this variable at such small amplitudes, the inherent error in the displacement response could be quite large. The data presented here merely indicate that there were sig- nificant increases of ground shaking on fill-type dams compared to that of the bedrock foundation within the limits of the seismoscope’s response range. Figure 105 shows the comparative records from San Fernando Dam in the San Fernando Valley of Los Angeles, San Gabriel Dam in the foothills above Azusa, and the single record from Pudding- stone Dam at San Dimas. The contrast in ampli- tudes is strongly evident in the first two sets of records, and although there was no abutment seis- moscope to compare with the Puddingstone crest, the large amplitude is of itself quite meaningful, considering the relatively large epicentral distance of 117 miles. RESPONSE SPECTRA By A. GERALD BRADY DIVISION OF ENGINEERING AND APPLIED SCIENCE, CALIFORNIA INSTITUTE or TECHNOLOGY The three acceleration records containing the strongest ground motion from the Borrego Moun- tain earthquake were the El Centro, San Diego, and San Onofre records. Parts of these appear in figures 98 and 99; the complete records were published in another report (US. Coast and Geodetic Survey and California Institute of Technology, 1968) . These three records have been incorporated in a current digitization project under way at the California Institute of Technology. One volume has been pub- lished thus far (Earthquake Eng. Research Lab., 1969). Part A of that volume includes two records, IA19 El Centro and IA20 San Diego Light and Power Co., and part B includes the 1340 San Ono- fre, Southern California Edison record (labeled 68—8T here, the original Caltech reference number). The project report includes digitized accelerograms (v. 1), corrected accelerograms (v. 2), and response spectra (v. 3). 154 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 TABLE 28. —Maximum seismoscope amplitudes from selected dams in southern California Distance Orientation Maximum Direction Amplitude . from ratio, Height Geologic Location of the am litude from Damt effiflgst? dam crest (tin. ) epicenter 5?th ype 1733“ (ft) foundation 13on Mountain earthquake of April 9, 1968 Puddingstone crest ...... 117 N. 62° E. 0.36 N. 30" W. ........ Earthfill, 1.080 147 Volcanic curved agglomerate. crest. San Gabriel: 124 N. 60° E. 16:1 Earthfill 1,520 320 Weathered Crest .......................... .230 N. 30° W. and rock, gneissoid Abutment .................. .014 straight granite. cres . Eaton Wash: 134 N. 60° E. 2:1 Earthfill, 1,545 63 Alluvium Crest ......................... to E.—W. .06 S 20° E curved Abutment .................. .03 crest. San Fernando: 157 N. 73° W. 6:1 Earthfill, 2,080 142 Tertiary Crest ........................... .24. - N. 15° W. straight shale. Abutment ................. .04 N. 70" E. crest. Dry Canyon: 166 N. 80° E. 2%:1 ...... do ............ 510 67 Do. Crest ........................... .15 N. 17" E Abutment ................... .06 N. 67° E. Barren Valley ear-tr “ of April 28, 1969, magnitude 5.8 (P ‘ \ Puddingstone: 117 N. 62° E. N. 56° W. 2+:1 See data above. Crest .......................... 0.09 Abutmentl ................. San Gabriel: 124 N. 60' E. N. 54° W. 16:1 Crest ........................... . Abutment ................... .004 Eaf‘ ‘ of unknown origin between June 1968 and March 1969 Dry Canyon: 166 N. 80° E. N. 56‘ W. ........ See data above. Crest ........................... 0.180 Abutment ................... .035 ‘A seismoscope was installed on the abutment at Puddingstone Dam after the Borrego Mountain earthquake of April 9, 1968. The response spectra described herein have been obtained from the uncorrected accelerograms of volume 1. They will be similar to those forthcoming in volume 3, since the particular baseline correction techniques are not critical in response-spectra calc- ulations (Amin and Ang, 1966). The response spectra in figures 106 and 107 have been calculated from the total length of the avail- able records, that is, approximately 87, 79, and 45 seconds. The period ranges from 0 to 10 seconds, and the five damping values are 0, 2, 5, 10 and 20 percent of critical. All three components of acceler- ation are analyzed. The computer programs used are those described by Nigam and Jennings (1968) ; changes were made on subroutines PCN05, PCN06 and PCN07 so that they could be run on the IBM 360 at present at Caltech. In addition, a few of the digitized time coordinates had to be further sep- arated so that the rounding to 0.005 second, per- formed in subroutine PCN05, resulted in nonequal increasing times in every record. These changes were of the same magnitude as the rounding off changes which are shown to affect response spectra calculations by less than 2 percent. Figure 106 shows the relative velocity response spectra for the El Centro record in the south, west, and vertical directions. The station at El Centro is approximately 41 miles southeast of the epicenter in the general direction of the San Jacinto and Imperial faults. (The focal depth, for comparison, was 11 km.) The two horizontal Components here are at the same angle, 45 degrees, to the radial motion from the epicenter, and therefore both components are affected by ground motion both longitudinal and transverse to the fault. Peaks in the spectra are well distributed over the longer periods, those greater than 1 second. Of particular note are peaks at 2 seconds in the south component and at 51/2 seconds in the west component. Vertical motion shows considerably lower spectral values with no predominant peak. Figures 107A, B, C are the relative velocity re- sponse spectra for the south, east, and vertical components of the record of the San Diego Light and Power Co., 67 miles west-southwest of the epi- center. This direction is nearly perpendicular to the San J acinto fault. Again the horizontal component directions of the record do not coincide with longitudinal or trans- verse motion, and therefore both motion compon- ents are present in both record components. Peaks are apparent at periods of 2 and 31/2 seconds in both these components. Figures 1070, E, F are the spectra for the record from the San Onofre Southern California Edison Power Plant, 83 miles west-northwest of the epi- center. The component directions are N. 33° E., N. THE BORREGO MOUNTAIN EARTHQUAKE San Fernando Crest San Fernando Abutment San Gabriel Crest San Gabriel Abutment SMM Puddingstone Crest FIGURE 105.— Seismoscope records from San Fernando, San Gabriel, and Puddingstone Dams. The vertical marks indi- cate the relative alinement of the crest of individual dams. 57° W., and vertically downward. This N. 33° E. direction is almost perpendicular to the general di- rection of the fault zone, but there appears to be no significant difference in the spectra of the two hori- zontal directions. Both have contributions in the short-period range of the same order of magnitude as in the remainder, as is also the case for the vertical component. When compared together, the spectra from the three locations‘show the usual rapid decay of motion as the distance from the epicenter is increased. The undamped curves show this well, but they are actu— ally a little misleading because the records from El Centro and San Diego are twice as long as the San Onofre record. This extra length has an appreciable OF APRIL 9, 1968 155 60F - 5° _ A EL CENTRO, SOUTH n Dumping. 0, 2, 5, IO, 20 percen! 40— — 30‘— _ 5 I I o ,— ,_ B EL CENTRo, WEST Damping O,2,5,l0,20 percent .p o l RELATIVE VELOC/TY {Ml/SEC) 8 l J n) o I 1 C EL CENTRO,VERTICAL Dumpmg‘ O, 2, 5, IO, 20 percent l 2 3 5 6 PERIOD (SEC) FIGURE 106.—Response spectra from El Centro. A, South; B, West; C, Vertical. influence (Hudson and others, 1969) on the un- damped spectra, particularly for the long-period components, which tend to be amplified by energy buildup during long excitations. The short-period components decay with distance more rapidly than longer period components; the El Centro spectra, obtained 41 miles from the epicenter, indicate this clearly. The predominant 2-second component ap- pearing at El Centro and San Diego has decayed to an insignificant amount at San Onofre. With no accelerometer closer than 40 miles from the epicenter, the vertical components recorded are not as important as the two horizontal components. In particular, the short-period vertical motion that is usually displayed only close to the epicenter was not recorded. SUMMARY AND CONCLUSIONS The Borrego Mountain earthquake was felt over 60,000 square miles of California, Nevada, and Ari- zona, but damages were generally minor owing to the sparsity of population in the immediate epi- central region. A maximum intensity of VII, based upon the routine questionnaire canvass, was as- THE BORREGO MOUNTAIN A SAN DIEGO, SOUTH * Dumping O. 2,5,|0,20 percem B SAN DIEGO, EAST ‘ Dumping O,2,5,l0.20 percent C SAN DlEGO, VERTICAL IO k Damping O,2,5,IO, 20 percent 0 SAN ONOFRE, N35° E Dumping 0.2,5.|O.20 pevcenv RELAT/l/E ' VELOCITY {IN/SEC) E SAN ONOFRE, N 57°w Dampmq O,2,5,|0,20 percent F SAN ONOFRE,VERTICAL Dumping 0,2,5,|0,20 percenl 4 5 6 PERIOD (SEC) FIGURE 107.—Response spectra from San Diego and San Onofre. A, San Diego, south; B, San Diego, east; C, San Diego, vertical; D, San Onofre, N. 33° E.; E, San Onofre, N. 57" W.; F, San Onofre, vertical. signed to an area near the Coyote Creek fault in the Borrego Springs—Ocotillo Wells region. The earthquake actuated 115 strong-motion seis- mographs operated by the Seismological Field Sur- vey; these instruments are located between 41 and 240 miles from the epicenter. The maximum ac- celeration and displacement occurred at the nearest station, El Centro, where values of 0.14 g and 7.25 cm were measured. Seismoscope plates were collected from 131 instruments, but owing to the small amplitudes, only five records from the Los Angeles area dams and four from El Centro can definitely be considered to have been written by this particular earthquake. The largest relative displacement re- sponse spectra that were measured at El Centro ranged from 0.82 to 1.00 inch. After examining the results of this earthquake, several conclusions may be drawn concerning the pattern of strong-motion seismograph operations and the evaluation of their records. (1) Since only 6 of 114 records were obtained less than 100 miles from the earthquake epicenter, it is evident that a greater geographical distribution of EARTHQUAKE OF APRIL 9, 1968 strong-motion seismographs would be desirable in southern California, particularly near the very active San Jacinto fault zone, where there have been eight shocks of magnitude 6 or greater between 1915 and 1968—an average of one every 7 years (Allen and others, 1965). As previously stated, . most of the existing accelerographs are located at structures of particular interest to the organizations that own the equipment; therefore, a cooperative program was instituted between the Seismological Field Survey and the California Institute of Tech- nology to select and instrument various new loca- tions to fill in the existing geographic gaps in the network. By 1970, eight accelerographs had been located along the San Jacinto and Imperial fault zones, and additional sites were selected for a num- ber of future installations. (2) At the time of the earthquake, there were only four free-field stations in the Los Angeles area, and only one was near the 20 buildings with multiple accelerograph installations. Because more ground- level stations would be advantageous, four sites were selected for future instrumentation to provide con- trol stations that would be on basement rock as well as being near the centers of present high-rise con- struction. Three of these sites were equipped with instruments by mid-1970. (3) The pattern of strong-motion seismograph triggering between 170 and 240 miles from the epicenter was correlated with the geologic founda- tion; that is, most instruments on basement rock failed to operate, but nearly all of them on rela- tively poor soils were actuated by the earthquake. (4) Geologic conditions seemed to exert a general influence on the level of acceleration at stations less than 120 miles from the epicenter; that is, those instruments located on or near basement rock re- corded lower values than those on alluvium or rela- tively less consolidated materials. The level of acceleration in the Los Angeles area, about 0.01 g, was too small to show any effective relationship be- tween soil conditions and seismic response, at least not without more comprehensive evaluation. (5) The displacement attenuation decrement up to a ZOO-mile epicentral distance was shown to be logarithmic and may be defined by the equation log A = 2.32 — 0.011d. Beyond the ZOO—mile limit there was no discernible pattern to the recorded data. (6) Strong polarized amplitudes at a large angle to the trend of faulting were observed on seismo- scope records from El Centro. Perhaps the installa— tion of paired seismocopes on opposite sides of active fault zones, a program currently in progress, may help to shed more light on this phenomenon. THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 (7) Seismoscope records from several earthfill dams show maximum amplitude ratios, crest to abutment, ranging from 2:1 to 16:1. Although these records were quite small and the ratios were ob- tained from unreduced scalings, it is sufficiently evi- dent that future investigation of this apparent amplification in terms of the dams’ structural be— havior under applied earthquake stresses would be of considerable value, especially if more complete instrumentation were made available. (8) For the first time in the history of strong- motion seismology, a large number of instruments in California were triggered by a single event, a pos- sibility that could become routine for even mod- erate-sized earthquakes, provided they occurred near an appropriate concentration of accelerographs. At the time of the Borrego Mountain earthquake, there were 108 accelerographs in southern Cali- fornia, defined as the area south of a line connecting Cachuma Dam and the Lake Hughes array (fig. 100). By the end of 1969 there were 200 instruments in this area, 150 of these in Los Angeles alone. It is quite clear that future shocks on or near the San Andreas fault system that are equivalent in magni- tude to or stronger than the Borrego Mountain earthquake will trigger a considerable part of the network, in any event probably more than 100 in- struments. ‘ (9) The three acceleration records showing strongest ground motion are included in a Caltech digitization project, and the relative velocity re- sponse spectra have been calculated. The distances from the epicenter are such that short-period com- ponents have been considerably attenuated at two of the sites, but not at San Onofre; here peaks in the spectra are well distributed over the whole period range. REFERENCES CITED Allen, C. R., St. Amand, Pierre, Richter, C. F., and Nord- quist, J. M., 1965, Relationship between seismicity and 157 geologic structure in the southern California region: Seismol. Soc. America Bull., v. 55, p. 753—797. Amin, Mohammad, and Ang, A. H.-S., 1966, A nonstationary stochastic model for strong-motion earthquakes: Illinois Univ. Civil Eng. Studies, Structural Research Ser. No. 306, 115 p. Carder, D. S., ed., 1964, Earthquake investigations in the western United States, 1931—64: U.S. Coast and Geodetic Survey Pub. 41—2, 264 p. Cloud, W. K., 1967, Seismoscope results from three California earthquakes in the Hollister, California area: Seismol. Soc. America Bull., v. 57, p. 1445—1448. Cloud, W. K., and Hudson, D. E., 1961, A simplified instru- ment for recording strong-motion earthquakes: Seismol. Soc. America Bull., V. 51, p. 159—174. Cloud, W. K., and Scott, N. H., 1968, The Borrego Mountain, California, earthquake of 9 April 1968: A preliminary engineering seismology reportz’ Seismol. Soc. America Bull., v. 58, p. 1187—1191. Coffman, J. L., and Cloud, W. K., 1970, United States earth- quakes 1968: U.S. Coast and Geodetic Survey, 111 p. Earthquake Engineering Research Laboratory, California Institute of Technology, 1969, Strong-motion earthquake accelerograms, digitized and plotted data: v. 1, pt. A, 164 p. Halverson, H. T., 1969, Some recent developments in strong- motion seismographs: Teledyne-Geotech Pub., Monrovia, Calif., 32 p. Hudson, D. E., and Cloud, W. K., 1967, An analysis of seismo- scope data from the Parkfield earthquake of June 27, 1966: Seismol. Soc. America Bull., v. 57, p. 1143—1159. Hudson, D. E., Nigam, N. C., and Trifunac, M. D., 1969, Analysis of strong-motion accelerograph records: 4th World Conf. of Earthquake Eng., Santiago, Chile, 1969, Proc., Sec. A—2, 17 p. Nigam, N. C., and Jennings, P. C., 1968, Digital calculation of response spectra from strong-motion earthquake records: Earthquake Eng. Research Lab., California Inst. Technology, 65 p. U.S. Coast and Geodetic Survey, Seismological Field Survey, and California Institute of Technology, Earthquake En- gineering Research Laboratory, 1968, Strong-motion in- strumental data on the Borrego Mountain earthquake of 9 April 1968: San Francisco, Calif., 119 p. Youd, L. T., and Castle, R. 0., 1970, Borrego Mountain earth- quake of April 8, 1968: Am. Soc. Civil Engineers Proc., Jour. Soil Mechanics and Found. Div., v. 96, no. SM4, p. 1201—1219. ENGINEERING GEOLOGY By R. O. CASTLE and T. L. Your) US. GEOLOGICAL SURVEY ABSTRACT Effects of the magnitude 6.4 Borrego Mountain earthquake that relate to enginering geology were surveyed between Coronado and Glamis, Calif. Of particular significance were the complexity and 0.15-1.0-km maximum half width of the main tectonic fracture zone along the Coyote Creek fault and the secondary (or spatially separated) faulting at least 2.4 km from the main trace. This pattern and distribution of tectonic ruptures, particularly where associated with pre- dominantly transcurrent movement, generally have been identified with shocks of magnitude 8+. Of possibly equal engineering significance was the surficial rupturing along the Banning—Mission Creek, Imperial, and Superstition Hills faults, which suggests that fault displacements may be seis- mically triggered along any discontinuity associated with accumulated elastic strain. Modified Mercalli intensities of VIII—IX were indicated for the epicentral region, in good agreement with those predicted for a magnitude 6.4 shock. Terrain effects attributable to seismic shaking included rockfalls and soil falls, slumps and incipient slumps or ground cracking, dislodged pebbles and cobbles, shattered crusts overlying poorly indurated dry deposits, seismic com- paction ruptures, and sand boils. Most of these features were located in the epicentral area, east of the Coyote Creek fault. Structural damage, which was confined chiefly to the epi- central and Salton City areas, was remarkably light for an earthquake of this magnitude, even when consideration is given to the very limited development of this region. Cracked and spalled concrete bridge piers constituted the most seri- ous damage to a modern structure in the epicentral region. Broken and cracked concrete-block walls were observed local- ly, and unreinforced masonry walls were toppled as far away as 65 km from the epicenter. Damage to wood-frame structures was generally minor, expressed chiefly as cracked plaster and stucco walls. Projected and displaced objects were reported from Coronado to about .8 km east of Brawley. The most severe damage beyond the immediate area of the epicenter occurred over the late Cenozoic lake deposits under- lying most of the Imperial Valley. IN TRODUCTION The magnitude 6.4 Borrego Mountain earthquake of April 9, 1968, was felt over approximately 155,000 sq km of the southwestern United States (Cloud and Scott, 1968a, p. 1187; Seismological Field Survey, National Oceanic and Atmospheric Administration, this volume) and an undetermined part of north- 158 western Mexico. Perceptible shaking extended from Dateland, Wenden, and Horn, Ariz., westward to the Pacific Ocean and from Yosemite Valley on the north (Cloud and Scott, 1968a, p. 1187; 1968b, p. 19) to at least as far south as Ensenada, Baja California, Mexico (M. G. Bonilla, oral commun., 1968). Per- manent effects, although generally minor outside the epicentral region (pl. 4), were reported from as far away as Los Angeles, about 220 km from the epicenter. Although the main shock was clearly associated with the release of considerable seismic energy, its location within the sparsely settled and generally undeveloped western Colorado Desert precluded ex- tensive damage. In spite of the generally light damage, however, this earthquake is considered sig- nificant to engineering geology in at least four respects: 1. The main shock was associated with the forma— tion of two partly overlapping or en echelon zones of tectonic surface ruptures along a 32- km reach of the Coyote Creek fault. (See pl. 5; Clark, “Surface Rupture along the Coyote Creek Fault,” this volume). This breakage was locally complex, particularly within the area of overlap where the zone of fracturing was as much as 3 km wide. Although shocks of magni- tude 6 and greater have commonly produced surface ruptures (Brown and others, 1967, p. 22), the complexity and width of the zone of tectonic rupturing generated during the Bor- rego Mountain event have not generally been reported for earthquakes of this or lesser mag- nitude, particularly along strike-slip faults. The complexity was such, moreover, that al- though the approximate locations of the rup- ture zones were generally predictable, the precise locations of the individual fractures were only locally predictable. 2. Temporally associated fault displacements oc— curred on the Banning—Mission Creek, Impe- THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 rial, and Superstition Hills faults at distances of as much as 50 km or more from the epi- center. (See Allen and others, this volume). This phenomenon suggests that we should not, as a matter of course, restrict our predictions of tectonic surface rupturing to the fault (or fault zone) associated with the main shock, even when dealing with relatively small earth- quakes. 3. Secondary terrain effects (those derived from shaking) were generated in profusion in the epicentral area. These effects, including both free-face movements and various liquefaction phenomena, identify the epicentral area with Modified Mercalli intensities of VIII or more. Similar terrain effects suggest that intensities were as high as VII over several thousand square kilometers of the western Imperial Valley. 4. Even when consideration is given to the low population density of the epicentral region, structural damage was remarkably light for an earthquake of magnitude 6.4 and Modified Mer- calli intensities of V or greater over at least 65,000 sq km. (See Cloud and Scott, 1968b, p. 26.) Local terrain effects and transient effects consistent with intensity VIII or greater suggest that structural damage should have been much more severe, far more pervasive, and experienced over a much wider area than it actually was. NATURE AND PURPOSE OF THE INVESTIGATION This report describes the effects of the Borrego Mountain earthquake of significance to engineering geology and attempts to relate these effects to the local and regional geology. The field study focused on a relatively thorough, systematic investigation of the effects (particularly terrain effects) identified along a belt arbitrarily restricted to 8 km, extending from Coronado on the Pacific coast eastward to Glamis, Calif. (pl. 4) ; a few excursions were made outside this belt, both north and south of the primary route. The main traverse was coincident with State Highway 78 over much of its extent. This route was designed to pass through the epicentral region at a high angle to the Coyote Creek fault and thus to provide a selective and representative sample of effects over contrast- ing geologic terranes at steadily increasing distances from the epicenter. In order to establish some sort of standardized scale of effects, special attention was directed toward several guide features, partic- ularly bridges and post-World War II supermarkets. Spot checks were made for evidence of damage to 159 other types of structures. Our direct observations were supplemented by eyewitness reports, both writ- ten and oral, press accounts, and information collect- ed in a questionnaire canvass conducted by the Coast and Geodetic Survey. (See Cloud and Scott, 1968a, p. 1187; 1968b, p. 19; Seismological Field Survey, National Oceanic and Atmospheric Admin- istration, this volume.) Fieldwork began the day after the earthquake with a brief reconnaissance of effects between Ban- ning and Ocotillo Wells. Systematic field investiga- tions were carried out during the periods April 11—17, April 24—27, and May 9—11, 1968. About two- thirds of the fieldwork was concentrated in the epicentral region — that is, within an area extending 6—8 km to either side of the main tectonic trace. ACKNOWLEDGMENTS The cooperation of many individuals and a number of public agencies contributed immeasurably to the preparation of this report. We wish to thank in particular the staff members of the Engineering and Utilities Department of the City of San Diego, the County Engineer and Road Department of the County of San Diego, the El Centro Office of the California Division of Highways, and the Public Works Departments of Imperial County; several significant effects would have gone unrecognized had we not been alerted to their existence by these groups. K. C. Crowther and D. R. Nichols of the Geological Survey ably assisted in mapping a variety of terrain effects. Partial support was provided by the Division of Reactor Development and Technol- ogy, Atomic Energy Commission. REGIONAL GEOLOGIC SETTING The epicentral area of the Borrego Mountain earthquake is located along the east margin of the Peninsular Range province of southern California (Jahns, 1954b). This province occupies the south- west corner of California and extends southward into Baja California; it is bounded on the west by the Pacific Ocean and abuts the Coachella and Im- perial Valleys of the Salton Trough on the east (Jahns, 1954a). Jahns (1954b) has described the geology of the Peninsular Range province, and Dibblee (1954) has described the geology of the Imperial Valley region. The geology along the San J acinto fault, the domi- nant structural feature of the eastern Peninsular Range province, was described by Sharp (1967), whose summary of the tectonic setting of the Salton Trough appears elsewhere in this volume. The inland parts of the Peninsular Ranges (pl. 4) are underlain chiefly by igneous and metamorphic 160 rocks of Paleozoic and Mesozoic age. These rocks are in turn overlain with profound unconformity by irregular patches of volcanic rocks and nonmarine sedimentary deposits of Late Cretaceous to late Cenozoic age and, in the coastal areas, by Pleistocene marine terrace deposits. Along the east margin of the province, and within the Salton Trough, the older crystalline rocks are mantled by locally thick chiefly nonmarine upper Tertiary and Quaternary deposits. These rocks consist largely of fanglomerate and conglomerate, lacustrine clay and sand, and various volcanic assemblages. The upper part of this section, the Pliocene or Pleistocene Ocotillo Con- glomerate of Dibblee (1954, p. 24), grades eastward into the Pleistocene lacustrine Brawley Formation of Dibblee (1954, p. 24), consisting of claystone and thin interbeds of sandstone. The upper Pliocene and Pleistocene deposits of lacustrine origin that can be conveniently separated from other types of Cenozoic deposits are shown on plate 4 as chiefly Quaternary lake deposits; elsewhere these deposits are included with the Pleistocene to Upper Cre- taceous volcanic and nonmarine sedimentary de- posits. The lake deposits probably thicken to at least 600 or 800 m and perhaps 1,500—3,000 m beneath the Salton Trough (Dibblee, 1954, p. 24, pl. 2). Poorly indurated Quaternary dune sand, alluvium, and nonmarine deposits are grouped together as “Q” on plate 4. The thicknesses of these surficial deposits probably vary widely, bufi they are almost certainly very thin compared with the underlying lake deposits or Upper Cretaceous to Pleistocene sedimentary rocks. (See Dibblee, 1954, pl. 2.) The Peninsular Ranges are traversed by a con- spicuous series of steeply dipping northwest-trend- ing faults, many of which project into or beneath the Salton Trough (Jahns, 1954a, p. 19; 1954b, pl. 3; Dibblee, 1954, pl. 2). These faults are most num- erous in the area eastward from Santa Ysabel and Julian, and most, if not all, have been intermittently active during Cenozoic time (Jahns, 1954a, p. 19; 1954b, pl. 3; Dibblee, 1954, pl. 2). Movement on one such fault—identified as the San Jacinto fault by Dibblee (1954, pl. 2) and Jahns (1954b, pl. 3) but distinguished from the San J acinto and mapped as the Coyote Creek fault by Sharp (1967, pl. 1) —— ap- parently generated the Borrego Mountain earth— quake of 1968. TRANSIENT EFFECTS Transient effects of the Borrego Mountain earth- quake are described and discussed in detail else- where in this volume (Seismological Field Survey, National Oceanic and Atmospheric Administration; Waananen and Moyle) ; those reported from along THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 the route of our reconnaissance are indicated on plates 4 and 5. Selected aspects of engineering sig- nificance are summarized here. The most significant and most generally observed transient effect was the ground motion generated in response to the passage of seismic waves. Clearly perceptible shaking occurred over at least 155,000 sq km (Cloud and Scott, 1968a, 1968b ; Seismological Field Survey, National Oceanic and Atmospheric Administration, this volume), and two reports of felt ground motion came from Yosemite Valley, about 650 km from the epicenter and roughly 300 km beyond the generally felt area. Estimates of the duration of felt motion ranged from 1/2 to 1 minute in the epicentral region, from 1A1, to 1 minute in the metropolitan San Diego area, and from a few seconds to about 1 minute in the metropolitan Los Angeles area. Strong-motion rec- ords were not obtained within the epicentral area of this earthquake as they were, for example, for the Parkfield-Cholame earthquake of June 27, 1966 (Brown and others, 1967, p. 51). Accelerations (as measured for a single horizontal component) of 0.03 g or more, however, persisted for at least 26 sec, 65 km from the epicenter at El Centro (Cloud and Scott, 1968a, p. 1189; 1968b, p. 104). Because the attenuation of acceleration between the epicen- ter and a point 65 km distant has been given as about one-third (Cloud, 1968), strong motion (accelera- tions of 0.10 g or greater) may have persisted for as much as 26 seconds in the epicentral region. PERMANENT EFFECTS Permanent effects associated with the Borrego Mountain earthquake are classified (pl. 4) much as were those associated with the Parkfield-Cholame earthquakes of 1966 (Brown and others, 1967, p. 41—43). The basic classification is adapted from Richter (1958, p. 81) and designates as direct or primary those effects due directly to differential movement along fault surfaces and as indirect or secondary those effects due to shaking generated by the passage of seismic waves. The permanent effects are further classified according to object affected and as effects on terrain or manmade struc— tures. Insofar as is now known, all the permanent effects described were associated with the main shock. PRIMARY EFFECTS ON TERRAIN TECTONIC FRACTURES ALONG COYOTE CREEK FAULT Tectonic surface ruptures, produced chiefly along a 33-km segment of the Coyote Creek fault, formed the most dramatic and probably the most significant terrain effect associated with this earthquake. (See THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 pls. 4, 5.) Measurable slip along individual frac- tures was generally right lateral and was as much as about 38 cm. (See Clark, “Surface Rupture along the Coyote Creek Fault,” this volume.) Near the Ocotillo Badlands, the breaks were distributed main- ly along two narrow en echelon zones along the margins of the hills, but an especially complex pattern of surficial fractures also formed within the badlands. (See pl. 5.) Allen, Grantz, Brune, Clark, Sharp, Theodore, Wolfe, and Wyss (1968, p. 1185) observed that “* * * In every area where a pre-existing fault was indicated by a scarp or ground-water barrier, the rupture took place almost exactly along the line of earlier breaks.” However, because clear evidence of earlier faulting was absent along perhapshalf the ruptured trace of the Coyote Creek fault and because the ruptured zones were locally as much as several hundred meters wide, indications of earlier faulting constituted imperfect guides to the location of these breaks. The width of the main zone of surficial faulting and the distances of the secondary tectonic fractures from the main trace are greater than would have been anticipated for a shock of magnitude 6.4, particularly one associated with predominantly strike-slip movement. Thus, Bonilla (1967, p. 29, table 1) observed that the half widths of the main zone of tectonic fracturing associated with earth- quakes of up to magnitude 8+ (accompanying trans- current faulting) have been as much as about 0.09 km and that secondary, or spatially separated, tectonic fracturing is not known to have occurred more than 2.9 km from the main trace. The main zone of surficial faulting produced during the Bor- rego Mountain event was characterized by maximum half widths of about 1 km in the area of overlap near Ocotillo Badlands and as much as 0.15 km else- where; secondary tectonic fracturing (excluding that associated with the Banning—Mission Creek, Impe- rial, and Superstition Hills faults) occurred at distances of at least 2.5 km and perhaps 3 km or more from the main trace. (See Clark, “Surface Rupture along the Coyote Creek Fault,” this volume.) The character and distribution of the Coyote Creek tectonic fractures, with respect to both their geographic occurrence and various significant geo- logic features, are described in detail elsewhere in this volume by Clark (“Surface Rupture along the Coyote Creek Fault”) and by Sharp and Clark. TEMPORALLY ASSOCIATED TECTONIC FRACTURES The most surprising of the major terrain effects 161 associated with the Borrego Mountain earthquake consisted of minor horizontal displacements of up to about 2.2 cm on the Superstition Hills, Imperial, and Banning—Mission Creek (San Andreas) faults (see pl. 4) at distances of about 45, 70, and 50 km, respectively, from the epicenter (Allen and others, 1968, p. 1185). Although these effects were probably triggered by the April 9 shock on the Coyote Creek fault and are thus indirect or secondary (in the sense that their existence could be attributed to the passage of seismic waves), they are nonetheless primary in the sense that the breaks are certainly due directly to fault movement. If these displacements can be considered secondary tectonic fractures (tectonic fractures spatially sep- arated from the main fault zone — see classification of Bonilla, 1967, p. 5), they are spectacularly unique among recognized examples of this phenomenon. Secondary faulting was previously reported at maxi- mum distances of 13 km (associated with an esti- mated magnitude 8.3 shock), 13.7 km (associated with a magnitude 7.1 shock), and, possibly, 34 km (associated with a magnitude 8.5—8.6 shock) from the main fault (Bonilla, 1967, table 1). These extreme examples were characterized by dip-slip or oblique-slip movement on the main fault; secondary faulting associated with predominantly strike-slip movement had never before been recognized at distances greater than 2.9 km from the main fault (Bonilla, 1967, table 1). SECONDARY EFFECTS ON TERRAIN Terrain effects attributable to seismic shaking consisted chiefly of slope failures of several types and various compaction or liquefaction phenomena. Slope failures, including rockfalls, soil falls, rock- slides, slumps, and block glides (classification after Varnes, 1958), formed the largest group of second— ary terrain effects. They were widely distributed within the almost universally dry surficial deposits from San Diego to at least as far east as the central Imperial Valley. Compaction effects occurred chiefly in the epicentral region and scattered areas east of the epicenter. The largest slope failure observed during our reconnaissance was a reactivated slump or block glide near Seely in the west-central Imperial Valley (see pl. 4) ; its significance lies less in its size, perhaps, than in its distance (65 km) from the epicenter. Soil falls apparently were generated over hundreds of square kilometers; the distribution of the resulting dust clouds closely matched the courses of many of the major arroyos over the area extend- ing eastward from Borrego Springs to the Salton Sea. 162 ROCKFALLS AND SOIL FALLS Rockfalls were unrecognized over the main route of our reconnaissance. They were dramatically de- veloped elsewhere, however, particularly in Split Mountain Canyon (about 15 km south of Ocotillo Wells), where several automobiles and camper vehicles were seriously damaged. Rockfalls and (or) soil falls attributable to the earthquake were also reported from the Sunset Cliffs district of San Diego; these falls, however, involved materials being actively undercut by wave erosion. Soil falls (falls composed of materials too poorly lithified to be identified as rock) were generated within many of the bedded alluvial deposits along the generally shallow but almost universally steep- sided arroyos that characterize this part of the southwest. They occurred at least as far west as Yaqui Well, about 25 km west of the epicenter, but they were only sparsely developed east of there to within 3 km of the main trace of the Coyote Creek fault. They persisted in abundance eastward from the fault for many km and were observed as far east as the south tip of the Salton Sea. The relatively few soil falls west of the Coyote Creek fault, as contrasted with their abundance east of this break, may correlate with an apparently abrupt increase in the thickness of the poorly consolidated sedimentary cover. (See Hamilton, 1970, p. 378.) Soil falls were particularly common along major washes and arroyos and in the badlands sections of the epicentral area. The most impressive soil falls or rockfalls were seen in the Fonts Point region of the Clay Hills, north of Borrego Mountain. Incipient failures of this type are shown in figure 108, where a series of blocks has been partly detached from the rim of this steep badlands slope. These particular failures are difficult to classify and could just as easily have been designated as slumps, for they clear- ly have been rotated or displaced laterally. One of the most intense concentrations of soil falls was along Fish Creek Wash (fig. 109), south of Ocotillo Wells. These falls were distributed west- ward along the banks of the wash almost without interruption for at least 2—3 km beyond the ruptured trace of the Coyote Creek fault. They presumably continued eastward for an equal distance; we did not, however, examine this terrain. The material in which Fish Creek Wash is incised consists of generally poorly sorted and poorly indurated but well-jointed sandy alluvium. (See fig. 109.) ROCKSLIDES AND DEBRIS SLIDEs Rockslides occurred along a number of cuts in the strongly foliated and highly fractured crystal- line rocks cropping out along the Montezuma-Bor- THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 FIGURE 108. — Rockfalls and soil falls in lacustrine silts and sands of probable Pliocene age. A, Complex rockfall and debris slide in the central Ocotillo Badlands. This failure is typical of many of those found in the badlands. B, Incipient rockfalls or soil falls in the Clay Hills, near Fonts Point, about 3 km north of the north end of the ruptured fault trace. rego highway, leading west from Borrego Springs. Many of the slides were along dip slopes, but the highly fractured character of the rock, which surely was aggravated by blasting during construction, THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 163 FIGURE 109.—Soil falls produced in sandy alluvium along the near-vertical walls of Fish Creek Wash, immediately west of the main trace of the Coyote Creek fault. must have contributed substantially to the failures. (Rockslides closed this route to traffic for a number of hours after the main shock; this was the only knOwn example of a road closure attributable to the earthquake.) Rockslides or debris slides (together with asso- ciated soil falls) also occurred locally on the over- steepened natural slopes of the Ocotillo Badlands underlain by the folded Borrego Formation of Tarbet and Holman (1944) and Ocotillo Conglomerate of Dibblee (1954), both of Quaternary age. (See pl. 1.) These failures, an example of which is shown in figure 108A, were generally too complex to be classi- fied simply as falls, slumps, or slides. SLUMPS Slumps and incipient slumps of various types were the most prevalent slope failures recognized in the epicentral region, if not within the entire area we examined. Most were incipient failures, in the sense that they were generally characterized by open cracks or fissures of the sort shown in figure 11.0 rather than by well-developed headwall scarps or clearly defined toes. Although slumps attributable to the earthquake were identified many kilometers from the epicenter, they were most abundant on the gentle to moderate slopes underlain by the Ocotillo Con- glomerate of Dibblee (1954), the Borrego Formation of Tarbet and Holman (1944), and the Palm Spring Formation (Sharp, “Tectonic Setting of the Salton Trough,” this volume) within and immediately north of the Ocotillo Badlands (fig. 110). The dens- ity of slump or incipient slump features was gen— erally very low more than 3—5 km from the trace of the Coyote Creek fault. We were unable to identify any good correlation between the distribu- tion of tectonic ruptures and slumps within the FIGURE 110,—Incipient slump in the low-relief hills north of the Ocotillo Badlands. 3- to 5-km-wide zone in which both were well developed. What was evident, however, was an al- most complete restriction of the slumps to the area east of the mapped tectonic ruptures. (See pl. 5.) Older landslides within the zone of intensive slump- ing were nonexistent or at least inconspicuous. The general configuration of the hills, however, sug- ' gests that slumping may have been locally operative for a long time. The boundary between the hills and the surrounding alluvial lowlands forms a con- voluted pattern in plan View that, together with the prominently steepened ridge crests toward the ends of their respective spurs, suggest great viscous flows edging out upon the adjacent plain. The best expressed and generally largest slumps were produced in the Ocotillo Badlands; a repre- sentative example of this terrain is shown in figure 111. The scarp shown in figure 112 could be traced for nearly 0.5 km; smaller, arcuate scarps of the sort shown in figure 1123 were the most commonly recognized slump features in both the Ocotillo Bad— lands and the hills immediately to the north. 164 FIGURE 111.-—— Gently rolling hills and steep-sided washes of the central 0cotillo Badlands characterized by abundant slope failures. View northeast. FIGURE 112.—Slump scarps in the gently to steeply folded Pleistocene 0cotillo Conglomerate of Dibblee (1954) along or near the crests of hills in the epicentral area. A, Failure along east-west-trending ridge in the Ocotillo Badlands, about 2 km southeast of Ocotillo Wells. This scarp extended for more than 0.5 km along the ridge crest. Photograph by E. W. Wolfe. B, Typical scarp produced in the gently sloping hills about 3 km east of 0cotillo Wells. THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 A reactivated slump or block glide Within the gen- erally silty Quaternary lacustrine deposits, about 65 km southeast of the epicenter at the Highway 80 crossing of the New River, constituted the largest single slope failure identified during the postearth- quake investigation. This failure was expressed as a simple break (fig. 113), about 125 m east of and roughly parallel to the river; vertical separations along this break of as much as 5 cm formed a sub- dued headwall scarp that extended south from Highway 80 for more than 0.4 km. Slumps were recognized in artificial fills near the south end of the Santa Rosa Mountains and in several areas around the Imperial Valley. The most significant of these consisted of small- to moderate- sized slumps in engineered fills forming parts of building pads in the Salton City area and along a recently completed highway between Borrego Valley FIGURE 113.—Headwall scarp of slump or block glide im- mediately east of Highway 80 crossing of the New River. Vertical separations of as much as 5 cm (see displaced curbs) occurred along a 125-m segment of this break. THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 FIGURE 114.—Slump in artificial fill along a recently com- pleted highway north of the Borrego Badlands. and Salton City (fig. 114). Slumps were also gen- erated in wet and possibly saturated soils forming fills or fill base along the edge of the Salton Sea and along river and canal embankments in the Im- perial Valley bottomlands. COMPACTION AND LIQUEFACTION EFFECTS Seismic compaction and liquefaction effects at- tributable to the earthquake were discovered at a number of widely separated points. Perhaps the most intriguing of these were the linear cracks and differential displacements produced chiefly along the margins of several recently irrigated fields. These ruptures were particularly conspicuous in the sandy alluvial deposits of the epicentral region, where they occurred in some measure within or around every well-wetted field we examined (figs. 115 and 116). Probably the least expected example of compaction- induced rupturing was found near the center of a thoroughly soaked alfalfa field in the Imperial Val- ley, about 15 km west-northwest of Westmorland, or more than 30 km east of the epicenter. Shaking of the silty Quaternary lake beds on which the field was established seemingly generated a small graben about 3 m wide, 150 m long, and characterized by vertical displacements of as much as 15 cm (fig. 117). Sand boils, the only direct evidence of seismically elevated pore-water pressures and soil liquefaction (or near liquefaction) that we discovered, were seen in only two areas: within a concentric array of compaction fractures in the epicentral area and landward of slumps along the New River northwest 476-246 0 - 72 - 12 165 \ \ \ \‘Z‘EA \ I) \%\\ \\60/\ \3\ \°2.\ \ \ \ WM 3 5 10 K 5-8 5 ~wfi" 0 I0. 20 30 4o 50 METERS \O, \ \‘5 \ \6\ EXPLANATION \‘a\ .‘_ r- \ . Open crock \‘a \ MEM- Open crack with vertical separation Showing amaun/ of yer/Ita/ movement in cam/meters. Hac/mres on down/hrawms/de —x——x——x—n——x— Fence FIGURE 115.—Sketch map showing distribution of seismic- compaction ruptures (see fig. 116) around irrigated field adjacent to main trace of the Coyote Creek fault, about 1.5 km southeast of Ocotillo Wells. Note that the cracking associated with branch fault zone does not project through field. of El Centro (fig. 118). The second association sug- gests that this slumping, at least, was attributable in whole or in part to liquefaction of the underlying materials. MISCELLANEOUS TERRAIN EFFECTS Many of the poorly indurated surficial deposits, such as dune sands, alluvium, and artificial fill, found within the western Imperial Valley are veneered with thin lightly cemented brittle crusts. Not sur- prisingly, these crusts were locally shattered by seismic shaking (fig. 119). The distribution of this effect, however, was seemingly random and largely unpredictable. The occurrence of shattered dunes south of the Ocotillo Badlands (fig. 119), for example, was apparently unrelated to their location with respect to the ruptured trace of the Coyote Creek fault. The most puzzling expression of crustal shattering discovered during our reconnaissance oc- curred about 13 km east of Ocotillo Wells (and about 6 km from the nearest strand of the Coyote Creek fault). The shattering there was confined to a linear zone that was as much as 100 m wide and extended for about 0.4 km (fig. 120). This peculiarly re- stricted zone, which was produced on a nearly featureless and gently sloping surface overlying ancient lake deposits, led eastward from the base of 166 FIGURE 116.—Seismic-compaction ruptures along the pe- riphery of an irrigated field adjacent to the main trace of the Coyote Creek fault. (See fig. 115.) Soil sample from this field had a mean particle size of 0.26 mm and a uniformity coefficient of 6. A, Parallel array along south side of field. B, Rupture along the north side of the field on which maximum vertical separation was about 10 cm. THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 FIGURE 117.—Fifteen-cm scarp (at arrow) along the edge of a compaction-produced graben in the center of an alfalfa field on the Elmore Ranch, 22 km west-northwest of \Vestmorland (pl. 4). a shattered sand dune and veered northward across Highway 78 near its east terminus. Differential displacements were unrecognized along the shattered zone, even where it was faintly expressed as fresh cracks in the asphalt pavement of Highway 78. Furthermore, we could in no way define the zone other than through the occurrence of the shattered crust itself. Thus, because this effect seems to be only very generally related to any evident geologic feature, its localization may be due to a peculiar focusing of seismic energy. Dislodged or displaced pebbles and cobbles (fig. 121) were found throughout much of the epicentral region. Their relative abundance showed a good cor- relation with that of the slumps and incipient slumps in the low hills of the epicentral region. THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 FIGURE 118.—Sand boils in late Quaternary lake-bottom deposits along the New River, about 55 km southeast of the epicenter. PRIMARY EFFECTS ON STRUCTURES Significant primary effects on manmade structures were limited almost entirely to rending or cracking of roads and highways. Chief among these was the cracking, displacement, and partial rotation of the pavement at the intersection of Highway 78 with the ruptured trace of the Coyote Creek fault. Minor cracking was also detected at a number of road intersections with the trace of the Imperial fault (Allen and others, this volume). Additional minor primary effects on structures included the rupture and offset of a 2.5-cm water pipeline that intersected the Coyote Creek fault about 1.5 km southeast of Ocotillo Wells and the extension and bending of an airplane tiedown cable at the Ocotillo Wells county airport on Benson Dry Lake. (See pl. 5.) SECONDARY EFFECTS ON STRUCTURES Secondary effects on manmade structures, al- though widespread, were generally minor. They were concentrated in two general areas: the epicentral region around Ocotillo Wells and the Salton City area, about 22 km northeast of the epicenter. Al- though our sample was poor owing to the sparse construction in this area, it is clear that the most 167 FIGURE 119. — Shattered lightly cemented crust overlying poorly indurated materials in the epicentral area. A, Sand dune south of the Ocotillo Badlands. B, Artificial fill about 1.5 km west of Ocotillo Wells. severe damage away from the immediate area of the epicenter occurred on the old, chiefly Quaternary lacustrine deposits that underlie most of the Imperial Valley. Damage west of the Coyote Creek fault, 168 HIGHWAY 78 Sand dune 0 IO 20 30 40 50 METERS THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 Freshly cracked Pavement EXPLANATION xzoe Penetration test location, showing cone index 75-cm dep/h, Corps of Engineers cone penetrame/er FIGURE 120.—Map showing distribution of shattered brittle crust overlying generally sandy lake-bottom(?) deposits about 13 km east of Ocotillo Wells. Relative difierences in penetration resistance indicated by Corps of Engineers cone penetrometer measurements along and across zone. FIGURE 121.—Cobb1es dislodged or displaced by seismic shaking, about 2.5 km east of Ocotillo Wells. particularly among the crystalline rocks of the Peninsular Ranges, was generally very light (pl. 4). Damage that impaired the safety or utility of “ structures occurred from San Diego east to at least Brawley and Calexico. (See Cloud and Scott, 1968a, p. 1190.) Frame structures, other than those that were poorly built or were particularly old, sustained very little damage. Unreinforced masonry structures were more severely affected, but we saw only two masonry failures that could be described as “im- pressive,” and both of these were well removed from the epicenter. Toppled or otherwise damaged chimneys were reported from as far west as Ramona (unconfirmed by our observations). We have no reports, however, of damaged chimneys east of the epicentral area or of damage to modern reinforced chimneys from any locality. The partly collapsed fieldstone chimney in Ocotillo Wells, shown in figure 122, was the most severely affected of any we observed. Masonry walls sustained damage over an area ranging from Borrego Springs to at least as far east as Brawley and as far south as El Centro. The most spectacular failures occurred in pre-1940 unrein- forced brick and adobe walls in Westmorland (fig. 123) and Seeley, 45 and 65 km east and southeast, respectively, of the epicenter. Unreinforced field- stone walls, which formed parts of two abandoned structures near Ocotillo Wells, collapsed almost completely during the earthquake. One of these structures (fig. 124), an abandoned cellar, was over- lain by a concrete slab roof, the inertial effect of which may have contributed to the failure. Lightly damaged reinforced concrete-block walls were ob- served from Borrego Springs east and southeast as far as Salton City and Brawley, respectively. Be- cause nearly all the supermarkets in this part of California are built of concrete blocks and because we examined a very large percentage of these, our knowledge of the distribution of this type of damage probably is good. Breakage in the block walls gen- erally consisted of cracking along construction THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 FIGURE 122.——Partly toppled fieldstone chimney in com- munity of Ocotillo Wells. FIGURE 123.—Partly collapsed unreinforced brick wall in Westmorland. joints, but individual blocks were rent locally (fig. 125). We have no substantiated reports of even partly collapsed reinforced block walls. Cracked plaster or stucco walls were the most common type of damage incurred by frame struc- tures. The westernmost examples of cracked plaster walls (other than hairline breaks associated with little if any debris) were seen in a pre-World War 11 Lakeside drugstore, just northeast of San Diego. Both interior plaster and exterior stucco walls were damaged enough to require repair. Plaster and stucco walls of frame structures in the Salton City area, none of which is probably more than 10 years old, seemed particularly susceptible to cracking and 169 FIGURE 124.—Collapsed concrete-roofed fieldstone cellar about 1 km east of Ocotillo Wells. FIGURE 125. — Cracked reinforced concrete-block wall in Sal- ton City building. spalling. Most of the cracking in the Salton City area Was minor, but the walls of a two-story wood- frame motel in the center of town were extensively damaged (fig. 126). The cracks in the motel general- 170 THE B FIGURE 126.—-Damage to plaster and stucco walls of a two-story motel in Salton City. A, Cracked interior wall. B, Cracked stucco wall. Note cracks radiating away from corners of sashwork and between floors. BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 ly followed the studs, jumped from stud to stud along straight lines (fig. 126A), or radiated away from sashwork corners (fig. 1263) ; the most conspicuous damage occurred on the first floor, particularly at the interior corners between wings. Separated or rent frame walls, partitions, or other structural units were seen in Salton City, Ocotillo Wells, and Borrego Springs. One very poorly built frame house in Ocotillo Wells was characterized by shifted, rocked, or possibly vertically lifted and projected walls. The displacement of a wooden col- umn at its junction with an overhead beam in this same house very nearly led to the collapse of a section of roof. Nonetheless, damage of this sort was very light, and the only examples of separated or rent frame units that we saw were ones in which the unit was poorly bonded to the adjacent member. Broken windows, which were observed locally over nearly the entire extent of our traverse, were, nevertheless, not generally evident. Press accounts of “hundreds of broken windows” in the San Diego area (Cloud and Scott, 1968a, p. 1190) may have been exaggerated, for a single cracked plate-glass window in an office building at 2d and Broadway was the only example of a broken window that we saw or were reliably informed of in the metropolitan San Diego area. Nearly all the highway bridges between San Diego and Glamis are of modern design and construction; most are supported by heavy wooden piles or con- crete piers. Of the tens of bridges we examined, only the monolithic reinforced concrete bridge at the Highway 78 crossing of San Felipe Creek sus- tained damage that could be fairly definitely asso- ciated with the earthquake. The concrete piers supporting this bridge were circumferentially cracked and slightly spalled at their junctions with the overlying pier caps (fig. 127); although the cracks seemed to be equally distributed about the piers, the spalling was most conspicuous on the northwest sides of the piers. This particular damage constituted the only evident deterioration of a large, massive, modern structure discovered in our in- vestigation. The only other shaking effects associated with the highway bridges along routes 78 and 99 were slumps or cracks within or along the margins of the abutments and aprons and cracking of the soil around the piers. The stream-bottom deposits within the arroyos intersected by Highway 78 about 10 km east of the San Felipe bridge, for example, displayed concentric and radiating arrays of cracks around the supporting bridge piers. It was not al- ways possible, however, to be certain that these cracks and slumps were derived entirely from seis- THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 171 FIGURE 127.—-Cracked and spalled 40-cm-diameter concrete pier supporting the Highway ‘78 bridge over San Felipe Creek, about 8 km east of Ocotillo Wells. mic shaking rather than from heavy vehicular pounding. A railroad bridge was damaged several hundred yards southeast of Glamis along the main line of the Southern Pacific Railroad (pl. 4). The con— crete-abutment retaining walls supporting this bridge reportedly displayed 1):, to 1-cm—wide cracks prior to April 9; these expanded at the time of» the earthquake into the 8- to 10-cm-wide cracks shown in figure 128. Ruptured buried concrete pipes on Jacobs Ranch east of Ocotillo Wells and cracked concrete canal linings on the Elmore Desert Ranch west of West- morland (see pl. 4) constituted the only other type of secondary structural damage of which we are aware. The cracks in the canal linings paralleled both the contour of the lining slope and fresh cracks in the adjacent soil; both sets of cracks probably were generated by free-face movements of the canal embankments. EFFECTS ON LOOSE OBJECTS Loose objects (those unrestrained or only partly restrained against free movement) are classified as heavy (for example, furniture, safes, automobiles) or light (for example, shelf goods, lamps, whisky bottles). Although heavy objects within the epi- central region commonly were displaced or shifted, we have few reports of this sort of movement more than 4 or 5 km from the epicenter. Shifting or pro- jection of light objects, on the other hand, occurred over almost all of our traverse. An abandoned kitchen stove in Ocotillo Wells that had been flipped upside-down (R. V. Sharp, written commun., 1970) and toppled bales of hay, about 3.5 km south of Ocotillo Wells, were the only reported examples of overturned heavy objects. ‘ ,, FIGURE 128.—Cracked concrete railroad bridge abutment about 0.3 km southeast of Glamis. Gaping cracks shown here reportedly widened at the time of the earthquake from 1/2- to l-cm openings to 8-10 cm. Shifted or displaced heavy objects, however, abound- ed within the epicentral region. Automobiles com- monly were shifted about, and a number of house trailers parked in the Ocotillo Wells area were knocked off their parking jacks or other light foun- dations. Large transformers at the San Felipe sub- station, about 5 km south-southeast of Ocotillo Wells, were shifted with such force as to shear their anchor bolts (Cloud and Scott, 1968a, p. 1189). Stoves, refrigerators, and various types of furniture slid around within the stores and dwellings of Ocotillo Wells; interviews with local witnesses in- dicate that these objects were displaced preferential- ly toward the northeast (that is, toward the trace of the Coyote Creek fault). Away from the epi- center, a large high-pressure fuel tank inside a restaurant about 5 km east of Borrego Springs shifted or pivoted about 8 cm to the south-southeast, 172 and heavy roller-mounted vaults in a Salton City bank moved about 25 cm west-southwest. Vibrational penetration of automobiles into loosely compacted sands occurred near the main trace of the Coyote Creek fault. About a dozen heavy vehicles parked near the Morton house, about 1.5 km south- east of Ocotillo Wells, were shaken into the ground to depths of 8—10 cm (fig. 129). FIGURE 129. — Light truck that sank several centimeters into the dry desert sand, owing to vibratory ground motion, about 1.5 km southeast of Ocotillo Wells. Projected or displaced light objects were reported from Coronado to about 8 km east of Brawley; they were particularly conspicuous near the epicenter (fig. 130). Interviews with three residents of Glamis (virtually the entire population) failed to elicit a single report of displaced objects at this most east- erly point of our reconnaissance. Goods shaken from shelves were both the commonest and most wide- spread permanent effect discovered. Two thousand dollars worth of breakage in a Salton City liquor THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 FIGURE 130.—Projected and displaced objects about 1.5 km southeast of Ocotillo Wells. All those objects for which the height of the center of gravity was equal to or greater than the diameter of the base were knocked over. Glass tumbler nearest the camera was placed in upright posi- tion before the picture was taken. store was, in fact, the single most distressing effect of the Borrego Mountain earthquake. The distribution of displaced and projected light objects is of interest. We have reports of extensive breakage in the metropolitan San Diego area from Chula Vista north to Mission Bay and reports of limited breakage in supermarkets in central El Cajon and various stores in Lakeside and Ramona. (See pl. 4.) All these areas are underlain by Quater- nary terrace deposits, beach or lagoonal deposits, or alluvium that range from a few to perhaps hundreds of meters in thickness. There were, by contrast, startlingly few reports of fallen objects from Ramona eastward to and including the west edge of the Borrego Valley; this region is underlain almost entirely by crystalline rocks. The occurrence of fallen objects diminished generally eastward from the epicenter, but local pockets of breakage were recognized in Salton City, Westmorland, Imperial, and El Centro, all of which are underlain by late Cenozoic lacustrine deposits several thousand meters thick. Several small-scale associations between the distri- bution of fallen objects and certain physiographic features attracted our attention. A relatively new supermarket about 0.3 km west of the center of Lakeside had no more than about $10 worth of breakage, whereas a similar supermarket near the center of town apparently sustained extensive break- age (exact amount unknown). The only apparent difi‘erence between the two situations was the loca- tion of the second market on what appeared to be a slightly lower (by 3—4 m) river terrace. The distri- bution of breakage in Brawley was particularly THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 interesting. The modern concrete-block Mayfair Market here sustained about $1,000 worth of break- age, yet according to the manager, breakage was minor in similar stores elsewhere around Brawley. The Mayfair Market may be unique because it is both the westernmost of the Brawley supermarkets and only about 0.6 km east of the high (approxi- mately 12 m) steep-banked New River flood plain. INFERRED INTENSITY Maximum intensities of at least VIII and possibly IX (Modified Mercalli) were attained during the earthquake. The highest intensities occurred within a 3—6-km-wide band along the ruptured 'trace of the Coyote Creek fault and along the west edge of the Salton Sea near Salton City. Intensities of VIII are indicated by partial loss of control of moving automobiles, minor damage to reinforced masonry, collapsed chimneys, cracked stucco walls, local changes in well flow, and cracks on steep slopes. In- tensities of IX may be indicated for the epicentral region by the destruction of poorly constructed masonry walls, separated or rent frames, broken underground pipes, local ground cracking, sand boils, and landslides in dry soils. (See Richter, 1958, p. 136-138.) Although some effects, such as large slumps in the Ocotillo Badlands, suggest intensities even greater than IX, the general survival of frame structures and the absence of serious bridge damage argue that intensities of X probably were not at- tained anywhere within the area we investigated. Intensities midway between VIII and IX were correlated with magnitude 6.4 shocks by Gutenberg and Richter (1956, p. 132), a correlation in agree- ment with what has been observed for this earth- quake. Gutenberg and Richter (1956, p. 131) also correlated intensities of VIII and IX with ground accelerations of 0.20 g.and 0.30 g, respectively. These correlations, in turn, are in good agreement with epicentral accelerations of about 0.22 g correlated with magnitude 6.4 shocks (Lockheed Aircraft Corp. and Holmes and Narver, Inc., 1963, p. 14). Epi- central accelerations of 0.35 g to 0.40 9 may be de- duced from accelerations measured at El Centro and attenuation curves presented by Cloud (1968) ; however, we have assumed in using these attenua- tion curves that differing initial accelerations decline proportionately with distance, and this procedure may have led us to overestimate epicentral accelera- tions. Clark (“Intensity of Shaking Estimated from Displaced Stones,” this volume), nevertheless, con- cluded from observations of displaced pebbles that horizontal accelerations approaching 1 9 may have been attained locally within the epicentral region. 173 Our field investigation indicates that intensities of VIII or more were attained over at least 500—600 sq km and that intensities of VII or greater were attained over several thousand square kilometers. Isoseismal maps are presented elsewhere in this report. (See Seismological Field Survey, National Oceanic and Atmospheric Administration, this volume.) CONCLUSIONS The occurrence of the Borrego Mountain earth- quake in one of the least developed sections of south- ern California scarcely minimizes the significance of this event to engineering geology and foundation engineering. The earthquake effects of greatest im- portance to these two disciplines are as follows: 1. The earthquake was felt over at least 155,000 sq km and was characterized by maximum inten- sities of VIII—IX within a narrow band ex— tending 32 km or more along the Coyote Creek fault. The inferred epicentral intensities are in close agreement with those predicted for a magnitude 6.4 shock. 2. The main shock was accompanied by the forma- tion of a locally complex zone of tectonic sur- face ruptures that extended about 32 km along the Coyote Creek fault. Zones of this complex- ity have not generally been reported for earth- quakes of magnitude 6.4 or less. The precise locations of the ruptures, moreover, could have been predicted on the basis of ordinary field criteria over no more than about one-half their extent. 3. Temporally associated surficial fault displace- ments were generated along the Superstition Hills, Imperial, and Banning—Mission Creek faults. Although very slight, they could have damaged sensitive structures situated on the fault trace. It now seems likely that even rela- tively distant shocks may trigger displacements along any discontinuities identified with accu— mulated elastic strain of whatever source. 4. The epicentral region was characterized by a variety of dry-soil slope failures and compac- tion effects that could be reasonably expected to accompany a magnitude 6.4 shock. Direct and indirect effects of soil consolidation gener- ated at distances of 30—60 km from the epi- center were less expected; these distant features were all underlain by probably thick upper Cenozoic lacustrine sediments. 5. Structural damage was generally minor for an earthquake of this size and was concentrated in the epicentral region and over the lacustrine deposits in and around Salton City. It seems 174 unlikely that this light damage can be due entirely to the limited development of the re- gion; it may be equally attributable to a rela- tively short duration of strong shaking or to generally good natural foundations beneath those structures close to the epicenter. The fortunate location of this shock should not, as we have observed, minimize its significance to those concerned with most aspects of earthquake engineer- ing. Had the locally complex pattern of tectonic surface ruptures and the various shaking effects associated with this shock occurred in a densely populated region (especially one characterized by water-saturated ground), the damage would cer- tainly have been much greater and locally severe. REFERENCES CITED Allen, C. R., Grantz, Arthur, Brune, J. N., Clark, M. M., Sharp, R. V., Theodore, T. G., Wolfe, E. W., and Wyss, Max, 1968, The Borrego Mountain, California, earth- quake of 9 April 1968—A preliminary report: Seismol. Soc. America Bull., v. 58, no. 3, p. 1183—1186. Bonilla, M. G., 1967, Historic surface faulting in continental United States and adjacent parts of Mexico: U.S. Geol. Survey open-file report, 36 p. (also U.S. Atomic Energy Comm. Rept. TID—24124, 36 p.). Brown, R. D., Vedder, J. G., Wallace, R. E., Roth, E. F., Yerkes, R. F., Castle, R. 0., Waananen, A. 0., Page, R. W., and Eaton, J. P., 1967, The Parkfield-Cholame, California earthquakes of June—August 1966—Surface geologic effects, water-resources aspects, and preliminary seismic data: U.S. Geol. Survey Prof. Paper 579, 66 p. Cloud, W. K., 1968, Strong-motion seismograph records, in State-of—the-art symposium, earthquake engineering of buildings: Earthquake Eng. Research Inst. Mtg., San Francisco, Feb. 5—6, 1968, Program. Cloud, W. K., and Scott, N. H., 1968a, The Borrego Mountain, California, earthquake of 9 April 1968—A preliminary engineering Seismology report: Seismol. Soc. America Bull., v. 58, no. 3, p. 1187—1191. 1968b, Field reports and isoseismal map, in Strong- motion instrumental data on the Borrego Mountain earth- THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 quake of 9 April 1968: Seismological Field Survey of U.S. Coast and Geodetic Survey and Earthquake Eng. Lab. of California Inst. Technology, p. 19—119. Dibblee, T. W., Jr., 1954, Geology of the Imperial Valley region, California, in Jahns, R. H., ed., Geology of south- ern California; chap. 2, Geology of the natural provinces: California Div. Mines Bull. 170, p. 21—28. Gutenberg, Beno, and Richter, C. F., 1956, Earthquake mag- nitude, intensity, energy, and acceleration, 2d paper: Seismol. Soc. America Bull., v. 46, no. 2, p. 105—145. Hamilton, R. M., 1970, Time-term analysis of explosion data from the vicinity of the Borrego Mountain, California, earthquake of 19 April 1968: Seismol. Soc. America Bull., v. 60, p. 367—381. Jahns, R. H., 1954a, Investigations and problems of southern California geology, in Jahns, R. H., ed., Geology of southern California; chap. 1, General features: Califor- nia Div. Mines Bull. 170, p. 5—29. 1954b, Geology of the Peninsular Range province, southern California and Baja California, in Jahns, R. H., ed., Geology of southern California; chap. 2, Geology of the natural provinces: California Div. Mines Bull. 170, p. 29—52. Jennings, C. W., compiler, 1967, Geologic map of California, Olaf P. Jenkins edition, Salton Sea sheet: California Div. Mines and Geology, scale 1:250,000. Lockheed Aircraft Corp. and Holmes and Narver, Inc., 1963, Nuclear reactors and earthquakes: Div. Reactor De- velopment, U.S. Atomic Energy Comm. Rept. TID—7024, 411 p. Richter, C. F., 1958, Elementary seismology: San Francisco, Calif., W. H. Freeman and Co., 768 p. Rogers, T. H., compiler, 1965, Geologic map of California, Olaf P. Jenkins edition, Santa Ana sheet: California Div. Mines and Geology, scale 1:250,000. Sharp, R. V., 1967, San Jacinto fault zone in the Peninsular Ranges of southern California: Geol. Soc. America Bull., v. 78, no. 6, p. 705—730. Strand, R. G., compiler, 1962, Geologic map of California, Olaf P. Jenkins edition, San Diego—~El Centro sheet: California Div. Mines and Geology, scale 1:250,000. Varnes, D. J., 1958, Landslide types and processes, chap. 3, in Eckel, E. 13., ed., Landslides and engineering prac- tice: Natl. Research Council, Highway Research Board Spec. Rept. 29, NAS—NRC Pub. 544, p. 20-47. INTENSITY 0F SHAKING ESTIMATED FROM DISPLACED STONES By MALCOLM M. CLARK U.S. GEOLOGICAL SURVEY ABSTRACT Characteristics of stones displaced during the Borrego Mountain earthquake of April 9, 1968, allow rough esti- mates of the accelerations responsible for displacement. Areas of displaced surface stones were generally restricted to certain high ridges in Ocotillo Badlands, indicating that topography magnified the intensity of shaking. Nearly every , unrestrained stone on the surface of one such ridge was either tipped, slid, or flung as much as 0.5 m in one direction, whereas only cobbles and boulders taller than about 50 mm moved on an adjacent ridge, sliding a maximum of 0.1 m in the opposite direction. These directions of displacement indicate that the horizontal shaking responsible was oriented 70°—80° from the general trend of surface breakage; however, the influence exerted on these directions by the position, shape, or material of the ridges is unknown. Theoretical considerations indicate that in the absence of adhesion, simple sliding of a rectangular block depends only on the values of coefficient of friction, slope, and acceleration and is independent of size and shape. Tipping depends only on slope, shape, and acceleration. Size of the block affects sliding or tipping only if adhesion or other surface forces are present or if rocking has begun. Estimates of coefficients of friction, combined with char- acteristics of size, shape and movement, suggest that hori- zontal accelerations on the first ridge were greater than 0.5 g and possibly as high as 1 9, whereas vertical ac- celerations were definitely less than 1 g. Stones of the second ridge experienced smaller accelerations, with the horizontal component apparently in the range of 0.4—0.9 g. Adhesion between stones and the surface of the second ridge prevented stones shorter than 50 mm from moving. INTRODUCTION This chapter describes and evaluates the implica- tions of stones near the surface rupture of the Coyote Creek fault that were displaced by shaking during the Borrego Mountain earthquake of April 9, 1968. Such stones offer a crude measure of intensity of shaking (as in Oldham, 1897, p. 129—133), and an analysis of the stones and their displacements per- mits some general statements about the accelerations responsible for displacements. The chapter presents observations of such stones in Ocotillo Badlands, an analysis of how stones may react to shaking, and estimates of local accelerations in the badlands. T. L. Youd, R. V. Sharp, and R. 0. Castle pointed out flaws and omissions in the analysis and made many suggestions for improvement of the manu- script. LOCATION OF DISPLACED STONES In general, surface gravels are not abundant along the 1968 break; however, a lag of granitic and meta- morphic gravel covers most of the low hills in the vicinity of the break and much of the ridges and lower flanks of Ocotillo Badlands. Scattered patches of gravels left by modern debris flows also occur locally near washes elsewhere along the break. Few of these surface gravels anywhere in the re- gion of the break were moved by shaking of the earthquake, except on several high parts of ridges of Ocotillo Badlands (for example fig. 131A, B, and C), where a majority of the pebbles, cobbles, and boulders resting on the surface were rolled, slid, or flung from their preearthquake positions. In con- trast, the gravels that cover most of the lower slopes of the Badlands did not move, except for those at location E, figure 131. According to R. V. Sharp (oral commun, 1969), about 20 percent of the sur- face stones at location E moved down the 10° slope on which they rested. Also, Castle and Youd (this volume) report displaced stones near slump frac— tures elsewhere in Ocotillo Badlands and in the low hills to the northeast. The ridges with displaced stones in Ocotillo Bad- lands evidently experienced significantly higher accelerations than elsewhere. As noted by Clark (“Surface Rupture Along the Coyote Creek Fault,” this volume), these same hills were literally shat- tered with small landslides and faults. Although the abundance of landslides and slips is probably a manifestation of magnified accelerations, the moved stones do not appear to have been directly affected by such landsliding. Several slides originated on ridges containing the moved stones described in this study, but none of these stones rested on such slides. 175 176 Base from US. Geological Survey l:24,000 Borrego Mountain SE,1958 and Shell Reef, I959 FIGURE 131.——Area of moved surface gravels. A and B, Location of ridges on which many stones were displaced by shaking. C, Ridge containing many displaced stones but which was not investigated. D, High ridges on which virtually no stones were displaced. E, Only area where many boulders were moved that was not on a ridgetop. Lines show surface ruptures associated with the earthquake of April 9, 1968. DISPLACED STONES OF OCOTILLO BADLANDS The distribution of the areas of abundant dis- placed stones in Ocotillo Badlands appears to be irregular. Stones possibly shifted on high ridges not visited, in addition to those that moved on ridges at A, B, and 0 (fig. 131). No stones, however, moved on the high ridges at D, suggesting that other factors such as orientation of the ridges, position with re- spect to breaks, or local topography, lithology, and structure might also determine locations of magni- fied shaking. Stones displaced by shaking were easily identified. Differences in tone and color marked the seats from which cobbles and boulders had moved and identified formerly buried parts of stones. Furthermore, many stones moved from seats embedded as much as 100 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 mm into the surface, revealing original position and orientation exactly (figs. 133, 134). The observations of moved stones that form the basis for this report were made on two parallel ridges (locs. A, B, fig. 131). Only enough measure- ments were made to characterize broadly the move- ment of the stones; hence, the data lead simply to general statements about shaking and displacements and to estimates of the probable ranges of acceleration. RIDGE A Ridge A lies about 100 m from ridge B, and both trend about N. 45° E. Ridge A is covered by predom- inantly granitic gravel that ranges from small peb- bles to subrounded and rounded blocks with minimum dimensions greater than 0.5 In (fig. 132). The gravels examined occupy the crest and west and northwest slopes on surfaces inclined from 0° to about 20°. The earthquake either tipped, slid, or FIGURE 132.—Gravel on top of ridge A. Shaking displaced most of the stones in view in a direction away from the camera and slightly to the right. Detail of boulder a is in figure 133, and detail of boulder c in figure 134. THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 177 FIGURE 133—Stereo view of a 0.6-m-high tetrahedron tipped over by shaking. a, marks matching points on the boulder and its seat. Cobble (I moved more than 0.5 m across pebbles from its seat (b') into that vacated by the tetrahedron. Most stones in this picture have moved, includ- ing all that occupy the vacated seat of the tetrahedron. View is downhill; average slope is about 10°. flung virtually every unrestrained stone on ridge A in one direction, about S. 70° W. Although this direction includes a component of downslope move- ment for most stones on this ridge, many others that moved either rested on the flat ridge crest, had level bases, or were originally nested as much as 100 mm below the surface. Boulders and cobbles commonly moved an amount at least equal to their own diame- ters, and many pebbles and cobbles moved into the seats of other displaced clasts (figs. 133, 134). Mov- ing cobbles and boulders also struck and displaced numerous pebbles, and one elongate boulder rotated 180° when it struck a buried boulder (fig. 134). The evidence of large and apparently rapid displacements shown in figures 133 and 134 suggests large local accelerations, and possibly displacements, during shaking of the ground at ridge A. RIDGE B Surface gravel on ridge B is not so abundant as on ridge A and includes many tabular schistose cobbles, which were scarce on ridge A. Furthermore, all the displaced stones examined here occupy the nearly flat crest of the ridge. As on ridge A, dis— placed stones were easy to identify because of the strong color contrast between newly exposed seats and the surrounding surface. In addition, many of the large cobbles and small boulders that did move (and many smaller cobbles that did not) rested on low sand-silt pedestals that not only clearly defined preearthquake position of their covering stones but also offered no material barrier to sliding. In contrast to ridge A, the displaced stones of ridge B were neither flung nor tipped but slid uni- formly in one direction, about N. 65° E., nearly _..,,...,.,-.{,€3 *' * ”a; FIGURE 134,—Boulder 0, located on ridge A, was propelled from its nearly horizontal seat (pencil) and rotated 180" after striking the partly buried boulder d. Pebbles moved into the seat of boulder c after it moved out. The scale is 150 mm (6 in.) long. opposite to those of ridge A. No displacements ex- ceeded 0.1 m. A striking characteristic of the stones of ridge B was a well-defined size difference between clasts that moved and those that did not. Most stones less than about 50 mm high, which included many tabular schistose cobbles, did not move, regardless of shape. A 500-g laboratory spring scale was used to obtain crude measurements of coefficient of sliding friction and adhesion between stones and the surface of ridge B. Four determinations of coefficient of friction yielded values between 0.4 and 0.7. The adhesion of 13 quasi-rectangular, tabular cobbles lying on the flat surface of the ridge was measured by lifting one end only of each with the hook on the end of the spring scale. Each of the 13 cobbles required a force greater than half its weight to break it free from the surface when lifted in this manner. Half the weight of each cobble was determined from the scale reading after the surface adhesion was broken. Theo- retically the net force required to break adhesion, divided by one-half the contact area of each cobble, equals the adhesive strength between each cobble and the surface (table 29). The measurements in table 29 show that the adhesive force may reach a large fraction of the total weight of small cobbles. If the two extreme values of table 29 are excluded, the average of the remaining 11 values of adhesive strength is 270 N/m2 (roughly 0.04 psi). The tensile adhesive strength measured this way is assumed to be roughly the same as the shear adhesive strength; measurement of the shear adhesive strength of one flat cobble on ridge B gave a value of 400 N /m2, well within the range of values in table 29. THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 TABLE 29.—Adhesion of cobbles to the surface of ridge B Force used to Net force break cobble Weight of used to Dimensions Contact Adhesive free from one end of overcome of cobble area 0f strength3 surface1 cobble? adhesion (mm) “huge (N/m2) (g) (g) (g) ("m ’ 120 110 10 90X50 4,500 40 170 150 20 80x60 4,800 80 120 100 20 90X50 4,500 90 120 100 20 70x50 3.500 110 320 280 40 70x90 6,300 130 100 80 20 50x50 2,500 160 160 110 50 80x50 4,000 250 140 100 40 70x40 2,800 290 140 70 70 70X60 4,200 330 160 100 60 80x40 3 200 360 240 170 70 90x30 2,700 500 320 ........ 140 180 80x70 5,600 640 400 ............ 200 200 100x50 5 000 800 1Lifting one end only of each cobble with a 500-g spring scale (see text). For these measurements, 100 g is assumed to equal 1 N (newton). 2Measured after cobble broken free from surface. aRounded to the nearest 10 N/mz. On ridge A, most of the unrestrained material on the surface was either tipped, slid, or flung as much as 0.5 m either horizontally or partly downslope in a west-southwest direction. In contrast, on adjacent ridge B only boulders and cobbles taller than about 50 mm moved, and they did so by sliding as much as 100 mm in a direction almost opposite to that of the displaced gravel on ridge A. These differences were observed on ridges less than 100 m apart. The parallel directions of displace- ment on the two ridges (although opposite in sense) indicate that the direction of strongest horizontal shaking here was roughly N. 700 E. to S. 70° W., which is 70°—80° from the general trend of the surface rupture along the Coyote Creek fault and about 250 from the trend of the ridges. Whether the direction of most intense shaking and the sense of displacement are controlled primarily by the posi- tion, geometry, and composition of the ridges or by the regional orientation of seismic waves cannot be determined from these observations. Slope may have determined the sense of movement on ridge A, but differences in the position and shape of the two ridges or differences, if any, in the material under- lying them could also have been important. The available data, however, may help to reveal the magnitude of the accelerations on these ridges and to explain why the shorter clasts of ridge B did not move. MECHANICS OF DISPLACEMENT INDUCED BY SHAKING Before attempting to explain the displacements on ridges A and B, we should consider the different ways in which a stone may react to the shaking of an earthquake. For simplicity, we will consider the forces acting between a rigid rectangular block and the inclined surface upon which it rests THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 \ FF =/_L[(w-Fv)c050 —FHsun9] x». x\ FIGURE 135.—Forces, F, acting between a block of mass, m, den- sity, p, and basal area, A, and the inclined surface on which it rests, as a result of friction, adhesion, and shaking of the surface. a.- and am, vertical and horizontal accelerations of shaking; g, gravity; ,u., coefficient of friction; I1, height; 1, length; 9, inclination of surface; c, adhesive strength be- tween the surface and block (assumed here to be the same for shear and tension, to be independent of weight, and to embrace all surface forces, including nonideal components of friction); Fr and F", vertical and horizontal inertial reac~ tions to shaking; W, weight; FF, frictional résistance to slid- ing; Fa, adhesion. (shown in fig. 135). The block will slide during shak- ing when the downslope components of both weight and horizontal and vertical inertial reac— tions to shaking equal or exceed the combined frictional and adhesive resistance to sliding: WsinB—Fysin0+FHcos02FF+F,,. (1) (Symbols are defined in fig. 135.) Substituting the definitions of these forces given in figure 135 gives m(g—ay) sin0+ mancosaa p. [m(g—ay) coso —maHsin0] +CA , (2) which reduces to chos0+(g—ay)sin0> ,u [(g—a;~)cos6 . c —-aHsm6] +35. (3) The critical horizontal acceleration, am, necessary for sliding is then (g—ay) (,u—tan0)+c//1pcos0 (4) 1+,utan0 ' When vertical acceleration is upward rather than downward or when horizontal acceleration acts out from rather than into the slope, sliding will be more difficult. It is assumed for this simple analysis that tipping moments do not change over— all frictional resistance of the block. In the absence of adhesion (0:0), the accelera- tion that initiates sliding, am, depends only on g, ,1, chZ 179 6, and av and is in no way affected by shape or size of the block, unless the block tips or is rocking. (See succeeding paragraphs.) Thus rectangular blocks of any size or shape on a slope will start moving at the same time when the critical horizontal accelera- tion is reached. It is important to remember that like inertial forces, frictional resistance of a block to sliding is proportional to the weight of the block rather than to its contact area, as is sometimes implied (for example, Richter, 1968, p. 56). Vertical accelerations are not likely to change drastically the values of am. If maximum vertical accelerations range from about one-third to two-thirds as large as the values of associated horizontal accelerations (Housner, 1970, p. 79), am, decreases little with increasing av, particularly with lower values of a (fig. 136). Even the extreme example of avzaH reduces 0L"c by only 35 percent, if #205. However, an increase in the slope, 0, of the surface may greatly reduce the value of a”, (fig. 137). For “20.5, an increase in slope from 0°-10° lowers a”, more than does an increase of av from 0 to 2A; a” (fig. 136). It is also important to note that rocking probably allows sliding in some situations in which sliding would otherwise not occur. Variations in the period, LO ' 00 04 3. ‘ 0.8 - E 9 ,\|‘b°\e\ (7') N K E 0.6 - 12,30“ °\I 2 9 o z of H a: 5 m 04 ' U 0 <1 _I <1 ‘3 ’2 0.2 - n: U 0 J I l I I 0 0.2 0.4 0.6 08 lo COEFFICIENT 0F FRICTION FIGURE 136.— Relation between coefficient of friction, a, and minimum horizontal acceleration, am, necessary to slide a block that rests on a horizontal plane, for different values of associated upward vertical acceleration, up. From equation 4, with 9:0 and (=0. Adhesion between the block and the surface on which it rests will always raise the value of the horizontal acceleration required for sliding. 180 LC [ O9 Q,’ 65' 0.8 ' 0 \0 0.6 ' 0.4 - CRITICAL ACCELERATION FOR SLIDING (g) 0.2 - O . I . J I O 0.2 0.4 0.6 0.8 I.O COEFFICIENT OF FRICTION FIGURE 137.—Relation between coefficient of friction, u, and the minimum horizontal acceleration, am, necessary to slide a block that rests on an inclined plane, for different values of inclination, 0. From equation 4, with vertical acceleration of shaking and adhesion assumed to be zero. Inclusion of positive values for av will depress the right- hand part of each curve an amount proportional to and smaller than the depression of the corresponding curves in figure 136. magnitude, and direction of shaking are likely to yield ground movements that can slide a rocking block at vulnerable positions in its cycle of move- ment. If either a" or av exceeds 1 g, the block will prob- ably be propelled from its seat. Oldham, however, pointed out (1897, p. 133) that there must be an upper limit to the size of boulders that can be lifted by the vertical component of shaking, which is deter- mined by the strength of the surface on which they rest. Because the ratio of mass to contact area in- creases as size increases, boulders larger than some critical size will simply deform the surface beneath them as it shakes. Adhesion may drastically change the behavior of shorter blocks. If adhesion is present, an increases, and the size term, h, becomes important in equation 4. Thus a lower limit, h, for the height of rectangular blocks that will slide exists for every value of at,”c greater than the value in the absence of adhesion. This limit exists because adhesive resistance to slid- ing varies with contact area rather than with mass and is proportionately larger for shorter blocks than are inertial forces. Thus when rc>0, the ratio of THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 surface forces, Fa, to inertial forces increases for progressively shorter rectangular blocks until sur- face forces become dominant. As a”, increases in equation 4, the critical height, h, that separates blocks that move from those that do not, decreases. The block may also respond to shaking by tipping or rocking. For tipping or rocking to occur, the block must be tall enough that the overturning moment about an edge of the block exceeds the resisting moment (fig. 135), or g- [chos0+(W—Fy)sin0]+ éFHsino 2%(W—Fv)cos0+cA. (5) The critical horizontal acceleration for tipping, then, is > (g—aV) [(l/h)-tan0] +(l/h) (c/hpcos0). (6) ”6/ 1—(l/h)tan0 ' Generally, stones tip if they are partly buried in the surface or if they are somewhat taller than they are wide, so that the critical acceleration for tipping is less than that for sliding. In the absence of adhesion (c=0), initial tipping of stones is not controlled by size but only by shape (l/h), slope, and the accelerations due to shaking. That is, all objects of a certain shape on a given slope will tip at some critical acceleration, regardless of their size. Rocking of an object dur- ing shaking, however, is controlled by size. The larger the object of a given shape the longer is its natural period of rocking. Hence, those objects whose natural rocking period is close to the period of strong shaking may be overturned or translated, while others of similar shape but different size may not be so affected. This size-controlled response manifests itself only after tipping starts; it has no effect if the acceleration is too low to initiate tipping or rocking. If the block adheres to the surface, however, a larger horizontal acceleration is required to tip it, and size again becomes important. Depending on the magnitude of adhesion, all blocks of a given shape that are taller than some critical height, h, will tip at some level of acceleration greater than that neces- sary to tip the blocks in the absence of adhesion. Thus, if rocking occurs or if adhesion is present, a size effect, in addition to a shape effect (h/l), will control which blocks tip and which do not. Thus in the absence of adhesion, simple sliding of a rectangular block depends only on slope, coeffi- cient of friction, and acceleration and is independent of size and shape. Tipping depends only on slope, shape, and acceleration. Size of the block affects sliding or tipping only if adhesion or other surface a THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 forces are present. Once rocking begins, size also controls its period. MECHANICS OF SHAKING APPLIED TO OBSERVATIONS AT OCOTILLO BADLANDS The preceding analysis treats very simple situa- tions. The stones observed at Ocotillo Badlands, however, show great differences in shape and size; few are rectangular blocks, and some are embedded in the surface. Although coefficients of friction prob- ably are fairly uniform, adhesion shows ‘greater variation. Richter (1968, p. 140) pointed out that even arrays of columns of different sizes and shapes designed to determine earthquake motion commonly yield ambiguous information, presumably because of rocking and complex shaking. Nevertheless, the ob- servations of Ocotillo Badlands do permit estimates of the probable ranges of acceleration. The evidence of flinging and large displacements of stones of all sizes of ridge A, including many resting on slopes of less than 5°, almost certainly indicates either that horizontal accelerations ex- ceeded 0.5 g and perhaps approached 1 g or that Significant vertical accelerations accompanied some- what smaller horizontal accelerations, or both. How- ever, because many boulders that were partly buried in the unconsolidated surface material showed no evidence of vertical displacement, vertical accelera- tions were definitely less than 1 g. Slope may have strongly influenced the sense, if not direction, of displacement and probably contributed to some of the larger downslope displacements. Probably some of the boulders tipped over or slid only because they rested on slopes or because their bases sloped. How- ever, the fact that virtually every unrestrained stone on the surface moved, including many whose bases were originally horizontal, strongly suggests that large accelerations, rather than slope, were primar- ily responsible for movement. On ridge B, smaller displacements and the fact that small stones did not move indicate that accelera- tions were smaller and that size was important. Either shaking produced rocking on ridge B or adhesion prevented the smaller stones from moving. Rocking of stones on ridge B is almost certainly ruled out, because clasts that were displaced have the same shapes as clasts that remained stationary. Furthermore, on ridge B, a block about 0.25 m high by 0.25 m Wide by 0.4 m long slid horizontally 50 mm parallel to the 0.4 m length. The most reasonable combinations of horizontal and vertical accelerations sufi‘icient to rock this relatively low block parallel to its long dimension and thus to permit it to slide would have been large enough to move virtually 476-246 0 - 72 -13 181 every stone on this ridge. Because many did not move, including cobbles with more favorable ratios of height to length, it seems very unlikely that differ- ences in the natural period of rocking caused the size difference observed between those stones that were and were not displaced. Adhesion to the underlying surface very probably prevented the shorter stones on ridge B from sliding. This conclusion is supported both by the observed height rather than shape difference between stones that did and did not slide and by the values of adhe- sion measured on ridge B (table 29). Figure 138 shows that these values are compatible with reason- able levels of excess acceleration over that necessary in the absence of adhesion. For example, if “20.5 and (111:0 in the absence of adhesion, aH,:0.5 g for all stones on a flat surface (fig. 136). If adhesion, however, is 270 N/m2 (the average of the measured values), an excess acceleration of 0.2 g (2 m/sec2; see fig. 138) , or (14,0207 9, Will be required to move all stones taller than 50 mm (fig. 138). In other words, with this value of adhesion, only stones taller than 50 mm will be sliding when a” reaches 0.7 9. These numbers indicate that reasonable values of acceleration can account for the observations on 103 k ‘ ' . 00 ‘ ' ‘ ’oo ’0: :02 lllll J3 7:5cmm 0 HEIGHT (MM) l l l lllllll l [‘l lll l 1 no" r IO [0'2 EXCESS ACCELERATION lM/SECZ) FIGURE 138—Relation between height, h (separating blocks that slide on a horizontal surface from those that do not), and excess acceleration, ch—Idg—av) (excess over that necessary to slide blocks in the absence of adhesion), between blocks and the surface on which they rest for different values of adhesive strength, c, (N/mz). From equation 4, with 0:0 and p=2.7>< 103 kg/ma; other symbols as defined in figure 135. 182 ridge B. Moreover, figure 138 shows that reasonable local variations in a, c, 04,, and av do not drastically change the value of h that separates stones that will and will not move. A determination of the likely values of accelera- tion on ridge B requires an estimate of both the value of coefl‘icient of friction and the magnitude of the vertical component of acceleration, av, in addition to the value of adhesion. By using the range of mea- sured coefficients of friction, 0.4—0.7, and by assum- ing that maximum vertical accelerations were one-third to two-thirds the magnitude of horizontal accelerations, figure 136 yields values of a” ranging from about 0.3 to 0.7 g. If, in addition, we include the excess acceleration of about 0.2 9, indicated by the preceding consideration of adhesion, the final esti- mate of horizontal acceleration on ridge B lies be- tween 0.5 and 0.9 g. It is worth noting that if values of adhesion on ridge A were similar to those measured on ridge B, the fact that virtually all stones moved increases the likelihood that a" was close to 1 g on ridge A. In summary, the observations of displaced stones of ridges A and B indicate that on ridge A horizon— tal accelerations were probably greater than 0.5 g and perhaps as high as 1 9, whereas vertical accel- erations were definitely less than 1 9. Ridge B was subjected to somewhat lower horizontal accelera— tions, probably in the range of 0.5—0.9 g. Stones were displaced in the same direction, 70°~~80O to the trend THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 of the main surface rupture zone, although in oppo- site sense, on these two ridges. The study also demonstrates that higher parts of the ridges experienced greater shaking than the lower slopes of the same ridges, other ridges, low hills, and flat alluvium. Studies elsewhere indicate that surface shaking generally increases with greater depth of unconsolidated material (for example, Bor- cherdt, 1970; McCulloch and Bonilla, 1970, p. 69—72). Observations at Ocotillo Badlands, supported by more recent findings at San Fernando (Nason, 1971), indicate that 100 m of topographic relief in poorly consolidated sediments may also amplify shaking. REFERENCES CITED Borcherdt, R. D., 1970, Effects of local geology on ground motion near San Francisco Bay: Seismol. Soc. America Bull., v. 60, p. 29—61. Housner, G. W., 1970, Strong ground motion, chap. 4, in Wiegel, R. L., ed., Earthquake engineering: Prentice- Hall, p. 75—91. McCulloch, D. S., and Bonilla, M. G., 1970, Effects of the earthquake of March 27, 1964, on The Alaska Railroad: U.S. Geol. Survey Prof. Paper 545—D, p. D1—D161. Nason, R. D., 1971, Shattered earth at Wallaby Street, Syl- mar, in The San Fernando, California, earthquake of February 9, 1971: U.S. Geol. Survey Prof. Paper 733, p. 97—98. Oldham, R. D., 1897, Report on the great earthquake of 12th June, 1897: Geol. Survey India Mem., v. 29, 379 p. Richter, C. F., 1958, Elementary seismology: San Francisco, W. H. Freeman and Co., 768 p. WATER-RESOURCE EFFECTS By A. O. WAANANEN AND W. R. MOVIE, JR. U.S. GEOLOGICAL SURVEY ABSTRACT The principal effect of the Borrego Mountain earthquake on water resources was seismic in nature and almost entirely transient, as shown by the charts from recorders on ground- water observation wells in California, Arizona, and Nevada, and at gaging stations in California. Hydroseisms were recorded at several sites nearly 500 km from the epicenter. At most sites, the effects were limited to water-level or water-surface fluctuations, with some minor and insignificant net changes in water level at many wells; changes in stream- flow were observed at only two sites. Hydroseisms were observed also in many swimming pools near the epicenter, in San Diego County, and near Los Angeles. The effects on water quality were minor, and no significant problems were reported. INTRODUCTION Seismic shocks associated with the Borrego Moun- tain earthquake of April 9, 1968,1 caused noticeable fluctuations in ground-water levels, and in the stages of streams, canals, and lakes, over a wide area ex- tending nearly 500 km from the epicenter near Bor- rego Mountain, about 27 km southwest of‘the Salton Sea. Fluctuations were caused also by a few of the aftershocks that were associated with the Borrego Mountain earthquake (centered near the town of Ocotillo Wells). Effects from the M (magnitude) 6.4 principal shock occurring at 6 :29 pm. (P.s.t.) April 8 (02:28:59.1 G.m.t. April 9) were noted on the records from more than 120 water-stage recorders; the effects of two aftershocks (M 5.2 and M 4.7) were recorded at a'few stations. The effects of the earthquake on water resources can be classified in four main groups: (1) Water- level fluctuations in wells, (2) temporary changes in ground-water levels, (3) changes in spring flow or in the base flow of streams, and (4) variations in the magnitude of surface-water fluctuations along chan- nels caused by variations in channel geometry. Some of these effects occurred immediately after the three principal shocks, but others, such as changes in the 1Greenwich mean time (G.m.t.) . Local time was April 8, 1968. Pacific stan- dard time (P.s.t.) used for time reference in this discussion. flow of streams, developed slowly over a period of days and were sustained for several days or weeks. No permanent changes in water level or discharge are known. A noticeable but temporary increase in sediment content of the water was observed in one well as a result of the Borrego Mountain earthquake. Selected terms will be used in the subsequent dis- cussions, as follows: “Hydroseism” refers to all seismically induced water- level fluctuations other than tsunamis (tidal waves). The term, as defined by Vorhis (1967, p. CZ), pertains to fluctuations of water levels in wells, streams, lakes, ponds, and reservoirs. “Hydroseismic data” refers to both the charts that record hydroseisms and the information taken from the charts. “Seiche” is a term applied to standing waves set up on lakes and similar closed bodies of water by wind, change in barometric pressure, and other factors. Seiches caused by earthquakes are termed “seismic seiches.” A special form of hydroseism, they are characterized by the symmetrical fluctua- tions (equal water-level rises and declines) set- up on lakes, reservoirs, and ponds at times corre- sponding to the passage of seismic waves from an earthquake. FLUCTUATIONS IN GROUND-WATER LEVELS The principal transient effect of the April 8, 1968, earthquakes on the water resources was the abrupt and pronounced fluctuation of water level in response to the seismic shock, as shown by the records from many water-stage recorders. These hydroseisms commonly (are shown by analog-type recorders in areas affected by earthquakes. At times of large shocks, the recorder traces show vertical lines caused by movement of the recorder float as a result of wave action in the gage well or observation well induced by ground motion associated with the seismic shock. There is a rough relation of the amplitude of the fluctuations to the severity of the shock, the distance 183 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 / \ ( \Jonom s ..._\ . | . lb :32 5:90"; I I‘ "'aomwa‘ ' ,1 ' l, /HING on VI 'WA I 51c. me as" ,J ' m Avlwllo ‘-‘ a Pain! Cane-anon sum .4." 5-" mm: m... 46 h . sm- non ulna“ .- l 2 0 ° Sula luau mar-l n EXPLANATION ox Wover~level observation well (\ \RCIum-nu l ‘ Au Gaqmq station I Companion-recorder site Approximate epicenter lulu too Cummnb 0 I00 MILES r 4 “3° E A O 25 50 75 ll6° FIGURE 139. —— Selected observation wells, gaging stations, and compaction-recorder sites that showed effects of the Borrego Mountain earthquake. Observation wells and gaging stations are given in tables 30 and 31, respectively. from the epicenter, and the kind of underlying soil and rock formations. The records for several obser- vation wells also demonstrate some postearthquake short-term changes in water levels as a result of the shocks. The greatest areal extent of the hydroseismic effects was shown by water-level recorders on ground-water observation wells in California, Ari- zona, and Nevada, and by compaction recorders in California. These recorders were operated by the Geological Survey and by state, county, and local agencies. The location of the recorders that showed effects of the Borrego Mountain earthquake is shown in figure 139. Three sites in‘Fresno and Kings Coun— ties in California (sites 1, 2, and the compaction- recorder site near Fresno) and one site in Esmeralda County in Nevada (site 41 near the California border 84 km southwest of Tonopah) are nearly 500 km northwest of the epicenter. Hydroseismic data for the selected wells shown in figure 139 (sites 1—47) are given in table 30. The wells listed are identified by both a site number and a well number. The well numbers indicate the loca— tion of wells according to the rectangular system for the subdivision of public lands. The site numbers identify the township, range, section, location within the section, serial number of the well in the section, and the baseline and meridian in accord with the standard practices in California, Arizona, and Ne- vada. THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 TABLE 30.——Fluctuations of water level in observation wells, 185 April 8, 1.968 Wells in California Net Well Depth to rise . . Double Source No. . Depth Water-bearing water below Time of . (+) (fig. Slte No.1 County Area (In) material land surface shock” anzglglide or fall do: 4 139) (m) (—)s a 3 (mm) 1 ........ 15S/16E—20R1 M ........ Fresno San Joaquin 109 Alluvium... 24.19 6: 29 p.m 61 —6 GS 2.. 188/19E—20P1 M King": lmmnnre 212 62.18 6 : 29 58 ~55 GS 3.. 23S/25E—16N4 M Tulare Pixley 76.2 24.84 6:29 18 —1 GS 4.. 24S/26E—34F1 M do Rinlnwrove 461 65.85 6:29 27 ~12 GS 5 ........ 7N/35W~28R1 S ......... Santa Barbara ....... Lompoc ........................... 156 Alluvium and 19.10 6: 29 24 —1 GS older deposits. 6 ........ 328/28E—20Q1 M Kern Rakersfleld 290 Alluvium ......................... 58.56 6:29 92 —91 GS 7 ..... 10N/9W—4D1 S .............. do ........................ Rogers Lake. 152 Lakeshore deposits 32.64 6 : 29 61 +9 GS 8.. .. ..San Bernardino ..... Daggett .......... 51.9 Alluvium ....... 29.93 6: 29 37 -— GS 9 ........ 3S/15W—13P1 S l’ n: Ang‘ El Segundo 101 Sand dunes... 1.1% $2 32 49 +3 LA . z 3 10 ........ 35/15W—24G S dn Manhattan Bent-ll 122 Alluvium ......................... 2.94 6:29 85 +6 LA 11 ........ 3S/14W—31D1 S do Hermosa Beach 101 Alluvium; uncon- 2.73 6 2 29 530 +30 LA solidated. 2.70 7 : 04 30 12 ........ 4S/14W—6J7 S do Rednndo Beach 156 Sand dunes .................... .53 6 : 29 140 +12 LA .5 7: 04 6 0 13 ........ 4S/14W—17N2 S do do 209 ...... d0 ............................... .14 6:29 125 +12 LA .13 7 : 04 3 —3 14 ........ 4S/12W—35K6 S ................ do ........................ Long Beach ................... 114 Alluvium.... 1.66 6:29 110 —3 LA .. .5S/12W—2G2 S do do 32.6 ...... do.. 2.49 6 : 29 340 0 LA ..4S/12Wa36P6 S .......... Orange ..................... Los Alamitos ................. 7 149 ...... do 2.51 6: 29 131 —9 LA ..55/12W—1IG4 S l as Annals“: l ong Bench 25.3 ...... do. .43 6: 29 230 0 LA ZS/l2W—1oqz s do Mnntebello 168 Older deposits. 27.48 6: 29 123 -—58 LA .. .. 2S/11W-5L1 S dn dn 30.8 Alluvium... 3.98 6:29 480 0 LA 20 ........ 28/12W—14P1 S do ...... do 29.9 ....do ........... 7.65 g: (2): 20(2) —§ LA 2S/12W—23N2 S do ...... do 54.9 6: 29 390 —27 LA ..5S/10W—32K3 S .......... Orange ..................... Santa Ana. ...................... 274 6: 29 119 (5) GS .5S/10W-3201 S do do 56.1 6: 29 49 0 GS ..1N/8W—26P1 S Lns Angelo Clarnmnnt . 153 6: 29 24 0 LA ..ZS/3W—2Nl S ............. San Bernardino ..... Redlands ......................... 83.3 6:29 43 —15 GS ..25/3W—11M2 S dn dn 65.9 6 : 29 113 —6 GS ..4S/3W—18Gl S Riverside Ferris 115 6: 29 24 0 DWR ..5S/1E—6P1 S ..................... do ........................ San Jacinto ................... (6) 6: 29 73 +43 RC .. 5S/1E—9J2 S do do 56.7 6:29 244 —12 RC 28/3E—12L1 S ................... do ........................ White Water ................. 52.6 3:32 183 —9 GS : 9 0 4S/5E—17L1 S _ do Palm Spring: (0) 6:29 24 —3 DWR 10S/6E—21A1 S ........... San Diego ................ Borrego Springs ........... 97.9 3:32 633 +670 GS . : 0 . 10:31 a.m. (Apr. 9) 26 —3 14S/17E—13 S ..... Imperial El Centro, 24 ENE ...... (0) do . 6: 29 p. m. —3 IID 34.. ..165/22E—29Gca2 .do... .Yuma, Ariz .................... 542 Alluvium, Bouse 4.80 6: 29 98 0 GS formation, Tertiary 4.78 7: 04 6 0 Fanglomerate 35 ........ 165/23E—10Rcc S ...... ....do.. ...... do.... 167 Alluvium.... 4.68 6:29 171 0 GS 4.68 7:04 6 0 Wells in Arizona 36 ........ (0—9—25)35bab G ....... Yuma ....................... Yuma .............................. 343 Alluvium ......................... 5.54 6:29 p.m 55 —-6 GS 5.54 7: 04 3 0 37 ........ (C—9-22)4dbc2 G do do 73.2 ...... do.... 23.08 6: 29 119 (5) GS 38 ........ (C—11—24)23bcb G do do 288 ....do... 3.20 6: 29 250 —6 GS 3.17 7 : 04 119 (6) 39 -------- (B—2—1)2aba G Mnricopa (‘lendnle 112 ...... do.... 82.11 6:29 6 —3 GS 40 ........ (A—2—4)23bba G ........ dn Scottsdale 57. 3 ...... 110.. 41.97 6 : 29 6 —3 GS Wells in Nevada 41 ........ 4S/36E—5bccl M F mernlda Dyer 59.8 Alluvium and 4.71 6 29 p.m 30 0 GS Paleozoic clastic rock. 42 6S/61E-32cd1 M Lim-oln Alamo 20.7 Alluvium. 1.98 6:29 3 0 GS 43 ...,Indian Springs... 424 224.75 6: 29 6 3 GS 44 I nfhrop Wall < 42.1 23.23 6 z 29 34 —6 GS 45 (7) .57 6: 29 230 (°) GS 46 18/54E—10aac1 M.. ..Clark ........................ Pahrump ........................ 244 27.21 6: 29 49 —12 NSE 198/60E—9bcc1 M do I as Venn: 253 37.20 6 : 29 152 +70 NSE 1Location referred to baseline and meridian as follows: M, Mount Diablo; S, San Bernardino; G. Gila. and Salt River. ”Time of shock in Pacific standard time (P.s.t.); add 8 hours for Green— wich mean time (G.m.t.). 3Approximate water-level ofi'set before and after shock. 4Agencies as follows: GS, U.S. Geological Survey; LA, Los Angeles County Flood Control District; DWR, California Department of Water Resources; RC, Riverside County Flood Control District; IID, Imperial Irrigation District; NSE. Nevada State Engineer. A water-level rise of 0.67 m in well 10S/6E—21A1 S near Borrego Springs (site 32) was the greatest change observed; most water-level changes did not exceed 0.09 m. The fluctuations reflected principally the response to the main Borrego Mountain earth- quake shock at 6 :29 p.m. (P.s.t.) April 8. The records for several observation wells indicated move- ment in response to the M 5.2 shock that occurred about 35 minutes after the principal shock. The well 476-246 0 - 72 - 14 5Not recorded; pen ofi' chart. 6Not known. 7Recorder in Devil’s Hole, a spring in cavernous limestone: exploration by scuba divers indicates depths greater than 60 m and diameter more than 2 m. near Borrego Springs (site 32) responded also to the M 4.7 shock at 10:31 am. (P.s.t.) April 9. Examples of direct responses to the seismic shocks are illustrated in figures 140 and 141. The hydro- seisms shown are representative of the more pro- nounced effects observed, including the range in the fluctuations and the changes in water levels. Most of the records available, however, indicate only short- term fluctuations at the time of the principal shock, 186 THE BORREGO MOUNTAIN 50.0 I l T Well at sile 32, near Borrego Springs, San Diego County, Calif. 50.2 - —4 50.4 i— 50.6 DEPTH BELOW LAND-SURFACE DATUM, IN METERS 50.8 I I 20.3 I I Well at site 30, along Mission Creek near White Water, Riverside County, 20.4- Calif. - 20.5— \ 20.6 ' I 3|.9 1 I l Well at site 33, 39 km ENE of El Centro, Imperial County, 32.0 ' s L.—\1F—~ Calif. - §——I\ 9 APRIL Isea FIGURE 140.—Fluctuations recorded at observation wells in the Salton Sea basin during and after the Borrego Mountain earthquake. Time is Pacific standard time. and a few demonstrate the response to one or more aftershocks. Parts of the recorder charts for three observation wells in the Salton Sea basin (sites 32, 30, and 33) are shown in figure 140, and those for three obser— vation wells in central and southern California outside of the Salton Sea basin (sites 21, 11, and 6) are shown in figure 141. The well at site 32 near Borrego Springs, 23 km northwest of the epicenter of the principal shock, was the nearest well observed. For several hours prior to the earthquake, the water level in the well showed a small decline and a slight rise. Immediately after the main shock, the water level started to rise rapidly; it rose 0.67 m in 15 min- EARTHQUAKE OF APRIL 9, 1968 I45 Well at site 2|, Monlebello, Los Angeles Counly, Calif m I E: 4.8— A a I... m 2 — _ Z l5.| ' 1 2.4 l l l" ‘1 SI Well at site II, Hermosa DEPTH BELOW LAND-SURFACE DATUM, l — Beach, Los Angeles _ County, Calif. 3.0 ' 58.4 I I Well at site 6, 32 km south _ of Bakersfield, Kern a County, Calif./ \' 58.7 ' ' 8 9 APRIL I968 FIGURE 141,—- Fluctuations recorded at observation wells in central and southern California during and after the Bor- rego Mountain earthquake. Time is Pacific standard time. THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 utes and then declined in 18 hours to a stable level about 0.09 In higher than that before the shock. The rapid water-level rise of 0.67 m may have been caused by compaction or compression of the aquifer, which resulted in the upward movement of water. As the aquifer pressures returned to equilibrium, the water level declined very gradually for several days after April 9, and the 0.09-m residual change subsequently disappeared. The record for this well also showed a minor response to the seismic shock at 10:31 a.m. (P.s.t.) April 9 (fig. 140). The record for the Mission Creek well (site 30, fig. 140) indicates multiple water-level fluctuations at the time of the principal shock at 6:29 pm. (P.s.t.) April 8 and for a short period after this shock. This multiple response possibly is the result of the aftershocks at 6:33 and 6:37 pm. April 8. This well also responded to the M 5.2 shock at 7 :04 pm. Compaction recorders in Tulare and Kern Coun- ties 50—60 km north of Bakersfield and one in Fresno County 16 km northeast of Fresno (fig. 139) showed movement resulting from ground motion at the time of the principal seismic shock on April 8. Compaction recorders, as described by Lofgren and Klausing (1969, p. B69), are special installations of recorders connected to heavy weights emplaced in the under- lying formations. These instruments record changes in thickness of particular sequences in water-bearing deposits as the water table declines. The amplitude of the fluctuations recorded on April 8 was small, but the time of the movement indicated that they re- sulted from the earthquake. The magnitude of the hydroseisms, as shown in table 30, demonstrates local differences in the re- sponses of water levels to the effects of seismic shocks. These differences reflect variations in well construction, distances from the epicenter, and geo- logical influences. Deep wells that have adequate perforations, for example, may have fluctuations of greater magnitude than shallow wells. The 274-m deep well at site 22 in Orange County 8 km southwest of Santa Ana had a fluctuation of 119 mm, while an adjacent well (site 23) , 56 m deep, had a fluctuation of 49 mm. The water level in this area was less than 3 m below the land surface. Wells in Los Angeles County in the Montebello area, along the coast near Santa Monica Bay southwest of Los Angeles, and in the Long Beach area south of Los Angeles had fluc- tuations in water level with amplitudes as great as 530 mm (site 11, at Hermosa Beach) ; the net changes, however, were small, less than 60 mm. In an earlier study of the effects of earthquakes on water levels, La Rocque (1941) reported that during 187 the Long Beach, Calif., earthquake of March 10, 1933, the water—level fluctuation in one well (4S/ 12W—28H2 S) reached an amplitude of 3.23 m with a net rise of 2.51 m. This well is located in Long Beach just east of Signal Hill; the epicenter of the earthquake was only 27 km southeast of Signal Hill. La Rocque reported also that lines of equal surge, sketched on the basis of recorded data, indicated the occurrence of surges greater than 3 m along a north- west-southeast trending band about 8 km long just east of Signal Hill. This area is closed to the so-called Inglewood-Newport fault zone. The underlying un- consolidated deposits contain thick impervious beds, and water in the deepest aquifers is confined under a hydrostatic head as great as 460 m. The pattern of the lines of water-level surge suggested, as noted by La Rocque, that the magnitude of the surge may have been caused to a considerable degree by the thickness, perviousness, and elasticity of the water- bearing zones and by the effectiveness of the confin- ing beds. SURFACE-WATER FLUCTUATIONS AND TURBIDITY Seismic effects were shown by water-stage record- ers at many gaging stations on streams in the Salton Sea basin and in basins in central and southern Cali- fornia. The effects, which were generated largely by wave motion induced in the stilling wells by ground motion associated with the seismic shock, caused movement of the recorder floats and consequent ver- tical tracings on the recorder graphs. This recorded rise and fall of the water surface constitutes an approximate measure of the severity of the earth shock. Two distinct events were recorded at some gaging stations on April 8, one corresponding to the prin- cipal shock at 6:29 p.m., and the second about 35 minutes later. The second event was also recorded in many ground-water observation wells. The location of selected gaging stations that showed the shock effects is shown in figure 139 (sites 48—84). Pertinent data on the range of the fluctua- tions and the net change in stage are given in table 31. These data demonstrate the area affected by the earthquake. Most of the recorder traces showed only a direct response to the shock or aftershocks, with no sustained changes in stage or flow. Parts of the recorder charts from five selected stations that show the fluctuations in response to the principal shock and the secondary event are repro- duced in figure 142. The record for Salt Creek near Mecca (site 69) consists of water-surface oscillations that continued for several minutes; this response is 188 TABLE 31.—Fluctuations and'net changes in water level recorded at gaging stations m Californza, Aprrl 8, 1.968 Fluctuations in Net gage well (double Changes, No. Stream, reservoir, or amplitude, in mm) rise (+) (fig. canal, and station ————-— or 139) 6:29 7:04 fall (—) p.m. p.m. (mm) 48 ........ Amargosa River at Tecopa ...... 3 ........ 0 49 ........ Salinas Reservoir (Margarita Lake) near Pozo .................... 121 ........ 0 50 ........ Santa Ynez River above Gibraltar Dam, near Santa Barbara ........................ 12 ........ 0 51 ........ Santa Ynez River near Santa Ynez .............................. 6 ........ 0 52 ........ Atascadero Creek near Goleta ........................................ 6 ........ 0 53 ........ Big Rock Creek near Valyermo .................................. 12 ........ 0 54 ........ West Fork Mojave River near Hesperia ......................... 6 ........ 0 55 ........ Lone Pine Creek near Keenbrook ................................ 30 ........ 0 56 ........ Lytle Creek at Colton ............... 70 ........ —3 57 ........ Plunge Creek near East Highlands ................................. 37 ........ 0 58 ........ Santa Ana River at spreading grounds near Mentone .......... 58 ........ 0 59 ........ Mill Creek power canal No. 1, near Yucaipa2 ......................... 18 ........ 0 60 ........ Mission Creek near Desert Hot Springs ............................. 15 ........ -—3 61 ........ Palm Canyon Creek near Palm Springs .......................... 18 ........ 0 62 ........ Temecula Creek at Vail Dam. 27 ........ 0 63 ........ San Luis Rey River at Oceanside ................................. 6 ........ 0 64 ........ Sweetwater River near Descanso .................................. 6 ........ 0 65 ........ Jamul Creek near Jamul ........ 30 ........ 0 66 ........ Cottonwood Creek above Tecate Creek, near Delzura. 21 ........ 0 67 ........ Borrego Palm Creek near Borrego Springs ..................... 6 ........ (3) 68 ........ Whitewater River near Mecca ........................................ 214 ........ 4—82 ........ 9 0 69 ........ Salt Creek near Mecca .............. 70 ........ +3 70 ........ Salton Sea near Westmorland ........................... 250 ........ —9 ........ 21 0 71 ........ New River at Mexican Boundary5 ................................ 95 ....... 0 72 ........ West Side Main canal at ........ 12 0 Dixie Pond5 ............................. 150 ........ +76 73 ........ Dixie Pond automatic spill gate-’5 .......................................... 6610 ........ +6 representative of gaging stations in the Salton Sea basin. Change in streamflow, a significant and common response of the water regime to earthquakes, is indi- cated by the records for Borrego Palm Creek near Borrego Springs (site 67) and Whitewater River near Mecca (site 68). Prior to the earthquake, the flow in Borrego Palm Creek, after a rise April 2 to 0.04 m3/sec (cubic meters per second) in response to 21 mm of precipitation, had declined by April 8 to 0.006 m3/sec. The shock at 6:29 p.m. April 8 caused only minor wave motion in the gage well. About 8 hours later, however, without any rain, the flow at the gaging station increased suddenly and was 0.03 m3/sec 16 hours after the shock. This THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 TABLE til—Fluctuations and net changes in ‘water level recorded at gaging stations in California, April 8, 1968— Continued Fluctuations in Net gage well (double changes, No. Stream, reservoir, or amplitude, m mm) rise (+) (fig. canal, and station —- 01' 139) 6:29 7:04 fall (__) p.m. p.m. (mm) 74 ........ West Side Main canal at automatic gate of No. 8 pond5 ......................................... 6300 ........ +82 ........ 6300 0 75 ........ West Side Main canal at Trifolium 11 lateral5 ............. 70 ........ 0 76 ........ New River at outlet5 .......... 183 ........ —24 77 ........ Vail cutofi" drain5 ................. 158 ........ 0 ........ 27 0 78 ........ Rositas Pond at Rose heading5 .................................... 265 ........ 0 ........ 21 0 79 ........ Alamo River at drop No. 95 ..... 37 ........ +6 80 ........ Alamo River at drop No. 25.... 280‘ ........ —3 81 ........ Alamo River at outlet5 ............. 43 ........ —21 82 ........ Lateral D at gate 305 ................ 195 ........ +3 83 ........ East highline canal at Z pond5 ..................................... 92 ........ +30 84 ........ T drain5 ....................................... 150 ........ —3 ........ 0 1Seiche; lasted 2 hours. '-’Data from Southern California Edison Co. 3Delay in response: rise in stage and increase in discharge at gage started 8 hours after principal shock. 4Flow decreased 0.57 mil/sec. 5Data from Imperial Irrigation District. “Complete rotation of recorder drum; exact definition uncertain. increase may have been the result of increased spring discharge at some point upstream from the gaging station. The increase in flow seems to have been sustained for more than a month. Flow at the gaging station ceased seasonally June 16. Substantial water-surface fluctuations were re- corded at the gaging station on Whitewater River near Mecca (site 68) at the time of the principal shock, and the stage declined suddenly (fig. 142) ; a secondary fluctuation occurred about 35 minutes later. The flow declined from 4.0 m3/ sec before the principal shock to 3.4 m3/sec after the event, and it continued at a reduced level for more than a week. Subsequently, the flow increased gradually to the preearthquake magnitude. The decline may be attrib— utable to short-term changes in the storage and drainage capabilities of the irrigated lands adjacent to the river. The record for the Salinas Reservoir (Margarita Lake) near Pozo, Calif. (site 49) shows a seismic seiche that had an initial range of 21 mm and con- tinued for about 2 hours. The reservoir is 470 km northwest from the epicenter. The Alaskan earth- quake of March 27, 1964, caused a seiche in this reservoir that had an initial range of 128 mm and continued for nearly 6 hours. In contrast, the Park— field-Cholame earthquake of June 27, 1966, caused only a short-term hydroseism with an amplitude of 55 mm (Waananen and Page, 1967), although the distance from the epicenter was less than 80 km. The Salton Sea, however, as shown by the record THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 0.6 Borrego Palm Creek near Bond:go Springs, Calif (site 67) .— I - ____,_———\ I - W 0.50 US , . . . Whitewater RIver near Medea, Calif. (snte 68) m _ n: u ___,_____,~_..~ '_ _ Lu I s - l _ Z | ‘ 0.80 ' '3 ll5 , (:5 salt Creek near Mecca,. Calif. (site 69) _ _ I _ E | In _ l I _ 0 | < I ‘ (D _ l _ l 0.95 i -7o.5 , m Sallon Sea near Westmolrland, Calif. (site 70) z 0! I < m WI Lu l- I 2 w * I ' u = l S E l a: --70. L <1: _, 9 z M39495 I E: Salinas Reservoir near Plozo, Calif. (site 49) I _ g ‘1 “V ‘l I.|J ‘1 I _I In - I _ m w i 394.80 ' 8 9 l0 APRIL ISGB FIGURE 142.—F1uctuations recorded at gaging stations in response to the Borrego Mountain earthquake. Time is Pacific standard time. near Westmorland (site 70, fig. 142), evidently did not develop any seiche, although the water-surface fluctuations in the gage well had an amplitude of 250 mm and the oscillations continued for several minutes; one aftershock was also recorded. Many water-stage recorders on streams, canals, ponds, and drains operated by the Imperial Irriga- tion District south of the Salton Sea showed large hydroseisms at the time of the principal shock and the secondary event April 8. Data for 14 of the stations (sites 71—84, fig. 139) are given in table 31. The amplitude of the fluctuations exceeded 150 mm at many sites and varied widely between adjacent sites, owing to differences in the width and depth of channels and the effects of canal junctions, automatic spill gates at ponds, and other manmade features. The fluctuations resulting from the principal shock continued for several minutes. The records for the stations at the automatic gates show extreme fluctu- ations that may have resulted from secondary wave 189 motion caused by operation of the gates in response to water-surface oscillation in the ponds. The avail— able records did not indicate any significant changes in flows that could be attributed to the shocks. Seiches, especially in small pools and ponds, may have been the most common surface-water effect attributable to the Borrego Mountain earthquakes. Countless swimming pools in San Diego County and over widespread adjacent areas in southern Cali- fornia were subjected to seiches and sloshing water. The effects ranged from minor oscillations to severe spill. The Ironwood Motel at Ocotillo Wells, about 6 km south of the epicenter, was flooded by water sloshed out of an adjacent pool (Cloud and Scott, 1968a, p. 1189; 1968b, p. 20). At Cypress, in Orange County, 13 km west of Anaheim and 32 km southeast of the Los Angeles Civic Center, the hydroseism in a pool was sufficient to cause a splash of more than 150 mm spill from the pool. A lesser motion, without spill, was observed in a motel pool near the commu- nity of Borrego Springs in Borrego Valley, about 40 km west of Salton Sea and less than 22 km from the epicenter. The full extent of this effect has not been determined. The earthquake may have caused the roiling or muddying of water in some wells and springs, a common effect with respect to water quality. A noticeable temporary increase in well-water turbidity was observed by the owner of a well along State Highway 67, about 8 km west of Ramona, San Diego County, and about 40 km northeast of San Diego. The effects of the Borrego Mountain earth- quake in most wells may have been so slight or of such short duration that they either were not visible or were not observed. REFERENCES CITED Cloud, W. K., and Scott, N. H., 1968a, The Borrego Moun- tain, California, earthquake of 9 April 1968—a pre- liminary engineering seismology report: Seismol. Soc. America Bull., v. 58, no. 3, p. 1187—1191. 1968b, Field reports and isoseismal map, in Strong- motion instrumental data on the Borrego Mountain earth- quake of 9 April 1968: Seismological Field Survey, US. Coast and Geodetic Survey and Earthquake Engineering Lab., California Inst. Technology, p. 19—119. La Rocque, G. A., Jr., 1941, Fluctuations of water level in wells in the Los Angeles basin, California during five strong earthquakes 1933-40: Am. Geophys. Union Trans, pt. II, p. 374—386. Lofgren, B. E., and Klausing, R. L., 1969, Land subsidence due to ground-water withdrawal, Tulare-Wasco area, California: US. Geol. Survey Prof. Paper 437—B, 101 p. Vorhis, R. C., 1967, Hydrologic effects of the earthquakes of March 27, 1964, outside Alaska: US. Geol. Survey Prof. Paper 544—0, 54 p. Waananen, A. 0., and Page, R. W., 1967, Water—resources aspects, in Brown, R. D., and others, The Parkfield- Cholame, California, earthquakes of June—August 1966: US. Geol. Survey Prof. Paper 579, p. 53-56. COLLAPSE FISSURES ALONG THE COYOTE CREEK FAULT By MALCOLM M. CLARK U.S. GEOLOGICAL SURVEY ABSTRACT Gaping collapse fissures that are locally as much as 4 m deep, 3 m wide, and 100 m long have developed along the Coyote Creek fault on many tectonic fractures associated with the Borrego Mountain earthquake of April 9, 1968. Most of these collapse fissures developed only in zones of active creep during rainfall sufficiently intense to cause surface runoff. Surface water, pouring into new tectonic fractures or into those kept open by creep, tunnels hori- zontally and cascades downward at many levels Within the fractures, eroding sediment from the walls and from found- ered blocks and transporting it deeper into the fractures. This removal of material leads to foundering and slumping of ad- ditional blocks from the walls of the fractures until flow of water ceases or until the fissures become clogged or filled at depth. Many fissures that remain open after flow of water ceases become clogged by slumping or are filled with windborne debris within weeks or months, unless additional tectonic creep keeps them open. Renewed creep may also reopen fissures that have been clogged for several months or years, thus permitting additional collapse. After a fissure becomes clogged, it slowly fills with windborne and water- borne material, a process that may take years in some loca- tions. Concentration of surface runofl" in fissures leads to vigorous growth of vegetation along them that may persist long after the fissures fill. Because collapse depends on fortuitous occurrence of runoff soon after tectonic movement, it may be a relatively unusual fate of tectonic fractures in this arid area. Assuming that tectonic fractures in the weak sediments broken by the 1968 rupture remain open to a depth at least as great as that of the water table (about 5—30 In), in most tectonic fractures the open space above that level is large enough to accommodate the material removed from the col- lapsed parts at the surface. Four groups of collapse fissures at the south end of the 1968 surface rupture predate the 1968 earthquake. The two oldest groups probably opened 15 years or more before the earthquake. Another group became active after 1953 to 1956 but apparently ceased enlarging several years before 1968. The latest collapse on the youngest group occurred less than a year and perhaps merely a few months before the earth- quake. Collapse fissures have developed elsewhere, usually in arid areas. In a few places they start from tectonic fracturing, but generally they start from fractures caused by other agents such as desiccation, soil piping, and subsidence. Never- theless, the development of collapse along fractures of any origin, which evidently requires conditions that are met in 190 arid regions, is probably similar to the development of col- lapse observed along the tectonic fractures of the 1968 earth- quake. INTRODUCTION By far the most prominent surface features to develop in the Coyote Creek fault zone after the Borrego Mountain earthquake of April 9, 1968 have been large gaping fissures along tectonic fractures (figs. 143—150; see also fig. 42). Many of these fis~ sures are as much as 4 m deep and 3 m wide, and some are hundreds of meters long. They form when runoff from infrequent heavy rain pours into open (typically 20—100 mm Wide) tectonic fractures and carries slumped and foundered material deep into the fractures; removing this material causes additional slumping and collapse of the walls of the fractures. The development of these collapse fissures, as they are called in this report, is one of the most obvious signs of active or very recent creep along the 1968 rupture. Collapse has also revealed the presence of many otherwise inconspicuous tectonic fractures that were overlooked during early phases of the postearthquake investigation because they were not along the main rupture. Because collapse increases the time required for sediment to fill tectonic frac- tures, it extends the period during which these frac— tures are evident. Collapse fissures collect surface runoff, encouraging unusually abundant and vigorous vegetation to grow along them and to mark the location of some original fractures for years or decades. Fissures constitute a hazard to the increas— ing numbers of high—speed dune buggies, jeeps, and motorcycles that frequent the flat terrain of this generally undeveloped region. Development of col— lapse fissures might also endanger structures that happen to be built upon tectonic fractures. One of the most important characteristics of col— lapse fissures is that their extent and time of devel- opment correspond closely to those of creep. Because slumping or blown sand and silt filled or closed fractures rapidly along the 1968 rupture, the only fractures open at the surface during periods of heavy THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 191 rain, and hence susceptible to collapse, were those that were either actively creeping or had been creep- ing no more than a few months before the rain. (Heavy rain falls in the area of the 1968 rupture only a few times each year; see fig. 40.) The first major collapse fissures to form along the 1968 rup- ture after the earthquake were caused by heavy rains in July 1968, the first to fall after the earth- quake. Collapse fissures developed then only along fractures, all part of the central break, that were actively creeping during or before the rains. Subse- FIGURE 143. —— Collapse fissure on the south break (loc. 30.3) that developed on a tectonic fracture after the rains of September 1969. A, October 1969. This recently collapsed fissure shows steep to vertical wall and peripheral cracks that bound the blocks that will collapse next. Small desic- cation cracks break the mud locally in the bottom of the fissure, but collapse of the walls continued along most of the fissure long after any ponded water had dried up after the September runoff. B, January 1970. Water inflow during rains early in November 1969 widened the fissure by additional collapse, but material accumulated faster than the water conducted it down into the fissure, thus causing the fissure to fill. Little collapse occurred after water ceased flowing. A grassy band is evident along the sides of the fracture and along the channel that conducted water into it from the right. C, December 1970. Runoff and wind during 1970 have added a small amount of ma- terial to the fissure, but the most obvious change is in vegetation. The grass grew taller than 0.1 m and died, and many shrubs have appeared in the grassy band. Creosote bushes along the fissure (to the right of the car) have increased markedly in size and vigor, whereas those distant from the fissure are little changed. quent collapse on the central break occurred only in areas where creep opened fractures immediately before or during episodes of runoff. No collapse fis- sures developed either in July 1968 or later on the north break, because blown silt and sand rapidly filled fractures created at the time of the earthquake, and no creep occurred to reopen them. No collapse fissures developed in July 1968 along the south break for the same reason, and none formed until nearly 11/2 years after the earthquake because of a delay in the onset of creep along that break. Thereafter 192 FIGURE 144. — Aerial view southwestward of collapse fissures and channels leading into them on south break (arrows) at location 31.4, January 1970. Vehicle tracks give scale. Branching fracture in lower half of View formed after January 1969 but before September 1969, and it continued to open and extend after September. Collapse shown here occurred in September and November 1969. The largest parts of the fissure are a and b, located where wide shal- low channels cross the fracture. Collapse developed on either side of the channels along the fracture, indicating subsurface flow of intercepted surface water in the fracture itself. Figures 145, 146, and 150 show details at a, b, and 0, respectively. The large plants are mesquite (Prosozn's); most of the small ones are creosote bush (Larrea) . collapse occurred on the south break only in areas of continued or renewed creep. This paper describes the collapse fissures that formed along the 1968 rupture and offers some explanations of how they form. They represent a potential, if not common, feature of tectonic frac- tures in this region, so they can be expected to form after future surface faulting here and perhaps else- where in desert regions. An understanding of these fissures, especially their connection with creep, can be important in future investigations of surface faulting in desert regions. Indeed, the presence of new collapse fissures in a fault zone constitutes pre- sumptive evidence of new fracturing. Furthermore, if the new fractures are caused by creep, subsequent collapse may be the first obvious sign of the presence of fractures and hence of the presence of creep. This report draws heavily from information in a preceding paper in this volume (Clark, “Surface Rupture Along the Coyote Creek Fault”), which describes the fractures that formed along the Coyote Creek fault during and after the earthquake. The observations supporting the explanation of collapse fissures reported here were made during periodic visits to monitor creep along the 1968 rupture. As will be explained below, development of collapse fissures along tectonic fractures is closely related to THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 creep, hence the reader should look at the sections of the preceding paper that deal with the postearth- quake investigation and with creep. This paper will use the names north, central, and south breaks for the three main sections of the 1968 rupture, and locations will be given in the coordinate system of plate 1. ACKNOWLEDGMENTS Because the investigation of collapse fissures grew out of the study of the surface rupture of the 1968 earthquake along the Coyote Creek fault, most of the people acknowledged in that report (see Clark, “Surface Rupture Along the Coyote Creek Fault,” this volume) also helped in this investigation. In addition, I have had helpful discussions with Arthur Grantz and E. A. Jenne of the U.S. Geological Sur- vey. M. G. Bonilla, Arthur Grantz, D. H. Radbruch, and R. V. Sharp made many suggestions to improve the manuscript. OBSERVATIONS OF THE DEVELOPMENT OF COLLAPSE FISSURES ALONG THE 1968 RUPTURE The most complete record of the development of collapse fissures comes from the south break, which not only has collapse fissures that predate the earth- quake, but also afforded the best observation of the evolution of new collapse fissures because of a for- tunate timing of creep, rainfall, and field visits (fig. 40). Although less complete, observations of collapse fissures along the central break support the findings from the south break. EFFECTS OF THE RAINS OF SEPTEMBER 1969 I first saw the collapse fissures of the south break in October 1969, about one month after the runoff that had created most of them. (See figs. 143A, 144, 145A, and 146A.) Most of the prominent fissures had developed on new and reopened long single fractures rather than on complex or short en echelon breaks. The roots of creosote bushes were stretched tautly across some fractures, indicating tectonic movement (fig. 145). Runoff from adjacent nearly flat terrain had eroded gullies as much as 2 m deep leading into the developing fissures. The water also eroded sub- horizontal tunnels along the plane of the fractures at various levels in the exposed upper 3 m. These tunnels evidently induced collapse well beyond the point of entry of the runoff. (Fig. 83 shows a trench excavated across a fracture of the central break about 20 m from the nearest collapsed section and 40 m from the nearest major entry point for surface runoff; this trench revealed a tunnel along the fracture 2 m below the surface; see also figs. 144 and 147.) The freshly exposed vertical walls and THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 193 FIGURE 145.—Detail of channel and collapse fissure shown at a in figure 144. A, October 1969. The tectonic fracture extends from upper left to lower right. Water entered the fracture from the right and eroded channels into the fracture. A taut root spans the open fracture at the left, whereas limp roots hang across the eroded channel to the right. Original tectonic opening of this fracture was 20— 30 mm. Peripheral cracks flank both the fracture and the channel. Desiccation cracks in the bottom indicate some filling of the fissure during the first episode of runoff and collapse in September 1969. Additional creep of several millimeters during 1970 and 1971 was not sufficient to lead to further collapse. B, January 1970. Runoff from the rains of November, 1969 has added 0.1—0.2 m of sediment to the fissure, but almost no further collapse has occurred. jumbled debris in the bottoms of the fissures on the south break indicated that collapse had continued after rain and inflow of the surface water had ceased, although how long after cessation is unknown (figs. 143A, 145A, 146A, and 148A). Some of the tectonic fractures (for example, the one in fig. 144) continued to extend after the rain and thus were prepared for further collapse during the next heavy rain. EFFECTS OF THE RAINS 0F NOVEMBER 1969 Heavy rains fell on the collapse fissures on the south break November 9 to 10, 1969, again causing surface runoff. Inspection in January 1970 revealed Rainfall was not intense enough to cause general sheet flow on the immediately surrounding surface. 0, December 1970. Runoff during one or more storms in 1970 filled and overflowed the sealed fissure but added no more than 200 mm of sediment to the bottom. Debris plastered on weeds growing in the channel immediately downstream from the fissure indicates a maximum water depth there of only about 70 mm; this depth implies a small sediment-carrying capacity for the channel. New desiccation polygons have formed in the fresh layer of silt and mud 2-10 mm thick on the surrounding surface. D, November 1971. Runoff from one or more storms since December 1970 has over- flowed the fissure again but has added little to the sedi- ment in the bottom. that rainfall had caused major changes in the fissures that formed in September 1969, most of which experienced little further creep after September, and in the few new fractures formed between September and November 1969 (figs. 1433, 1453, 1463, 1483, and 149). Most open collapse fissures that developed in Sep- tember were partly filled with waterborne and slumped mud and silt by January 1970; this filling created shallower rounded and polygonally cracked bottoms that contrasted with earlier steep walls and deep irregular bottoms (figs. 143 and 146). Many 194 THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 FIGURE 146.—Explanation on opposite page. THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 had been widened by additional collapse, and some of the gullies leading into collapse fissures had enlarged by both headward erosion and widening, although they were now also shallower because they and the fissures had been partly filled (figs. 143 and 148). The U-shaped cross sections and cracked mud of the bottoms of the fissures indicated that draining or drying of pools of muddy water were the last major events to occur, rather than foundering and slump- ing of walls as in September. In contrast, many of the fissures which had just begun to collapse in October, plus the tectonic frac- tures that developed or opened further after the September rain, had now collapsed more as a result of the November rains. Collapse, rather than ponding of water, was the last event to occur in these new fissures after the November rains, as it was for the older fissures after the September rains. A significant clue to the collapse mechanism, noted in October 1969, was the development of fractures parallel to the main fractures or fissures and usually within 1 m on both sides (figs. 143, 145, and 150). These cracks initially bounded small graben along the fracture and represented the first step of col- lapse visible at the surface. After the central blocks foundered, the side walls progressively caved in along similar cracks that developed farther ‘away from the fracture (figs. 143 and 146). Although a small number of peripheral cracks accompanied new tectonic fractures, most developed during ensuing weeks or months (fig. 150), presumably either by delayed collapse toward the free faces presented by FIGURE 146,—Detail of the collapse fissure at b, figure 144. The fracture runs directly away from the camera, near the center of the photograph. Water entered the fissure via channels from the right. A, October 1969. Collapse result- ing from the rains of September 1969. Walls are nearly vertical in the 3-m—deep fissure. Peripheral cracks define blocks that may collapse. B, January 1970. More water entered the fissure during the rain of early November and caused additional collapse before the fissure started to fill and flow ceased. The fissure evidently contained standing water at the end of the runoff episode, although some blocks collapsed after the bottom dried. 0, December 1970. Heavy rains of 1970 caused some additional collapse in the foreground of this photograph and extensive additional col- lapse in the background. Runoff in the foreground over- flowed the fissure. The high level of water in the fissure probably caused much of the new collapse. D, November 1971. Water overflowed the fissure again after December 1970 but did not deposit large amounts of sediment. Blown debris is accumulating along one wall of the slowly filling fissure. Vegetation alongside has grown rapidly since the fissure developed. 195 FIGURE 147.—Collapse across a silty mound demonstrates subsurface flow of water in the fissure from the channel extending from left to right beyond the mound. Almost no runoff poured into the parts of the fissure visible in this photograph. The tectonic fracture extends from the bottom of the photograph to the right middle distance. Loca— tion 19.2. the tectonic fractures or by removal of support by running water during subsequent rainstorms. These peripheral cracks bounded nearly all material that fell into the fissures, and their development appears to be a necessary step in the formation of collapse fissures. A noteworthy event after the heavy rain of No- vember 10 was the appearance of bands of short thick green grass in the zone of peripheral cracks along many collapse fissures and the gullies leading into them (figs. 1433, 1483, and 149). The grass greatly aided the search for small collapse fissures in January 1970 and made them particularly obvious from the air. At those places Where peripheral 196 FIGURE 148.—Collapse fissures on the south break (loc. 30.5). Fractures extend from lower left to right center. Water entered the fractures from the right. A, October 1969. September rains created deep, steep-walled fissures; a peripheral crack parallels the left side of the large fissure near the center of the photograph. B, January 1970. November rains have caused headward erosion of the chan- nels leading to the fissures. A block bounded by the pe- ripheral fracture visible in A has collapsed. Short grass surrounds the fissures and the gullies leading into them. C, December 1970. Rain and wind have further modified and filled the fissure, but the dramatic increase in vitality cracks were present, full-sized grassy bands even flanked some tectonic fractures that had not yet collapsed. In contrast, grassy bands were missing from the few uncollapsed tectonic cracks that lacked peripheral cracks. The grass almost certainly grew in response to the increased penetration of rain and surface water afforded by the peripheral cracks. (The low permeability of undisturbed surface silts in the vicinity is indicated by large areas that consist of small desiccation polygons that are typically 20— 50 mm wide and 2—6 mm thick; see fig. 1450.) Ex- cept for these 1/2- to 1-m-wide bands of grass along collapse fissures and newly opened tectonic fractures, grass grew elsewhere after this November rain only THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 and amount of vegetation next to the fissure is the most obvious change since January 1970. The same storms that caused overflow of the fissure shown in figures 145 and 146, 1 km distant, did not fill this fissure nor those shown in figure 149 with water and added very little sedi- ment to the fissures at this location. D, November 1971. Unusually large amounts of blown silt and sand during 1971 filled this and many nearby fissures. Although the grass and most of the small bushes have died, the large creosote bushes along the former fissure have grown noticeably since December 1970; they are far larger and greener than other bushes nearby. in Widely scattered small surface depressions of the otherwise nearly featureless surrounding terrain. The width of the bands of grass showed no relation to the greatly varied depth of the fissures — further evidence that the grass resulted from increased pene- tration of surface water into the cracked zone flank- ing the fissures rather than from a high level of water in the fissures. Many parts of fissures more than 2 m deep showed no evidence of having been filled with water, yet the grass was just as well developed on the surface next to these fissures as it was on the surface next to shallow fissures that showed evidence of ponded water less than 0.5 m from the top. THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 197 FIGURE 149. ——January 1970. Collapse fissures on the south break formed by runoff in September 1969 (10c. 30.2). Rain in November 1969 led to growth of grass in the zone of peripheral cracks flanking the fissures and the gullies leading into them. The rounded cracked bottom of the fissure shown in A is typical of fissures in the early stages of filling. The November rain caused little additional collapse except that shown in the lower right in B, Where peripheral cracks have opened further and some blocks with grassy tops have collapsed after the grass reached its present size (see text). The grass also demonstrated that some fissure walls collapsed long after flow of water ceased. Fig- ure 1493 shows grass-topped segments of slumped walls in a fissure. The segments were much too small to have supported the root structure of such thick grass. The‘blades of grass were perpendicular to the sloping tops of the segments, hence the slumping happened after the grass had reached its present size. Because the grass beside the fissure continued' to grow after the photograph (fig. 1493) was taken (for example, fig. 148C), slumping must have oc- curred shortly before the photograph was taken in January 1970, 5—8 weeks after water last entered the fissure in early November. CHANGES DURING 1970-71 Except for minor collapse along a few new frac- tures formed after November 1969, most collapse fissures gradually filled during 1970—71 (figs. 143C, 145C, 1460, and 148C). The small amount of creep during this period on both the central and south breaks has not lead to significant further collapse. Strong winds during 1971 filled many collapse fis- sures along the central and south breaks that were scarcely affected by blown debris during 1969 or 1970 (see fig. 148D). GENERAL OBSERVATIONS ON COLLAPSE FISSURES The history of development of collapse fissures shows that at some time after the onset of water- aggravated collapse, either the fissures became clogged with sediment and could no longer conduct water, or they filled with water and detritus. At many of the fractures described above, this clogging or filling evidently occurred during the rain of No- vember 1969,.which generally caused some further collapse before the fissures started to pond water. The low gradients (about 0.003) and relief of the areas surrounding the south break apparently do not enable runoff to carry large loads of sediment to the fissures. Most channels that lead into fissures on the south break are merely wide swales, some less than 0.1 m deep. They drain small areas, typically thou— sands to tens of thousands of square meters of nearly 198 FIGURE 150.—October 1969. Detail at 0, figure 144. Blocks are foundering between the tectonic (center) fracture (arrows) and the peripheral crack to the left. In the foreground the block next to the open fissure has dropped about 20 mm. The tectonic fracture is prerunoff (Septem- ber 1969), whereas peripheral cracks are postrunofl"; these facts indicate that some peripheral cracks form slowly after subsurface excavation by intercepted water flowing in the fracture. flat terrain, and transport relatively little sediment into the fissures (see figs. 145 and 1463, C, and D). Each of the increments of waterborne sediment shown in figures 145 and 146 is a small fraction of the total amount of sediment that has collapsed from the walls of these fissures. These observations indi— cate that most fissures on the south break do not act as receptacles for large amounts of waterborne sedi- ment from the surrounding surface. Indeed, the observed persistence of shallow sinks and depres— sions for years in areas free of large amounts of Wind-transported material along apparently inactive fractures of the south break implies locally low rates of water erosion and transport. THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 In contrast, parts of the few fractures and fissures along both the central and south breaks that devel- oped across major channels apparently filled with sediment during the initial or subsequent episodes of collapse. Evidently, the relatively large sediment load in these larger channels either rapidly clogged or filled nearby parts of these fractures and fissures. As a consequence, little or no collapse occurred on parts of many fractures that cross larger channels. The final stage of the existence of these collapse fissures, in the absence of renewed opening by creep or rapid filling by blown debris, appears to be grad- ual filling, with the persistence of alined shallow depressions or sinks for perhaps years or decades. The sinks may represent either residual unfilled parts of fissures, minor renewed opening and col- lapse, or later subsidence from compaction or grad- ual collapse of cavities in the fissure. Because an unfilled fissure collects surface water, small plants and the larger creosote bush and mesquite tend to concentrate and flourish along them (figs. 143 and 148). The sequence of formation, growth, and filling of collapse fissures, as determined from the foregoing observations, is summarized in figure 151. This se- quence should apply to the development of collapse regardless of whether the initial open fracture is caused by tectonism, desiccation, subsidence, or in FIGURE 151. — Hypothetical development of a collapse fissure. A, A new tectonic fracture may develop small peripheral cracks after the earthquake; slumping then takes place along these cracks. Without subsequent surface run- off, windblown sand and silt will eventually fill or seal the fracture. B, The first episode of surface runoff occurs before the fracture is clogged with debris. Water entering a nearby part of the fracture erodes a tunnel above a level that became clogged before or during runoff. The tunnel, by removing support from the overlying material, promotes the formation of peripheral failure planes. Blocks slump along the planes into the tunnel. C, End of the first episode of surface runoff. Water running in the fissure has removed most of the slumped material to other parts of the fissure and fracture, but after the water ceased flowing more material slumped in. After runofl" ceases, peripheral cracks continue to form back from the walls of the fissure. D, The second episode of surface runoff causes more slumping; the fissure, which became clogged after the first runofi" ceased or during the second runoff, begins to fill. Standing water at the end of the second episode of runofl" leaves a rounded muddy surface in the bottom of the fissure. Concentration of water during the first and later episodes of runoff causes pro- nounced increase in the vitality of bushes growing next to the fissure. E, The collapse fissure gradually fills with wind- borne and waterborne debris and leaves a line of shallow sinks and prominent bushes that may persist for years or decades. DEPTH {METERS} THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 8 FIGURE ISL—Explanation on opposite page. 199 200 some situations, piping. In photographs, these other features appear indistinguishable from the collapse fissures described here. A brief description of the development of desiccation fissures in Texas and Chihuahua by Underwood and DeFord (1969) ac- cords with the development observed in Lower Bor- rego Valley. MECHANISM OF COLLAPSE THE PROCESS or COLLAPSE For foundering of blocks and collapse to occur at all requires that the walls of fractures remain rela- tively coherent and that they not fail by crumbling or general slumping soon after opening. These condi- tions prevail in much of the unconsolidated Pleisto- ~ cene and Holocene alluvial and lacustrine sediments along the 1968 rupture. Vertical and overhanging walls are common along freshly cut margins of nearby channels, and some of these exceed 6 m in height. These observations indicate that tectonic fractures should also remain intact and open to a depth of at least 6 m below the surface. Collapse begins when blocks 0.5 to at least 4 m tall founder into tectonic fractures along nearly ver- tical surfaces (the peripheral cracks described above) that form as much as 1 111 back from the fractures or from the walls of the subsequent fis- sures. Water flowing in fractures or fissures removes the material of the foundered blocks; removal of this material leads to additional foundering. This removal of the foundered and slumped blocks is a necessary part of collapse, for, in general, the major tectonic fractures along the 1968 break opened less than about 100 mm. This amount of opening permitted only a comparable amount of settling of foundering blocks before they contacted the other side of the irregular surface of the tectonic fractures and stopped. A few small collapse fissures (less than about 0.3 m wide and 2 m deep) did develop without the help of any water. Some formed along the main breaks during the days immediately after the earthquake. Before any rain fell, some blocks dropped 10—50 mm between bounding tectonic fractures less than 1/2 m apart. Smaller graben along some new fractures completely crumbled out of sight. Such collapse must have occurred when blocks foundered and slumped a short distance and then disintegrated into the space created below in the tectonic fractures. How- ever, the amount of collapse caused in this fashion was very minor compared to that caused by water flowing into open fractures. Clearly, water removes to deeper levels the mate- rial that slumps into a growing fissure, as is shown by postrunoff fissures 1 m wide and 3—4 m deep that THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 develop from cracks that exhibit as little as 20- 30 mm of tectonic opening. The subhorizontal tunnels described above and the extension of collapse beyond the immediate points of entry of runoff indicate that some water flows in channels along the plane of the fracture near the base of the collapsing zone. Obser- vations of collapse fissures in different stages of evolution show that water flowing within fractures enlarges fissures in two ways. It removes material that slumps into and would otherwise fill and support the walls of the fissures, and it also undercuts the walls, inducing additional foundering and slumping. The evidence for extensive subsurface flow of water along low gradients within collapsing fissures shows that local clogging commonly impedes the downward movement of water in fissures. Such lat- eral flow may occur at many levels in the fractures and may lead to collapse between levels. Certainly, during collapse, large volumes of water must be continually eroding and widening conduits, whereas falling and collapsing walls of the enlarging fissures must be clogging them. The arrangement of passage- ways probably changes rapidly in the fissures as erosion and removal of material by water causes, but in turn is affected by, collapse. Collapse fissures cease enlarging when water can no longer leave the upper collapsed zone of the fis- sures as fast as it enters. This may happen during or immediately after episodes of runoff if waterborne sediment or the collapsing material clogs or fills the conduits and channels by which surface water de- scends the fractures, or it may happen if Windborne material clogs or fills the fissures between runoff events. Both types of blockage occurred along the collapse fissures associated with the 1968 break. Of course, it is important to remember that continued or renewed tectonic displacement (creep) can keep fissures open or reopen clogged ones, allowing addi— tional collapse. SPACE AVAILABLE IN TECI‘ONIC FRACTURES Probably the most intriguing aspect of the collapse fissures is the great volume of both water and debris (mud, silt, and sand) that disappears into them. The volume of sediment eroded from the fissures near the surface alone indicates large storage capacity at depth. Although direct observation from the surface was limited to the upper 4 m of the fissures, indirect evidence reveals many of the conditions that prob- ably exist at greater depths. It seems very unlikely that the soupy mix cascad- ing into the depths of a tectonic fracture penetrates very far into porous strata that may exist below. Porous strata next to recharge wells that inject water containing suspended sediments rapidly be- THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 come clogged (Sniegocki, 1962, p. F14—F15). This type of clogging constitutes a major problem with injection wells (for example, Meinzer, 1946, p. 197 ; Todd, 1959, p. 262; Sniegocki, 1962, p. F12). Per- haps a small amount of sediment penetrates strata in the lower parts of new fractures, but because of the tendency for the same fractures to break repeat— edly (see Clark and others, this volume), strata sur- rounding reopened fractures are likely to have been clogged during earlier episodes of infiltration. There- fore, it seems unlikely that much solid material removed by collapse and erosion from above goes anywhere except into lower parts of the fracture system itself, although the water may eventually penetrate through the clogged areas into porous strata. The existence in this area of cavernous voids underground or interconnected conduits emerging at the surface some distance away at lower elevation can be neither established nor disproved with avail- able observations. If underground conduits to remote surface openings or large underground storage places do exist, it would seem probable that at least some collapse fissures would enlarge without limit, but none have. All preearthquake and postearthquake fissures along the 1968 rupture have stopped grow- ing after reaching a maximum width of about 3 m and depth of 4 m. N 0 other evidence so far even sug- gests that cavernous voids underground or inter- connected conduits may be principal methods of removing collapsed solids.l The fact that collapse fissures along tectonic frac- tures have not increased beyond a certain maximum size implies that each tectonic event creates a finite storage volume. The most reasonable storage volume is that of the newly opened fractures themselves. In order to estimate the volume available in a tectonic fracture, we must determine the depths to which fractures can remain open. If water-saturated clay strata are present at depth, fractures in them could probably not remain open long because extru- sion of these moist clays would close the fractures. Clays make up perhaps 30—40 percent of the upper 200—300 feet of strata underlying the central break, according to drilling records of wells 5—10 miles north and northwest (Moyle, 1968). Water is 15— 20 m below the surface in wells 3—4 km east and west of the collapse fissures near Old Kane Spring Road 1A minor example of piping occurred on the south break at location 30.6 at the intersection of a. channel 1—2 m deep with a fracture that reopened after January 1970. Two subhorizontal tunnels formed in the fracture at the level of the bottom of the channel and conducted runoff into the channel. Nearby fractures, however, collapsed to levels deeper than the bottom of this channel, so this piping could not have been the principal mechanism for re- moval of water and debris. 201 (Moyle, 1968). In 1969, water was 27 m below the surface in an abandoned well about 11/; km southeast of the collapse fissures on the south break. These depths represent minimum distances to saturated sediments, because the water in the wells may have risen from deeper strata under artesian pressure. Thus, 15—30 m seems to be a maximum depth that fractures might remain open along the central and south breaks, if extrusion of water-saturated clay limits the maximum depth of open fractures. Whether the clay, silt, sand, and gravel strata above the level of water saturation are able to main- tain an open fracture remains a problem. This prob— lem cannot be completely solved without knowledge of the physical properties of the material at depth. T. L. Youd (written commun., 1970) estimates that the material would require a cohesive strength of roughly 70,000 N/m2 (newtons per square meter) to maintain an open crack 30 m below the surface and half that at 15 m. These strengths fall within the 35,000—100,000 N/m2 range (about 5—15 psi) determined for dry loess (Gibbs and others, 1960, p. 126—127) and seem at least plausible for the allu- vial and lacustrine sediments underlying the 1968 break. If this estimate of strength is correct, the fractures probably remain open to a depth of at least 15 m and possibly as much as 30 m’ below the surface. The space available in such fractures to receive material carried down from the surface can be calcu- lated. The net volume in a fracture that is 30 mm wide and 15 m deep is 0.45 m3 for each meter of length of the fault. If the fracture is open to a depth of 27 m, then the volume is about 0.8 m3 for each meter of length of the fault. If the fracture opens 60 mm wide, the volumes double, and so on. Considering only the volume of sediment derived from collapsing walls, the space in a fracture 30 mm wide and 27 m deep could accept enough material to create a surface fissure about 0.9 m wide and 0.9 In deep or 0.4 m wide and 2 m deep if all the water entered surrounding strata and if solids filled the fracture at the same density they had before collapse (for this approximate calculation, we ignore the fact that the collapsed part of the original fracture at the surface is not available for storage). These collapsed dimensions are compatible with those of the most continuous collapse fissure near the south break (0 in fig. 152; see below) ; this col- lapse fissure opened probably no more than about 30 mm and formed where the water table is probably at least 27 m deep. The cross-sectional area of this fissure ranges from about 0.1 m2 near the north end to perhaps 0.5—1 m2 locally elsewhere. The average collapsed volume along the entire length of the fissure 202 is probably between 0.2 and 0.8 m3 per meter of length. Assuming a tectonic opening of 30 mm, the fracture can supply the necessary storage volume for this observed range of average collapsed volume if it remains open to depths of 7 m and 27 m, respectively. However, fissures 1—3 m wide and 3—4 m deep that developed on fractures 20—30 mm Wide have already been described, and the volumes of these fissures cannot be accommodated in the fracture below, even using the 27 m depth. A solution to this problem is that these large collapsed sections are seldom more than 3-5 m long, and they generally occur on frac- tures that have many uncollapsed or only slightly collapsed parts (fig. 144). Indeed, most fractures that have collapsed since 1968 near the south break —where the deepest and widest fissures occur— have about as many uncollapsed as collapsed parts. Water flowing within the fissures must commonly remove material from one part of the fissure and deposit it elsewhere in the lower parts of the same fracture beneath sections that did not collapse. Hence, to an initial approximation, each collapse fissure studied along the 1968 break appears to have enough volume at depth to receive the amount of material that has been carried below plus minor amounts of sediment brought in by runoff. CONDITIONS CONTROLLING THE DEVELOPMENT OF COLLAPSE FISSURES If the preceding explanation for the development of large collapse fissures from tectonic fractures is correct, then the conditions necessary for their cre- ation are fairly restrictive. The primary require- ments are simple: The fractures must remain open to a depth of at least 10 m or so, and water sufficient to erode or to undermine material from the walls, but not so loaded with sediment as to clog or fill the fractures, must enter before other agents, such as wind or slumping, close them. Many factors may affect these primary requirements. Certainly, the physical properties and conditions of the material in which fractures form are critical. Such properties as cohesion and shear strength and such conditions as moisture content and depth to the zone of water saturation not only affect the depth to which the crack will remain open but also prob- ably influence the nature of the fracture itself—— whether it will be a single fracture or a narrow band of multiple, interconnected, or en echelon fractures, which may not be so likely as a single break to offer an open path to lower levels. Delivery of a relatively large volume of water with a low suspended particle load to many places along tectonic fractures requires some minimum in- THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 tensity and quantity of rain and is facilitated by surfaces with low slope, relief, and permeability. If the surface is also largely free of vegetation, a lower intensity or total amount of rainfall will supply the necessary amount of runoff, although lack of vege- tation may also increase the sediment load. Thus collapse fissures presumably will not form in soil too weak to maintain open cracks or in places where surface runoff is either small or quickly localized into narrow channels. Indeed, collapse fissures may not be a certain con- sequence of each episode of surface faulting in Lower Borrego Valley. The relations observed between creep and collapse imply that at least some collapse fissures depend on a delicately timed sequence of tectonic opening and heavy rain. For large collapse fissures to occur at all, heavy rain must fall before tectonic fractures are sealed near the surface. Wide- spread development of collapse fissures, as in 1968— 69, may be a relatively unusual part of the tectonic activity of this part of the Coyote Creek fault. OTHER COLLAPSE FISSURES IN LOWER BORREGO VALLEY PREEARTHQUAKE AND RELICT COLLAPSE FISSURES ALONG THE COYOTE CREEK FAULT Four groups of preearthquake collapse fissures are recognizable along the south break (fig. 152). The IKILOMETER <_ . Bose from U.S. Geological Survey I 24,000 Horpers Well, I958 FIGURE 152.——Relict (preearthquake) collapse fissures (a—d) near the south break of the 1968 rupture along the Coyote Creek fault. THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 oldest developed before 1953; the youngest opened shortly before the earthquake, and abundant vegeta- tion has not yet grown. Alined vegetation, with or without shallow sinks, marks the three older groups. These older linear features can be identified as relict tectonic collapse fissures because ( 1) they are par- allel to major en echelon fractures of 1968, (2) they are located next to the 1968 break, and (3) they are the older members of a gradational series that ends with nearby younger fissures created since the earth- quake. The oldest two groups of fissures (fig. 152, a and d) are now mostly filled and are represented by alined creosote bushes, small brush-stabilized dunes, and linear sinks less than 0.5 m deep and 2 m long. These oldest relict fissures are distinctly more con- spicuous on 1953 aerial photos than they are on 1968 and 1969 photographs; indeed, on the 1953 photos, one at d (fig. 152) appears to have an open collapse fissure, although the scale does not permit positive identification. The relict fissure at b (fig. 152) consists of alined bushes, dunes, and sinks like those at a and d but is evidently younger. Unlike the older fissures, most of it is not visible on the 1953 photos, and in 1969, the alined sinks of the southern part were nearly con- tinuous and deeper than those of the fissures ‘at a and d. Furthermore, the alined bushes of this fissure have become increasingly prominent since 1953, when only a few grew near the northern end. More bushes appear on a 1956 aerial photograph (fig. 38), and by 1968, large and prominent bushes marked the entire length shown in figure 152. The discontinuous nature of this relict fissure and the rounded shallow sinks along it indicate that the fissure stopped grow- ing several years ago. Moreover, it showed no evi- dence of fresh cracking or opening at the surface either shortly before or since the earthquake, even though it nearly joins one of the two new fractures formed by postearthquake creep in September 1969 (e, fig. 49). By far the most prominent collapse fissure — pre- earthquake or postearthquake — along the entire fault is at 0 (fig. 152). When first seen on April 10, 1968, it was a straight, almost continuously open fissure nearly 500 m long and as much as 2 m wide and deep (figs. 153—155). Moist polygonally cracked mud in the bottom, presumably created by drying after inflow of water during heavy preearthquake rains of April or March 1968, did not crack further during the earthquake. This fissure is not visible on the aerial photographs of 1953 and 1956, and thus was almost certainly not in existence thenbecause it is very conspicuous on 1968 aerial photographs of similar scale. 203 FIGURE 153. —-Aerial view southward over part of the most prominent preearthquake fissure (fig. 152, c). This photo- graph shows a 400-m length of the fissure as it appeared on April 10, 1968. The southeast part of the south break passes immediately in front of the dense vegetation in the middle distance (figs. 154 and 155 show details at a and b). Photograph by J. P. Lockwood, US. Geological Survey. At least some of the collapse took place less than a year before the earthquake, although parts of the fissure could have been older. The southernmost section of fissure 6 had the steep walls and deep narrow bottom typical of initial collapse of a new fracture. Moreover, a relatively fresh set of tire tracks crossed the fissure at a place that had subse- quently collapsed. As seen a few days after the earthquake, the tracks were obvious and appeared fresher than many l-year-old tracks studied along the break after the earthquake. Other parts of fissure c looked as though they had been subjected to two or more episodes of surface runoff, because they had the U-shaped cross section and mud polygons of a fissure that is filling. The abundance of small bushes in and along the northernmost part of the fissure suggested that that section was more than several months old. In 1968 much of this fissure had not yet stimulated a vigorous growth of creosote bushes typical of the older collapse fissures nearby; thus, it was at most only a few years old. Two large mes- quite-stabilized dunes athwart the fissure imply the existence of an earlier fracture in the same location, but the position of these dunes could be fortuitous. Preearthquake rainfall as late as March and April of 1968 (fig. 46) might have caused most of the lat- est collapse, although it seems more likely that rain- fall in 1967 or earlier caused some of the initial collapse. Nevertheless, the condition of the fissure, the tire tracks, and the pattern of rainfall in early 1968 indicate at least some collapse shortly before the earthquake. Although this fissure is longer, is farther from the fault, and has a slightly different trend (fig. 152) 204 FIGURE 154. — South end of the relict collapse fissure shown at a in figure 153. Water entering from the left has eroded deep gullies into the fissure. Peripheral cracks bound the blocks that would slump next or that would founder into the fissure if it were growing. However, the fissure has been slowly filling with waterborne material since it was first observed in April 1968, although it opened slightly and some minor collapse occurred during 1970. than the other relict and younger fissures nearby, it appears to have had the same tectonic origin. Frac- tures and fissures somewhat similar in appearance to these relict fissures have been described, but they result from other causes, such as desiccation (for example, Lang, 1943; Knechtel, 1952; Willden and Mabey, 1961 ; Neal and others, 1968; Underwood and DeFord, 1969), piping (for example, Carroll, 1949; Fletcher and others, 1954), broad subsidence, pos- sibly related to lowering of artesian head (Robinson and Peterson, 1962; Kam, 1965), or near-surface compaction (hydrocompaction) at the time of initial wetting of alluvial deposits (Bull, 1964, 1970; Lof- gren, 1969). THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 FIGURE 155.—A taut root across the relict collapse fissure at b in figure 153 may have been stretched by tectonic opening of the original fracture. The bottom of the fissure shows the shrinkage polygons and rounded cross section typical of a collapse fissure that is filling rather than deep- enmg. The preearthquake fissures of figure 152 along the south break lack important characteristics of each of these other types. The fissures of the south break have neither the polygonal arrangement nor playa setting of desiccation cracks. They do not trend toward the walls of Carrizo Wash (pl. 1), the only nearby free face sufficiently high to account for depths observed in the fissures (about 4 m), if they have been caused by collapse of subsurface pipes. These fissures bound no apparent subsiding area; the closest pumping of ground water since before the earthquake is at a ranch nearly 10 km to the north, and the older fissures predate these wells. Finally, these fissures show neither the proximity nor orientation parallel to channels that might be causing hydrocompaction of adjacent sediment (Bull, 1970, p. 55, 152). The straight channels that cross the south break at location 30.6 (across the relict fissure at a, fig. 152) are actively deepening their channels, which evidently follow old roads or man- made courses. These channels receive runoff concen— trated by culverts under the nearby railroad. If this augmented flow across the fault were causing hydro- compaction, the channels should subside and frac- tures should form parallel to them, but neither is apparent. Thus there is no convincing evidence that these relict fissures have developed from anything except tectonic fractures associated with the Coyote Creek fault zone. Moreover, groups of small alined preearthquake sinks along the south break at loca- tions 28.4 and 32.0 provide further evidence of tectonic activity shortly before the earthquake. These sinks represent partly filled collapse fissures and THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 A 0 100 200 I; l 1 300 1 205 400 l 500 METERS B l FIGURE 156. ——Vertical aerial photographs of concentric collapse fissures at location 18. A, April 1953. Circles (apparently older) are faintly visible north and northwest of the prominent circles. B, April 1969. Although many new concentric fissures have formed between the two prominent sets, many fissures visible in A are much less prominent or no longer evident. The distinct E-W lineament is Old Kane Spring Road. appear to be comparable in age to the young pre- earthquake fissure at 0 (fig. 152). OTHER FISSURES IN LOWER BORREGO VALLEY Two groups of fissures and relict fissures that are not obviously related to the fault occur near the 1968 rupture. The larger group lies 1—2 km southwest of the south break alongside and Within one-half kilome- ter of the base of the Fish Creek Mountains between locations 28 and 32. The smaller group is about 10 km to the northwest near uplifted sediments at location 18, about 3 km west of the central break (fig. 35). For the most part, these fissures consist of continu- ous and discontinuous shallow sinks in circular, straight, or irregular alinements. Bushes are concen- trated along all the fissures, which range in length from tens to hundreds of meters. A few sinks are as much as 1 m wide and 2 m deep and are nearly continuous for tens of meters. Some fissures are merely conspicuous lines of bushes that connect widely spaced small (less than 0.5 m wide) sinks. Many of the fissures are apparent on 1953 aerial photographs. Most of the fissures fall into two classes. All those in the smaller northwestern group and many in the larger southern group form clusters of as many as seven or eight concentric rings within a diameter of 30—150 m (fig. 156). The other class, entirely in the southern group, comprises many nearly straight fissures that roughly parallel the adjacent front of the Fish Creek Mountains. Although these features were not systematically mapped, aerial photographs taken in 1969 recorded some of them and revealed definite changes from 1953. The concentric fissures in particular show ob- vious evolution between 1953 and 1969. In the southern group, several circular fissures formed in new locations after 1953, whereas in the group to the northwest, many new concentric fissures devel- oped around those present in 1953. In contrast, many of the less prominent fissures of 1953 (presumably shallow sinks) had almost entirely disappeared by 1969 (fig. 156). The nearly straight fissures along the Fish Creek Mountains show similar changes, and partial aerial photographic coverage of the southern group since the earthquake suggests that new frac- tures are appearing more rapidly than old ones are disappearing. There seems little reason to doubt that these fis- sures developed by collapse from open fractures, but the origin of the fractures remains uncertain. Except for some of those in the northwestern group, none of the fissures showed signs of active enlargement at the time of field observations in December 1970, nor were any uncollapsed fractures found. It is difficult to imagine that concentric fractures 150 m or less in diameter could have originated at great depth or been formed by tectonic forces. The circular 206 fissures closely resemble “ring fissures” described by Neal, Langer, and Kerr (1968, p. 86). Ring fis— sures supposedly develop from desiccation contrac- tion of soil around a phreatophyte as the plant lowers the water table and capillary fringe directly beneath it. Although no single central phreatophyte (mesquite, in this area) is obvious in the rings in figure 156, several of the concentric fissures in the southern group encircle individual large mesquite bushes. In the absence of any direct evidence for some other form of local compaction or subsidence, the phreatophyte explanation seems the most reason- able for the circular fissures. The position of the straight fissures of the southern group alongside and subparallel to the mountain front suggests that more widespread subsidence of the sediments might result from either tectonic movement or, possibly, compac- tion. Regardless of their origin, these unusual out- lying fissures demonstrate the potential for collapse in the sediments of Lower Borrego Valley beyond the immediate vicinity of the Coyote Creek fault zone. DEVELOPMENT OF COLLAPSE FISSURES ALONG OTHER ACTIVE FAULTS Few investigators have reported collapse of tec- tonic fractures elsewhere. Neither the detailed de- scriptions nor abundant illustrations of the report on the 1906 California earthquake include collapse fis- sures (Lawson and others, 1908). None were men- tioned in the brief report on the Imperial earthquake of 1940 (Ulrich, 1941). The intensive investigation of the San Fernando earthquake of February 9, 1971, revealed no collapse fissures (US Geological Survey, 1971) despite intense rain February 17th, 8 days later (Environmental Data Service, 1971, p. 47). However, in 1907, W. D. Johnson photographed and described sinks as much as 1 m deep that had devel- oped and persisted on the breaks of the 1872 earth- quake in Owens Valley, California (unpub. data, US Geological Survey library, Denver). Many of these sinks were little changed in 1971. Some of the sinks developed, however, in sandy alluvium within 100 m upstream of a scarp 6—7 m high, and piping could have caused these, rather than the mechanism proposed for the fissures in Lower Borrego Valley. Warne (1955) described fractures caused by the Arvin-Tehachapi earthquake of 1952 that inter— cepted irrigation water, and he included a photo- graph of what appears to be subsequent collapse of such a fracture (his fig. 20). Although no collapse fissures were reported from the ground breaks that formed at the time of the Parkfield-Cholame earth- quake of 1966 on the San Andreas fault (Brown and THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 others, 1967), R. E. Wallace has photographs of some fissures (written commun., 1971) that were present 1 year later along the fractures. What con- nection, if any, these fissures have to creep and rainfall is unknown. The fact that a few collapse fissures have been associated with other earthquakes, plus the develop- ment of similar fissures elsewhere from piping, des- iccation, subsidence, or unknown causes, indicates that the conditions for their formation are not unique to Lower Borrego Valley. Fissures will develop along earth fractures whenever all the necessary conditions are present, whatever those may be. That most of the reports of fissures in the United States are re— stricted to arid or semiarid areas strongly suggests that in desert areas soil and sediment, surface char— acteristics, rainfall patterns, ground-water condi- tions, or combinations of these or other factors are most favorable for the development of fissures. ENGINEERING SIGNIFICANCE OF COLLAPSE FISSURES Collapse fissures of the size that developed along the 1968 rupture present a distinct but manageable hazard to manmade structures. Moreover, if the mechanism of collapse proposed here is correct, col- lapse can be prevented by diverting runoff away from new fractures. Once collapse has started, it can be stopped by dumping sand and dirt into the growing fissures. Fissures that have stopped grow- ing, of course, can easily be filled. Structures threatened by collapse of fractures are also threatened by the generally greater hazard of ground displacement. Tectonic opening or continuing creep sufficient to permit collapse is likely to do more damage to most structures built across fractures than will subsequent collapse. Moreover, collapse can be controlled; tectonic movement cannot. SIGNIFICANCE OF COLLAPSE FISSURES IN FAULT STUDIES Probably the most important characteristic of collapse fissures in fault studies is that their pres- ence in a fault zone is one of the most obvious signs of recent fracturing. Along the 1968 rupture, new fractures formed by creep were discovered only by investigators who crossed them on foot, but collapse fissures that developed from these fractures were easily visible from cars and airplanes or on aerial photographs. Although collapse is never a necessary consequence of fracturing, new collapse fissures in a fault zone are an almost certain indication that new fractures have formed, either during an earth- quake or, afterward, from creep. THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 REFERENCES CITED Brown, R. D., Jr., Vedder, J. G., Wallace, R. E., Roth, E. F., Yerkes, R. F., Castle, R. 0., Waananen, A. 0., Page, R. W., and Eaton, J. P., 1967, The Parkfield-Cholame California, earthquakes of June—August 1966—surface geologic effects, water resources aspects, and preliminary seismic data: U.S. Geol. Survey Prof. Paper 579, 66 p. Bull, W. B., 1964, Alluvial fans and near-surface subsidence in western Fresno County, California: U.S. Geol. Survey Prof. Paper 437—A, 71 p. 1970, Prehistoric near-surface subsidence cracks in western Fresno County, California: U.S. Geol. Survey open-file report, 256 p. Carroll, P. H., 1949, Soil piping in southeast Arizona: U.S. Dept. Agriculture Soil Conservation Service Regional Bull. 110, Soil Series 13, 21 p. Environmental Data Service, 1971, Climatological data, Cali- fornia: Natl. Oceanog. and Atmospheric Adm., v. 75. Fletcher, J. E., Harris, K., Peterson, H. B., and Chandler, V. N., 1954, Piping: Am. Geophys. Union Trans., v. 35, p. 258-263. Gibbs, H. J., Hills, J. W., Holtz, W. G., and Walker, F. C., 1960, Shear strength of cohesive soils, in Research con- ference on shear strength of cohesive soils, Boulder, Colo., 1960: Am. Soc. Civil Engineers, Soil Mechanics and Found. Div., p. 33-162. Kam, William, 1965, Earth cracks—a cause of gullying, in Geological Survey research 1965: U.S. Geol. Survey Prof. Paper 525—B, p. B122—B125. Knechtel, M. M., 1952, Pimpled plains of eastern Oklahoma: Geol. Soc. America Bull., v. 63, p. 689—700. Lang, W. B., 1943, Gigantic drying cracks in Animas Valley, New Mexico: Science, v. 98, p. 583—584. Lawson, A. C., and others, 1908, The California earthquake of April 18, 1906, Report of the state earthquake investi- gation commission: Carnegie Inst. of Washington, v. 1, 451 p. U. S. GOVERNMENT PRINTING OFFICE: 1972 O — 476-246 207 Lofgren, B. E., 1969, Land subsidence due to application of water, in Varnes, D. J., and Kiersch, G., eds.,- Reviews in engineering geology, v. 2: Geol. Soc. America, 350 p. Meinzer, O. E., 1946, General principles of ground water recharge: Econ. Geology, v. 41, no. 3, p. 191—201. Moyle, W. R., Jr., 1968, Water wells and springs in Borrego, Carrizo, and San, Felipe Valley areas, San Diego and Imperial Counties, California: California Dept. Water Resources Bull. 91—15, 16 p. Neal, J. T., Langer, A. M., and Kerr, P. F., 1968, Giant desiccation polygons of Great Basin playas: Geol. Soc. America Bull., v. 79, p. 69—90. Robinson, G. M., and Peterson, D. E., 1962, Notes on earth fissures in southern Arizona: U.S. Geol. Survey Circ. 466, 7 p. Sniegocki, R. T., 1962, Problems in artificial recharge through wells in the Grand Prairie region, Arkansas: U.S. Geol. Survey Water-Supply Paper 1615-F, 25 p. Todd, D. K., 1959, Ground water hydrology: New York, John Wiley & Sons, Inc., 336 p. Ulrich, F. P., 1941, The Imperial Valley earthquakes of 1940: Seismol. Soc. America Bull., v. 31, p. 13-31. Underwood, J. R., Jr., and DeFord, R. K., 1969, Large-scale desiccation fissures in alluvium, Trans-Pecos Texas and northern Chihuahua [Abs]: Geol. Soc. America Abs. with Programs, v. 1, pt. 2, p. 31. U.S. Geological Survey, 1971, Surface faulting, in The San Fernando, California earthquake of February 9, 1971: U.S. Geol. Survey Prof. Paper 733, p. 55—76. Warne, A. H., 1955, Ground fracture patterns in the south- ern San Joaquin Valley resulting from the Arvin- Tehachapi earthquake, in Oakeshott, G. B., ed., Earth- quakes in Kern County, California during 1952: California Div. Mines Bull. 171, p. 57-66. Willden, Ronald and Mabey, D. R., 1961, Giant desiccation fissures on the Black Rock and Smoke Creek Deserts, Nevada: Science, v. 133, p. 1359—1360. UNITED STATES DEPARTMENT OF THE INTERIOR - PROFESSIONAL PAPER 787 PLATE 1 GEOLOGICAL SURVEY EXPLANATION 120 50 \ \\\’—‘\.~\_\\~\‘\ \l /’-—" W10m HBOLL . 1 Surface rupture I] V J l \ ‘l ’y “ 1 I, I x“ . Y " 3 4 I I x ’ ‘ Many en echelon lines are schematic and include some fractures too small to be ( H V I 1 V l l V I ' .I » shown at the map scale. Elsewhere, en echelon, multiple, and complex breaks, as well as single fractures, that are too small to show at the map scale are represented by a single line. Arrow and number indicate apparent direction and amount of horizontal displacement, in millimeters. Left-lateral offsets are further identified with LL. Number without symbol represents amount of vertical displacement, in millimeters; number on upthrown side.W10 m, width of fracture zone, in meters (80-100) [210] ~ \Tx:~\\\\x\\\\ \—\R\\ Displacements in parentheses were measured in June 1968; those in brackets were measured in March 1969 along the central break and in October 1969 along the south break. All others were measured in April 1968 blown sand ‘ A4048 1L6“ 30” 33°07’30“ 2 I9 , 33°07’36" 33"l?.’30" 7'W30m; no consistent horizontals-LII , _I,orver'tical movement l l ‘ , l16°07l30“ \\\" \ I > //' / a 2'9"” I “l /10 Small right and left- fl" lateral offsets BIreakgenerIa’lly singl ) withIn/ I1. 5m of base of, ‘slope; t ’pically _ ,Ilr’ IIOO righI't lateral, 3 )— 8 vertical II \ / Ill ,, overtic lor T2"; I 21911 aieralfl§ 030‘ \ \ \ Base from US. Geological Survey 1:24,000 Shell Reef and Harper Canyon, 1959; Bor- rego Mountain SE, 1958; Harpers Well, 1956; and Borrego Mountain, 1960 ‘xI \ ‘\ W?m(‘~VeLtIl10-2olLL 30 50 \\ , TX“ \\\\\l180 cumulatIve crc§s zone 11601230" ——'IW1Om\ ago/N wsorri\ ‘S‘ON W20m ,\_ /40l, ‘\x\ . (20)\\ \I‘ (4e5p)o~II'\- ||\\ cu ua iv Vl<3¥n fin ec align \ 100’ urfto 1mdeep . k4/DE‘;\ County IEWZIl 7 W0.2-0.8m; steps over IV I “ swine“ \- :I was approximately But every 100m 10045.10, Northeast side up?! (150-230)“, at this location; I250 300) Graben 50-80 deed“; 33” 10' 116W)?! I (SQ openI—u (109pen)/’ ‘w _ (30 ‘openl/ //l //,(10,Iopen)/g on cracks not YinedI one savvtooth crack .730 open ~ (gdépe ) Z/i/HOO open) Inn\ \II l‘ (10 ' l I (ZOZZEeTIIZ: j / /“II\\ {$2} I ‘ .W15m ’/ \fo ‘OCOIILLO/VlI/ELIS \ \ ' \COUNW AIRPORT Rs Wl/Difxf‘use fractures, W45m -, a /——(5 open) / 2/(20 openl / a \ Bensonx \ \If\ Lake \W /// (20 open) 5 openl/II g, I, 11 / Hairline breaks; \ offset or opening ".7 less than 5 / . 0‘7 _ \offset or ope r \ less than 5 I?) ~ WV 'xm yx'flw,1“r 33"2’30" {3 ‘-‘ 10; sage fractu H . . most show 5—1 33” 07'30” V14? TRUE NORTH I l. :2 o z 9 a Lu 3 0 w E APPROXIMATE MEAN DECLINATION, 197 2 116° 07'3 O” ‘*~\A rea of many \craeks-Y I12“, offset note)d SCALE 1:24 000 (30) ”A‘s \ \(30-60I, 1 1/2 0 MILE I———I l—l lj l l L i l _._l—| 1.5 O l KILOMETER l'—l_l—l I—I l-—-l I—-I l-——-———-———-—————--—-:l CONTOUR INTERVAL 10,20, AND 40 FEET SUPPLEMENTARY CONTOURS AT SVFOOT INTERVALS DATUM IS MEAN SEA LEVEL 115‘ 5.7130” \ Fractures mapped between April 9,1968 and March 1971 '\ by C.R. Allen, FLO. Castle, M.M. Clark, Arthur Grantz, R.V. Sharp, T.G. Theodore, E.W. Wolfe, T.L. Youd, and geology students from California Institute of Technology llfi"()0’ R, 9 E: l MAP SHOWING SURFACE RUPTURES CREATED AT THE TIME OF AND AFTER THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 (G.M.T.) 116°05’ 476—246 0 — 72 (In pocket) No. l PROFESSIONAL PAPER 787 PLATE 2 UNITED STATES DEPARTMENT OF THE INTERIOR GEOLOGICAL SURVEY EXPLANATION 5R 90° U 0-5V D 800 Single break or closely spaced short en echelon breaks, showing meas- ured displacement in millimeters, and dip of fault plane R, right-lateral horizontal displacement; V, vertical displacement. Arrows indicate relative horizontal displacement. U/D, relatively upthrown and downthrown sides where vertical displacement locally accompanied the dominant horizontal displacement. Breaks are estimated to be generally within 75 feet of true position in hilly terrain and within 150 feet in featureless terrain \\\ Generalized location of long en echelon fault segments consisting of single breaks or short en echelon breaks Base from US Geological Survey 1:24.000 ‘ Kane Spring, 1956; Superstition Mountain, 1956; and Brawley NW, 1957 JOINS 7STR|P A Fault trace mapped from poorly preserved (sandblasted or sand-and silt-filled) breaks AC” Superstition Hills Geodetic network across fault CIT creepmeter >\. . .~ :1 . ‘ 7 » ; 71177 ~ , 7 ‘9R KANE sari/N6 25' t 7 @7777 7 58 . 1 x ~ SUPERST/T/ON MTN77.5' I UI‘CITcéfpmeter 77 , . 1 ,7 , ‘ ‘ \ 77/ \‘\ 7 ’ J 7 1 7 77(1/7NA , 1 , 7 - , 1:; M v.77 _ 177 7 7 w L 77 7 .\7 1 W 7 77 1 “\777 7 l l ,, l . \ x, \ ‘10- 1511 A “”3 . ‘ Row of bushes ' without cracksv Cracks step right 77 Lo , . r 8 N to N E E 2 >. 0 Lu ' I: \1 1: E ii (I) a: 0: I. 1 I 1 ‘67” 0:1 11615 115 45 D1 ,. . . o: - - filfil’e‘é‘ co . Borrego Springs 33°15; E Q: ’_— (D Z O -> 33L plate 2 - A f 0 5 10 MILES Strip 8 fligeifiii‘sa LI_J_1_.J_l_L—l—l—J—J EICentro 'il ‘2) 32°45— ——————————— 11:1 .3132 \ INDEX MAP SHOWING LOCATION OF \7 STRIP MAP SEGMENTS \ w 7\ 1 A E :AL \\ BRAWLEY NW7 5' . . l ‘ :1' L Breaks mapped by Arthur Grantz and Max Wyss, May 13— 15 1968 37);] 7‘ I” Pig‘i‘il IMEEE’ ‘5 -1-” 7 7 ,,,,, - D Ll‘ l 1 I 1 .. l .1 ’ V‘Eamhest NW point of break. ‘ ‘ 7 . __ ,_ ‘ F" 1 SCALE 1:24 000 7 . N‘onfrac'tures found on 1‘77‘ . , _ “5‘ 7“ 71:: 1" "’ “ ' - 7» »L 1 L ‘ 1’1'7 77 1’ projectian toNW-5/8/68 H 7 ‘ ‘: “ {H “'7'" “=7. 5—H: ‘ 1 7/ 1 1/2 0 1 M‘LE ,7 7 77 7 7, .. . 1 E ‘5 ‘ ‘5 l ‘ T 3- 6R; ’rfia‘tel ‘ '~ ,_ .L7 zone afiPYOKI y 77 1 .5 O l KlLOMETER r—-11——11——11——11——11———————E CONTOUR INTERVAL 10, 20, AND 40 FEET SUPPLEMENTARY CONTOURS AT 5-FOOT INTERVALS DATUM IS MEAN SEA LEVEL 0. 5m wide with intermittent en echelon cracks “”m‘,‘ ‘ CIT creepmeten 1 01d scarp 0.5m high Very prominent 6-18Rf é‘ghelon i , tint/11.11 \ ‘61le. NorthlShore 6‘8R fractures 1-3m7 long; i '3 i " 77 f\7 X , , - zone 1-1.5m Wide ’3 ‘ \77 7‘7} ir‘lM. . 1‘11: En echelon fractures, 3R; Ven‘gbhdlon /\ strike N“ 120 W. fractures strike \ ----—e~—*1 em Mme-Liv? ~17 #1 Old scarp 0.3-1m high, 1 upon NE Base from U. 8. Geological Survey 1: 24,000 9.5»; 2‘6R in mud crack_ .1 zone 0.5m Wide ' , 1 77 New cracks over old 6- 13R; zone 0.31 .5111 wl‘db// 7 7 7 en echelon fractures N. 22& JOINS STRIP D i” 7‘ I‘ ‘1‘" 1 \ 33 \ 7\\ EXPLANATION T 7 i 7 i l V \ \ echelon ffidlures 1 g _—..—. eN¢2§°W w. \ 7 , ~ , L 7 1 /5R,0V _7 u A ‘f \ 1 3-13 R; along crest f dune 77 ": ‘ r“, 1. ‘ : /‘ o D 77\ \\ 7 7 " ’ “Ht” 80 7_ 7 7 “ 1’ En echelon breaks, showing measured displacement, in millimeters, "‘ :‘g‘ 631., , .. DUN 11 7/ l and dip of fault plane 7 77 - \7 77 l 7 ‘5 11:“ h 777 7‘7l :. i 7 ‘ 777 l 7_ ~ 1' 77 R, right— lateral horizontal displacement; V, vertical displacement. Arrows indicate \ "" 1 ”” “f m” f' 17" ' "7" 7' 1 ' ' 7 in: , .\ '1~.\ 7/ E ‘ \ ‘ . NZ“ .7, 7 relative horizontal displacement. U/D, relatively upthrOWn and downthrown \\ 1 ‘ 777 7. \ i En echelon fractures, 34’ 1313717 I" zone 112m mag; ’ 7/ _ sides along older fault trace. Breaks are estimated to be within 75 feet of .\7 I 17 \ - 77 .. \\ 29 strike N. 25° -30 MW. 7\ @R 7 helm? 7877777795 1' :f\7\ true position in hilly terrain and within 150 feet in featureless terrain 7 7 77 .\ 7 .s i A 7 \ en ec 7 ‘ \“ l ‘7‘ ‘ '* 1 11‘ K “HALL. _. ,, "H7 \ I. , 1.: 7Q. 3- 2m long, i_‘(e‘NL7H71570-30 \ 7 .. 7 7 . 7 . , l :/ v;_\ ‘ \7 1 CIT Red Canyon 7 . 1 .. _ 7 117,7. {3 \\ ‘1 Brea‘1ks in tar‘strike Nlrregular hairline crack 7 17 1‘ _ \,7 \ ‘I7 1’ .7 7 , . . ‘x‘ | N.I 52° W. No offset \7 7. , \ , “7“. chem fractures , 1 ‘ ‘ 7‘ 1 . Geodetic network across fault \\ * of whiteline; no _ “ . t . 1‘ x, ‘ ‘ “ 97in. if”? 20"” 0§m “”13 ‘ ‘ r ‘1 i <11 ‘ \ 33- en meme-7.13777 7 .77 : 1, _ . '7’ 7 \\ 1 _ cr cksin adjacent“~.77 7 7_ L177 7 1 ‘ 7\ . 0.5m wide , .7 Sky/17$ , 7‘ Rex/.7 4 .\ ' \\ \ l7‘l gr und77 7 7 7 ,7 4 I , l ‘ ' ‘ - ‘ \ . 1 / 1,... . 1 , . \_ 7 1‘1 .5 » . '7Shtrr105m “3,71%;7 .. i_fi..- wwj, (AW 77\\ \7 7 . 7 \ CIT creepmeter \\ en echelon .1... ° 1;} THEWAMAWW’ ,1, \7 \ l " T 1 ~ ~ '9 8R; across zo‘n .Ewid \ . _,. I. ' \ "‘J‘“ \ , atbaseofO. mscéfime 03TMAR7‘5, \ , _ 7 \\ “1 ”In _J‘\/ TVH\77\ 777:\77\ Thermal Canyon 1965; Mecca 1955; 7 7 7 7 7 77 7 7 Mortmar 1958; and Durrnld and . Zone 6m w‘ide- “391:9” . 1 FreaksflEm longelanre [/4117 ‘ 1 2 . 1 a1 1': : 1 , ‘ .1 :— .' ~ ' . -1: , "*11’4491er'3/‘Lwi 1.“...7‘, ‘ 57,-“ "i \‘1‘\ , ‘ \x ‘ ‘. § 132‘ Salton, 1956 7 ‘® 7 7 in side buulch dip 952 ) 7/777, , ‘6-25R' lfractutefi gem/1339“ .7}, 7 777 Nw 177:” 7 .7 7 7 , 7 77 .7 77 7 7 7 7 / 7777,,,-\7777’,7"..- -1 1 71 1 SAL TO/V 7 5 15,71 ~11 :1 11’ H.,\ 1.7 77 7:7)th :roKIVEn Qzuge 7777 .v‘ ,f .1 . W ‘73.:713 1n1 zoherjaém‘deetJenechelon \ “7—1, {‘6 ‘ . , 7 . . 77 . _7 1 ,_ . 7 77,. 7\ 17 3687::3757 " 5'0931 l L \7 ,- ‘11.; a; 6:5 in f 1 in Q 12;” 7fracturgs/gtrlkeN 5-15" ' L '1 1 1 ‘1 , 77/11 1 . . 77, 1 7 <1 7 7 70177:???“ 7771777171177 0'19: x. ,\ NJ ./ 1ngth 7 c: aw racturc‘es‘ wept 1,. , W . . . 7 7; 1 7.; , . ‘1 77 ‘ :15 ‘ é {‘5 led (arm w? ‘ 7722777777 5 ‘ Er <36R en echelon breaks 1 u‘J1 ' 7“ 1‘ \\\ ‘17", ‘ ‘ 7i 7 7 3- 3R; erfechelon ftac ures “"9 77 777 77 . 77:77? 2 7777 . 7. .. 7 7 7 7 0 115- 2m long' in narrow/ ,7 1 \ 1.77 7 »._.77777 \ f 7 _7 . . ‘ .3 111510me n1 ‘ \ ' ‘ l7 7Col7lalpsed fracture 1 ‘ \137. 15‘ l 17 , \ 1 H. ‘1‘ $1" I‘ I, /" \\ 1 \, r"“ l \\ ‘ 7 1‘ , tac‘tures ‘l-1.5m ‘\ ‘13: \\, 1 \ .1" ‘\ »1‘ K 1 ‘ ‘ , ‘ ‘ ‘\ “\7 /| ‘gfiobsmuredby san ‘:‘\ \\\ ‘ . ~ I. 1 a 1 1. .. bteaki‘im mi “g7 /, ""17” -~ — 1 7—, . I17 ’ 7i 5 7‘L lZon‘e‘ 10m w'd e;.c0ntain \Oo 7 \ \1\ \ ‘ : in zone 0.3m wide, 1‘ l I, 17 10-12 subp allel fractures, “613(6— 2mm x 25mm crac I. T 19m up on fideh'“ "7 ‘ “1 I ,/ zone 10m NEE scarp in ‘7 .5, sand ohliterated 5 68 ‘1 1 7 \\ \77 17 17 \ A (I *L’: \ l \ .I ‘17 ‘1 1 ‘1gra7y soil ‘“ “ «« -25\\7l . \ \ \ ~ . i 1L, / “I l 1“ ‘ (I ‘ \‘M‘x rd‘ \\ ‘ \\ , _ ~ 1 ~~~~~~~ H , 1 1 l .' ‘ 1 \.\ 1 1 1 O ,,,,, 1 O 2 . \\ ' .1 \\ .._7 I, 17 . <\~\(, 3R; intermittent fractutes l . 1 \ l \‘\\ 1‘ ' “1.7 fi\, 0.3-1m long; zone 0.5-1m‘\\ ‘ \ \, i‘ ‘ l 1,7 M 1- \‘W “ ~ , wide; generally right _ \w7 \ ‘l x‘ l ‘31 7 ‘ lateral but some fractures ‘ 7.\\7 . “~ 77 7 1 ‘1 I ‘ “\ 1 3‘R;\ll71airlirie fractures \T‘Tinear parallel to zone; ‘\L,77 Ax 1: 1 \ 7 \7 I7 7 ,7 in7 z‘one 0.5m wide may fractures obscured 7 7 g 73..., \ LIN , “1,7 ‘17 l ‘ 77 ,l‘ “Yuk. 7 7 \ by sand 519/758 17 7 ‘\\77 x 7 17 \77 77 ‘1‘ 7 [.1277 _.\7\7 x ‘1 Ii 1“. L ‘1 ll 1 k 1 new \ \\ : \ 6-13R; e?! echelon fractures :1 7‘7, 7 .1777 M H..\ 17 ‘ \ ”"xin zone 0:3; mwide ' ‘ ”""' ’’’’ 5"” Fl """"""" #7 . ‘ ’ ‘ 'T. I ’f’ E" R ’ f ‘ ‘ ,7 “l .\ ‘K, “\7 ~.,_ I /7; l7 / l /7// \f \77\ ‘NG-‘13Rren echelo‘n‘fi'act‘ures » 1 i I Lv.7 ' ET ‘ ‘\ r \7 7 A, , '1 3.- ‘lm long In zone 1n717 wide;_77 , __M. ,.,_::-~. Im/J / if: 7 ' CIT BePtnam aetures stf‘flze N. 105115" W. ”/ L //‘1/’ 7.\7,j‘1 \\ l H“ / H4"? \ \"1 ' V” ’ 2R \ 5R3 ‘en echelon fractures , 7 7 77 77 ‘77 777777 7 7 (7/7 _ 7 <7 . r 7 1.7, , .. , , in 77z70§e 0. 5m wide at NE edge ‘ . 1 7 ' ‘ x" I ~74". ‘ ‘ ,7 “in. ‘ ‘ ‘ 1: - ‘M - '1 , ‘ 1 H. \ of re gouge ‘ ‘ \. 1 ' . ' ~ ~ ' " ‘ n " ‘ ‘ a n 1 No fractures observed 7 .. 7 _ 7 : 7 7 7 7 7 . ,, ~. , \ , _7 long fault trace S A A N D R E A S F A U L I . 1 . 1~ .1 . . ~ . 1 ,_ . . OIdsIcarp.both , . . . , 7 , . _ , _ 7 a , N l 1 ‘ ' '1 - . z ,, S '5 1» 7 , x, ‘ sides dissected‘":\ 1 1 w, 1 . _ \ 1 .1 7 7 Ml . . , = 7, 5/9/68 7‘ \‘P‘K 1 2 . 1 ‘ ‘, . . 7 ,, .7 777 \777 ‘11:.77777 “ ,77 ‘ \77 \1 , . “1.77777 7 ’ 777— 77 ‘1 ‘ __77 7 7 ‘ 77 ‘ - : 7/ 1 7 . : 7 5' .: . ‘1 .120» - 1 1 ’ v I Q71 L11 #1 \ ,7 7 7 77/. 7 7 77777777 7 777 77 .7 l“(‘1“'.lll‘,lli ' 7 77 7 7 7 7 7 77 7 :f7 7 7? 717777.997 77 Dy/iM/Q 7,75 " 1‘ ‘ JOINS STRIP E 11 '31 Breaks mapped by Max Wyss and R. E. Wallace, May 8—10,1968 MAP SHOWING BREAKS ALONG THE SUPERSTITION HILLS AND SAN ANDREAS FAULTS FORMED BY DISPLACEMENT TRIGGERED BY THE BORREGO MOUNTAIN EARTHQUAKE OF APRIL 9, 1968 476—246 0 - 72 (In pocket) No. 2 UNITED STATES DEPARTMENT OF THE INTERIOR GEOLOGICAL SURVEY R.7E, 33°15’ 316. 15 0 Borr : 900 V L42.“ Badlaéc 1.. 4 116°15’ Base from U.S. Geological Survey 1:62,500 R. 7 E. Plaster City, 1957; Carrizo Mtn. and Borrego, 1959; Borrego Mtn.,1960; and 1:24,000 Harpers Well, 1956 3 L, 67 «'1 \\ \ Pliocene EXPLANATION § 3 Qa S Q . m Alluv1um 33 Older alluvium 8 o f; S 5.1 Ocotillo Conglomerate of Dibblee (1954) 2 a L) E E Undifferentiated sediments °~ Includes Imperial Formation, Palm Spring Forma- E tion, and Borrego Formation of Tarbet and § Holman (I 944) Crystalline rocks Contact Anticlinc Syncline Fold axes Dotted where concealed 28 _|__ Strike and dip of beds Highest shoreline of Holocene Lake Cahuila Y ' ! k “ .uxuueuguncgv I e} El 08 I, , K4 Eli/LEO, éfliwq, / flaunt]; Well /1 '1 6 D 00’ "" “-IPI’Yl 5' ,, W S . ' ‘- SCALE 1:62 500 1 1/2 o 1 2 3 MILES l-—l 9—4 l-—l l—l l—l .7 j F———i 1 .5 o 1 2 3 KlLOMETERS ELL-«HHH; CONTOUR INTERVALS IO, 40, AND 80 FEET DOTTED LINES REPRESENT 20~ AND 40-FOOT CONTOURS DATUM IS MEAN SEA LEVEL GENERALIZED GEOLOGIC MAP OF THE AREA NEAR THE SEGMENT OF THE COYOTE THAT RUPTURED DURING THE BORREGO MOUNTAIN EARTHQUAKE CREEK FAULT W QUATERNARY TERTIARY PRE— TERTIARY AND QUATERNARY PROFESSIONAL PAPER 787 PLATE 5 Geology by R.V. Sharp and MM. Clark, 1969 476-246 0 — 72 (In pocket) No. 3 UNITED STATES DEPARTMENT OF THE INTERIOR PROFESSIONAL PAPER 78 7 GEOLOGICAL SURVEY PLATE 4 ll7’30’ 15’ ~o . » ., 15’ a" 43 I 11/300 R14 5- 30’ R, 5 E. R. 6 E. 15/ R. 7 E. R. 8 E R. 9 E. 115°oo' R. In E. R. II E, 45’ R12 5- R- 13 E» 3_ 30 __ I ,. “ a I « F r - » I 3 i . . ’ I _4/1 NEW . . , s ' a5347 ‘ Engineering geology effects of the Borrego Mountain earthquake, April 9, 1968 3141-13351 Camn ammo M jag ‘\ M33 fl ‘ ”my 1 Classification ofeffects modified after Richter (1958, p. 81) , ‘ ISL. , w l Palm 0959“ \., ‘ T' 5 5' 5 '13 r337 ‘1 6'] I ‘ Palm DOSEI'K3 ’f‘ 1 3 1 13/1 s3, 1 3/.1 3 3 . 3 , . Tran51ent effects Permanent effects 1 v " " raw “ WW” ""‘~‘~» \ 33”,-1 ER 1 I ~ “i JV. ,/ 3 1‘ ts I a” ‘ I 2o I“ of " 3 Ef ec Map . Map Direct or primary Map Indirect or secondary I {why \3" 33333§< ‘ EXPLANATION Symbol Symbol (due to fault movement) Symbol (due to passage of seismic waves) l {31 IT‘V,’ 3 mm mm, 1 R ‘l E/TEALDTAo—ufifizl / R E 3. R *1 E3 Predominant direction of horizontal motion and / Fractures and offsets Slope failures 1 ALM VIEW mm ‘2 l A -~.. , ’ND’O MOUNIA‘N ‘ . . . . . / / . 3 I \ .3 F 3 0 75 duration of shaking, in minutes. Perceptible , Or Rockfalls ; - 3 / 3, / 3 3: ' shaking experienced over entire area of plate 4. Pressure ridges .5 Soilfalls 8 Ram}. : . 3 1 fife?“ \\ \\ Chlriaco 5mm 1 . ' ‘ . ' /~ r t, . v' 1 . . ‘ It.» / . . I3 21 Visible surface waves Developed extensively along fracture zones, see A ROCkShdes , . . _ . ) "WW I / . 1 131,15”; 41; \ Y Quaternary dune sand, alluVium, and nonmarine Clark, this volume A Slumps and inCipient slump cracks (fissures generated in // \K Q" S -_ I .- /' . 33, Y C‘Q, ._ terrace deposits . . . . . "' 3 , ’l . =’ i ,1), r ~ . , , / . Sca ps . . headwall regions of evolvmg slides) in natural materials \w ,/ 3 I ASBES‘TOSXAOUNFAW/ 3I/ . 1 , '"i‘f‘d' (4y Unindurated to poorly indurated river, stream, and fan I Developed along fracture zones that comcrde AF Slumps in artific1al fill \ —~~-’—'— :1qu :5 ~ _,./ . deposits Includes beach deposits delta deposits and Terrain With preeXisting scarps; see Clark, this volume AFl Slumps in artifiCial fill along shoreline (canal, river, or lake) Range, 3‘ 33;: ii“ Mme ; 3 \\ 'r‘fl’ey ./ 33/ filled land along coast and scattered to abundant eolian E Shattered crusts; natural depOSits ‘ £3?“th ”me deposits in Imperial Valley region 21 Shattered crusts; artifiCial deposits x, Compaction features 3 mam Sarina R Cracked surficial deposits ., l- 1/ / I O R F 1t (1 f. . l d .t Wei; Huh 1/1 .4 3, 33 _...L F au 6 sur 1c1a eposr s S NTA R0 ~ 33” «,1 ,p/ . Rs Sand bells mm... 3 3 O D Displaced pebbles or cobbles ‘ ll 7 S 1 \ MARTINEZ“ / C . . . X Collapsed animal burrows 33.2'RE5ERVAI'ON i\ 1 \3 ‘1 ~ "“1” Pleistocene marine terrace depostts ‘ ‘ \ , Sandstone, siltstone, and conglomerate veneering older rocks . ' . . . . in coastal re ion Deformed by bending, twrsting, or rending Damaged by shaking: » ' _ g l Roads V Cracked highway :- ““““““““ " ~11r--~-——»«-~I~-m>-~—3l;)~- 3 2 Pipelines PL Broken pipeline -. 1» ~ 3. l 30’ C1 Collapsed or toppled chimney 3 3 1 30' _ C2 Cracked or partly collapsed ceilings _{E 3,._.,.L, __ ._ 1 Walls (masonry) HFvalion 1234 feet below 1 g 1 .' W, Cracked __._Isea IevleLMgv 1959 ,_.., j TORRES l v \ Chiefly Quaternary lake deposits . ' W2 Collapsed 1 - 1 1 i 1 MART‘NEZ I 9‘ Claystone sand and beach gravel. Confined to Imperial SALTONISEAI INDIAN . o . » ; Walls (wood frame) m » a- L...__ ’1/ Valle region ' - 1 - l “ RFSERVATION “- y M d W3 Separated at joms With annexes i l . . 1 J $351313”: W4 Cracked plaster or wallboard seam - ’ ; K Concrete slabs, liners, or pipes shifted or cracked l j Bridges / />zMz ‘I . Bl Cracked or spalled piers ///// i 32 Concrete deck cracked; abutments cracked or displaced; fill ~ W ~ _.__ m --"- - ” ~ . buckled or settled at aprons .3 Pleistocene to Upper Cretaceous volcanic . 3 A Anchored objects (housetrailers, washbasins) pulled or snapped W" l 3 and nonmarine sedimentary rocks 1 from fittings 01' foundations m __.g§ TORRES Unmetamorphosed but generally well-indurated sandstones, P Pounding at base 0f structure m :— MARHNEZ 1 siltstones, conglomerates, and fanglomerates. Includes p3y- . F1 Occasional cracked window gm ‘NDIAN 3 LN") roclastzc rocks and lavas along western edge of Imperial Sc Swimming pool cracked \ IRESERVAT‘QN NAVAL l \ Valley ‘ . L . I —~~ ._ 1, l V /3/\/‘\/\ I RESERVAHON \ (NV—WI" . . s . - \\ ‘ / \ / \ , \ I \ \ \ - I Q Anchored objects (chandeliers, flagpoles, trees) S Automobiles or trailers shifted laterally l ’1 “6’41. J ’ ‘/\\I‘,‘:‘F/‘\ ‘/\ Q zf/ \7': . ' - , P. .2 A / Lg . . I 0131““ H2 Heavy objects overturned --—:~},;L_v: .l,£.I_../:_._. ’ /l \ \ \_ \ .................. I -‘ L1 Light objects shifted, projected, 0T overturned .-/ I’ " I ' \3‘ l\ I37I’:‘, Mesozoic and Paleozoic igneous and metamorphic rocks . I g . \ 1\ I“ . 3 3 1 34133543103] 3 3 "I / Elevation 234 feet below sea May 1959 22 Seiche in swimming pool or storage tank M Automobiles damaged by falling rock ,Jfi ‘\_N ,3 ,3 , » .3 / ‘1 ‘ — . , ,,,,,,,,,,,,,,,,,,,,,,, ....__._. \ . J Experienced difficulty in walking /’ LOS cox/39758 IND ' ‘ ' k G} Miscellaneous A1 Sand in water following shock I" I.:/IREI 'R‘ZATION \I . A2 Increased flow from well following shock 1 XI“ B" . k Prehminary instrumental epicenter . ..3 ' -' ‘\ 4 .J i '9: at 3/ \\ 1Classification and terminology modified after Varnes (1958) 9 ./ ‘-‘ L j , \ .- ' \_/ \ — r ,‘ \ ,\ 15, R.5W RAW. R.3W. sz. RIW. RIE. RZE. R35 7— \\/\l>\/,\//’ ‘ .-.- 15, ,4 / . " "n“ - /‘ / \ /l \/ ' \ MOUNY 1 M//\ 3 /3\_‘\/*¢\\_I\ 3 3 .................. -. ., .-_. ~ ; ADOBE FLATS \/ \ _ / ‘ / (“X‘— ' ‘ ‘ 2 . p45 ,, . . I I ,I , Ex * . SALTON SEA TEST 8 E c b .1» u \ .1 A, ./ a. . W”? C’lifl’lw” . I: weaves? . 7:9 2' “W 2 I '-\ , ' 3 j / / I’ ~. ,1 E’s/0:41;: 3‘ 1,57 5320’ MUIIq'I IslandD \\ 3 SAN MARCOWOUNT . “”97 0"”?! "\ 5m 3 Q l Ilim EL moum- ’/;’IL/1\/I~\‘ ‘ c/(é Lwa‘ SALTON SEA I.._,.I \ \ r i v ‘ . ‘ ‘ _ . _ \ {>767 '\ ”I P/ ”g /\,'= 3 _l H 14.18% cl ,3 \" V ’ NATIONAL I _ Ocean‘side \ 3 O: “W ‘1 1/4 GTQW‘I l ,._. 3/“ ’ WILDLIFE «.4 sand ,3 . /7faEZm SERVO/FLAT l \ \r. .Lmr.._.,_:‘_,__3 REFUGE \\ ‘\ ‘ dam 2‘ . ./ 1......\ l. ’1’” %MND’E I -\ x \ s c. {K ’ , was. \. South Oceailside , 8E; “MINE/1.3 : Mgr: G" \ -\\ e I . f 1 . \‘x ‘x \ \ 131y ' \_\ ‘\\Ca{lsl\:had "1/ . \3 \ \fihlp . GU 3 l \\\- \ \‘ ‘ L, IS 1‘ W It. I? \ I / ‘ \ .3 \33 3,.3357 \ 1 _I’ \g a ’I)’\>_\\,<\'/\ / thlfigrge Qua“? qua Hedionda a , 't' V g! [E ’2 " "l \:\\ 1‘ £7: / /‘ \ s. ; all I 0 El \ ”A 9 urn \ n I ‘ 1/! Im ”'1 1g” 5 \\I‘/L‘f‘\/I’I<7//= 1 I2 5 I ~ w»: u ,- =~ a :i ' \ I . / . . . , - I ’mud mud ( DOUBLE PEAK ; , / E w“:11! 15 \_\/\\’\ f, \\ SALTON 3 I ,/ _r/ l / Y 2 5 I ,3 3333333333 33 ,A g; m . _x \Q NV I“: , NATIONAL. . ' “ i E/ ‘ /“ 1 71V; 4 /.~/.:‘>Igg;,, ’ L" We / \ \ ‘ \ ~— , ‘ ‘ Pclroglyphs Qt- Cem‘ Y?\I\1/ l\ " L7: :1:l7\'//— "/ \C‘1\"\/' \/\/\\ l ‘ \—/\ /)\ 3/._\l 1 \'/ ‘ / I ; \ (I \<: ‘ mud Pass ’1‘ / \ / \ / ., 1‘ f 2 ~ _ ,\/\\ \ . " \ / PCT/1‘31 \\ 111/ \Ch ’ / \LCLJ. . ~ ll" , / ‘ Ct- 'Al I LV/I AI“ \\_//11‘ v 1 011%,13SM“ ’3 r'- fiaxl/yJI’d/Xym mud ’ / "1‘, :‘ Lay/aim) "'53 /I\I‘\,1\\/ 5’50; \ l\i‘€l5“:lWPI-l\ AL, 33°00 /‘”3\ \I\/'\‘ / \T/ \1 \— 1/)?751’ \/\~"\ ‘ H \ N / / \ \ / May / ' ’2‘ /\/fS/\,‘"}l\y_w“_\l\\/\(74)W‘ ‘ \ \71:\‘ “\l'“>‘/\’\"‘/\”\/\“L/—— mayo 2: l\ I?! .‘_.I_\.l,\ A 3/01! IOIIWTDW f) I30: /3’\3/ 3; SRanch ~i 3/ CUYAM A \ A -1 / \3 T..l3 5. Danger area ll? Val V W1 NORTH PEAK Y" ‘ //3 "\\m 33"OO'I1-1 a, e .. L, (3 3 < 3/ 3§9m191r 1 \\r I ‘. Solan‘a Beéch a l 3‘ :3 “£13 ‘ ‘ 3). ‘\ ‘Kl " f /l PE :w s. . ‘ l : m3ud33 I 1 \3 1 ‘e p .«’ 33, 433 13% SI «3,. Wis/22;“, T53“? \\ A 3322:?" \.._l D'AN RE 0‘30”}95 ORIFLAMME . ; 5"" 1 13 “5“. IL/”' . / _ ,._._._n \\/I‘\ feel» 3333 I \ __ -l--\ - mouwmw i 1“: ° wakfix ‘ /\‘ : "ma 1 1 ~ Randi ~ ”M < 3 » ‘- ‘ \ ‘\ l ---M._ —\ / TO . " 3 , 1 . » ROCKY ‘ » 1 a. 14 S ' /1, c , .3 (\o _‘ ,Pgfiyw l 1 M/ 0 ”('5 /7 Vallqfiito C l K m. 3‘99 ‘ fl 1 "/‘ l\\ . . .. Ioee'/,’- . Ale 1. v ’ T m s AS737 l ‘ 1 ~. 1 7-1 1/ v“ x 9/ : I... . ~. ”CW” Ca: \ I \\e~l .~-"‘*+‘ ./ ‘ " \w\: an ca Ga rd .1 .1 \‘1 1,, 1 . Q‘ \ ’ . D\" . . Calienle ‘ " \ 1 5/ 1 374 l mud 62ml” .foo’rMarIMesa ’ f . / > StatIIm H ’ 1 1‘ \‘GAIéNEi MOUPITAW 6“ /\\“9"‘/\ Agda Gamma SD” 3215 1 / \ “\3 530' l J" > . r _L 4-1 " v . 552x . ‘ I ‘ \ .1 ‘~-- « w.» 2 l . -. I _ -' —- ,_._.- ' 3 mud 3/ Cem- J x,» 1 "I, C EVE N l Lookout fiver r.\ 31 , '\3FQ33565933.1 _ 3390 ““4../ . \ \3 3. . «20/ 3‘ '1 3 ‘T3. 14 S. I ;’ CARMEL 3w” .. ~' . 37;. cm \’ A 01% Alf—Q R‘l ./ )8 l . ~ . 31 1> CARRIZO J3 3 DANGER AREA :33 QWMOUNWN/ZW I _ 3 3_ 3/1 A . 3, />/ Lifer-(Ugo (‘L . l g l ABORREGO: ESERT STATE PARK '7; z; - ,s @ARRIZO33 I U 8 NAVAL I .3 "f/s‘r /I/ I3; 63: a I 4/ ,\ I timer 0‘“ {A 3f I ””\. E 1% IMPagT AREA I to I 333I 3 3 333 2 Go (43939 " '1/\1’ \ ~171._1__./ 13 . 3:- \ 3 / \3 1 33.313 Q33. 3».» . a, JIV'PACI I l q 1 IMPACT RANGE 3/ g a . : - m- iiiii . . «:3: . . . ‘ : E «w ~ ' . ‘v ' If ,» _qefiv I f 0 RAMA-1911‘” Swizz“ 3333 /(/§’ /’ ~ XSX may .3 0 1o ~_ 3n 1 I 3 ; 353 I (SAND HILLS) I \Tungsteb Mme 3 1. : '\. 1. ‘) 3:393 l 2: \AK , 1 3W3 AREA I ' E3 1 mud \ ”T" w “r: l M" 3’ “a ‘ 3;. ....... _ \(j; MONaMENr. 3 r" :‘1 ’ z 1 \éi/{itfig Lagu'm flake/.1," \1 g 3' ‘ ‘ _. n"“‘_ 1 mug: /" " 13“." , 2‘ ‘\\ I‘Lagun‘a Radai Station -~"K,’. ”Ir—K”? 3W4“ . .J ‘ I 33/ “‘2’ \\\§\a 1 {f- .. Q.§FT3_¢/ L‘Urrlk‘o‘ 63:23,]: .. f :2, [2 L Y C3 L -33., 3 H I — A if ‘ xx 23‘” /\’1’ I ll! F :17, “"fa‘éyl ' ff 1‘ ‘ /“‘ $0» 1 / / T a /- , c. g (13 A 3 [In ”a in M. I . _______ _; l ~~ ‘ . 3/ I l,AdOL£23¥ ‘0 “535/ . ,:’/“T~.“« ‘9 .l CUYAF’AIPEf7’l// 3' ./~-*/§ firings/4r ../ / 11—3-71" ‘‘‘‘ " “ "‘ w ' (mm, L“? r ""’ , // Descar Guatay For Cami/(7‘ ‘L/ T art/VSM 14, ‘ “I H I . ~ / ROLM Valley I T 15 S 33 I“ Kali) / ...,_- I 885 (2/ 331mm Ranch 1 1 1 lND‘l‘ REIS l \ T4” In Q”Cree/l 4'1 10 O I 1 Cami! ' Sanamnum . . ' ' mud II mu 2 ‘13 .1... . , 3 ‘ 3 3 x I 33. 33 .- “£3 , 3. l’ 3. 3 3 3 3 3. WHSI . m3 ATAY MOUNTAW flat“ Warns Ranch-{VA 33333131237 \_3\ j I l O\ SOMBRERO PEAK $13, BM 833389313333 Pa” O f 5 CARRl/O ( Hollwlle AuxIIIaIy \ 33 , 3, I l x 3293 A 310N911] "(j/IA); 1. \LI 1 ‘1} l 3 83333 $31 Ranger \\ ,5 ,,IOUN;,’IN . ‘l’ 30’ I ’ I I A ~' EEPH AD '01] TA ., . ’ ".1" ,, I/ I I' ate”! ~/. 1 \333 J 5i; i ’1" N W 11 LL73} 1 / \ ’3’ r» 1 \ I44 0 0 Lorain l mud I; l mud l w ‘ I _.__ '._,__ w... ,,,,, ' ‘ ' ' x’ ,9 hell ' .1 ,3 ,_.-~" \ ‘1 A / o ~I ’(flfimig Ranch ,3 '3 In \ "T '3— ll 5 S \jmld Kfiip\\ 831 f on \W“ Oil \\408 ) 15g] 1‘ I‘ lN ~13? \3_3'Ijlll‘a[llll§; If ‘3, ‘\ T C'VW __ ‘3' . _ . l g Ni. \‘\ ss‘lnl —- 67(7):!904 Tiff/"rift" J ONG VAhf‘; ‘ 0‘0 W\F S 1."... fl‘ " ' / mud .1 ~ I -I /\,/‘ D ~ pm [JIqu‘d 1/, 3 I I (3 ,3 I \/1/‘ ehesa 83' W ’ mud { I; \. \3 School ( . I I \. V. (y 5 1 Mission " .\ ‘I 'd I. \ SAND DUNES \\ nm \33\ 3 3 L \\ 1 , . “a ‘ . 521:1", Ilieteryf‘;§~._ "\ [I ’3 mm" “/91!" , ' T‘ ‘6 Si 32°45?”- ‘x . ”w” , ‘ »J 134/39 I a.» M 1 //’ dunes ’ "148 7;.4 32545‘ .3 I , / ,...__ / / 3 \ ~._3 1 I , 3 . \3 m1 1 Flats 3‘ 13 1 3 3DIESERT £3 . m I Kel/p , . . . * ,1 S l \ sand ‘I, ,3 l V4 1 3r \- q; ’ \ ‘ x l 1 , ’ . . . . -.3 .3 ud31 , I 1 < , ‘ _ .3 x. mud :14 R1ckI’ MBHW Ram“ M‘"”°J 33’3 ”\R / \ _. mmmmm 41‘ \\ \313 \ I“: \l 7* l ”h“ 565'” T 151/2 5 Hm" I . .. . 5 a , e. .4 , 3. .- mud IUS NAVAL. RESER 33 “IONS v 5:sz "2433335 \ (f3 3mm . \ 3X7“ II\?§‘§~\33 .3 W mg l 63‘ _., g; = ,3 ., . A ‘ r'“““““ . > . - ,3 ‘ .' a, 3., -- I ( IN“ 3738 \l. Barrett Lake 33%,le els I, 1 la: 33 ll 1 “3c 3 ‘3 W3 " 1-3,, _______ “_____,__g _,__._..... - M l 7‘ I '\ T I? 8 ~{’ _3eru)lr 415/ l 135\ “’5‘ lbilgi’l Aczes Ranrh 1 SIGNAEMTTT' ' 7 er """"" .——- - - ”"" I »,_ , 1 333 1—1-1 395; / I.._3_zsw. 3",. L. ,\33r333333c3;3.3.33_33;v l PARAGHUTEI I “”3 _ 1 ‘3 , ., I” . ,3 r , ~ ' Is ma 33333 .‘ 13333333 3 ”8333 _3 CuflouwoOd 019° pmfir l c1514 P0, !33 . \\ 3 n , DROP AREA 98 c .. .. l I 953 53"“ Breakwater Illeservali‘ori‘ ~ / ' “NW-AN RQS‘ . “1 Tale eksd-flflcdr‘fil-I ' . . 3 ‘3 3/ 13 133mm 35m _33 _________ [3 l I \r” l .~ 33 (333I33333333 1 Locations of tectonic fractures from plate 1. 3/ \. i \3 \.3 3 i 2%} ":1 1; 1 €\ 1 5 V m???” .3 Geology generalized from Strand (1962), Rogers (1965), (\.,/ 1. 3/ ,' 3‘ . 2.13}. Jim Kelsi: ;\. ”2 1 \\ m I SCI 3‘ 33-03:) I - and Jennings (1967). Data on earthquake effects ‘x I 3‘” \ NA . ‘ <° : , Q ' ‘ ‘1 com iled b RD. 3 tle‘ n . . "mm sand ' \ Sawdgtfi Channel. J's/San Déégo C; " ”'21“ \ ‘3 Spring '2 L..5" Q ‘5 a n E If.” —-— r“, '- p y C s a d T L YOUd “\V ‘ \3 Intuit 3/ \I 39% feetultpril 1953 (33" 4 3 \ EDF‘DQP ngl R 4 E \ “ I \J ' A' Davies R ‘0 E \ I. ~. 1 x p ' I. 3m ~ ’59 ROCKY ‘ 2” .im . .. v . — —“ “i l “x I “Wreckage of Eggnge‘go‘k L ‘ (\ 1 r W} I “A???” MOUILg _' G‘\- A i ‘8 b alley“ ' ”T. — - M \\ ‘ ‘3 mid“. Sand 1 ‘\ DIEGO may ' \ 3 0/" (A, 330 3 u 1 3‘ W 3333 '3': M \\ \3 \I 3 \, ._\ ‘3 l» \3 BA YES" 3 MR \5 K1 1 ’ 3. / U I - I. \ A, I f “\. x I ‘ d ‘ ' I ‘ ‘I ' \33 I33 8 .g‘ 3 ,1 1mm :33; sea 33% 33 ,3 M'R» K (I 3) 3K//v££\ ‘/ I460 , Cameo \ \ mud l 1 I RESTR’CTEQ AREA ‘ I I; “ - ' AJ/ * 214 mm: rECArE " ”x ., 1 ‘\ (rind ‘ " ”-1 ,, ‘w - \ 4' " 7750 OTAY M 2366 ' K .OR \. \_ K. - e ‘I I ~.«. ‘ 1 B ‘11 3/68‘ 1 \ \ CAI—1F 1 F: ‘ ' ‘\ “ ,t I gravel “\. \; mpgfif‘amrkyeal: Canyc" .357 . \ l 1 f3053‘_\__ M ~ " IA \ ~ / ‘ ‘\ . L I/ , , \ ‘_ a, "Av—ur- \ \\ ’ I, slum sand shells). /' \ 3:!” 1%“ S A Y S D O , 1TECATE PEAK ""4” 3 A CALIFORN l \ \. a x ~- A III I , _..... 1A \\ \\\\\ \ \3 33 L. .._.. . we. .,33I g-.. .. e . ti 3/ ell) \vjt'tM’O/U 3T I N 5 L/' doggy/:qungggflf “w”, a ‘ \\ ‘\ \\ m1 ‘\ 1 K6“) “F/ 3 S\TATE$ 7-HT" :1"). 1"“ T TM I 1 \ $2.” g 1. Nl‘AS 331x f UN, 59W5flJ—Eé’“ “ SCALE 1:250 000 E 1 \fi. \_ " ‘z, Kelp“? ’ a «Y I — - W‘“ " " 1 I00 2 I a mud “\ DANGER AREA \, » MEX 5 o 5 10 15 MILES m ‘»I [I ‘I sand I find I 80R R Fig} I—l l ' l—l 1,7 ' I D \ a 1 . I 1 9: I3 .33 NAVAL R SERVAT ‘- .: ., 5 o 5 10 15 KlLOMETERS 11 7 “ 30' 1 b r--l l—l H l—-_—l 4| APPROXIMATE MEAN DECLINATION, 1972 Base from U.S. Geological Survey 1:250,000 San Diego, 1958; Santa3Ana, 1965; Salton Sea, 1959; El Centro, 1964 MAP SHOWING GENERALIZED GEOLOGY AND DISTRIBUTION OF EFFECTS OF THE BORREGO MOUNTAIN EARTHQUAKE BETWEEN CORONADO AND GLAMIS, CALIFORNIA 476—246 0 - 72 (In pocket) No. 4 UNITED STATES DEPARTMENT OF THE INTERIOR PROFESSIONAL PAPER 787 GEOLOGICAL SURVEY PLATE 5 Engineering geology effects of the Borrego Mountain earthquake, April 9, 1968 Classification of effects modified after Richter (I 958, p. 81) _ Transient effects Permanent effects 116°07’30" 1 1905’ Effects Map Map Direct or primary Map Indirect or secondary Symbol Symbol (due to fault movement) Symbol (due to passage of seismic waves) , Predominant direction of horizontal motion and \\\\\ Fractures and offsets Slope failures1 .......... , finger“ I," WWW _W X duration of shaking, in minutes. Perceptible .r Rockfalls Epicenter 0.75 shaking experienced over entire area of plate 4. Pressure ridges .5 Soilfalls , _ Developed extensively along fracture zones; see: A Rock slides ' C 1 Z1 V1s1ble surface waves Clark, this volume A Slumps and incipient slump cracks (fissures generated 1n “ headwall regions of evolving slides) in natural materials i Scarps _ _ ‘F Slumps in artificial fill l Terrain Developed 31,0135; fracture zones that comcrde AF] Slumps in artificial fill along shoreline (canal, river, or lake) 1 With preex1st1ng scarps; see Clark, this volume 2 Shattered crusts; natural deposits " E 1 Shattered crusts; artificial deposits ‘ Compaction features ! R Cracked surficial deposits RF Faulted surficial deposits , ‘ Rs Sand boils W047”? Welltirii ‘\ l D Displaced pebbles or cobbles (my) 4 _ » ¥ ' X Collapsed animal burrows i Deformed by bending, twisting, or rending Damaged by shaking: l l Roads V Cracked highway i 2 Pipelines PL Broken pipeline 4 Cl Collapsed or toppled chimney i C2 Cracked or partly collapsed ceilings 1 Walls (masonry) , W1 Cracked W2 Collapsed I Walls (wood frame) xx; W3 Separated at joins with annexes Manmade W4 Cracked plaster or wallboard seam i structures K Concrete slabs, liners, or pipes shifted or cracked T H S R Surface ruptures ; Bridges p 1 April 9,1968 I B1 Cracked or spalled piers \\ B2 Concrete deck cracked; abutments cracked or displaced; fill (km -. ‘ buckled or settled at aprons " A Anchored objects (housetrailers, washbasins) pulled or snapped from fittings or foundations P Pounding at base of structure F1 Occasional cracked window SC Swimming pool cracked Q Anchored objects (chandeliers, flagpoles, trees) S Automobiles or trailers shifted laterally Loose swung or whipped Hl Heavy objects (furniture, water heaters, safes) shifted or projected objects H2 Heavy objects overturned L1 Light objects shifted, projected, or overturned 22 Seiche in swimming pool or storage tank M Automobiles damaged by falling rock _ J Experienced difficulty in walking Miscellaneous A1 Sand in water following shock A2 Increased flow from well following shock 116°02’30” rs LLO,’ Mitt GUN??? AIWORT I,” E / -_ \.1 , ‘ (x r~ i ‘ngfuer iemtos . st l‘ I'Marker, \ era-~22- , _ _ ,, \ w y, , _,/ \ t, ‘ , -__ / ,’ .> ; .» SEA LEVELW‘I‘: Base from US. Geological Survey 1:24,000 i‘ I, ,7 ,, " , ' :- . ' ‘1, ' _ , » ,, 1 _ \, , x _, VVVVVVV , , _ _. _» ,_ _y .~ , , \ x, x , a ,. 1, V; 1r ‘ Borrego Mountain, 1960; Shell Reef and ‘ ’ - f ,/ 1‘ _ ~ ., ‘ 4 , , , _ , 'f _, , x , . , , ; ‘ , C, ,. ', r ,g _ , ff A '\ f :2 Harper Canyon, 1959; Borrego Mountain SE, , , > _ J _ _ ‘ ‘ I; I, Harpers Well, and Kane Spring, 1956 W, _ K .Water I: ,, ,, W" , ,, s o ,- {ffi/gé: Kg» 1 ,.’ I ' :1 " 1 x i / ‘~ " ‘ _ , , _ V ' " \\ .,,._§f‘fPf ; 4 . a l ‘ ,2 , , ( , I“ , _ . , , /, ,\ ,_ i" I: , k7 , . t, , v . _ , = . . , , Q}, ’/ R/J \ \ F; - {Lt/1176 \ E, ,\ \ 21 \ t \ 22 \ V \ \: VA \ , \ \\ V x O t \ \N ,W-\/>qp a, ¥\\ \\ I “K”; 213°07’30” ‘ a} '\ \ ‘=., Baileys W "\\ “xx, (mm 0 V, 3 I SCALE 1:24 000 1 V2 0 1 MILE MW «\V M ;\V E H—i T J r J T I : 116°05’ ,\ __\ 1 _5 o 1 KILOMETER , \j E l——l l—‘l |-—4 l——l r—————-————-—d WU \_ CONTOUR INTERVALS 10, 20, AND 40 FEET B 28 / DASHED LINES REPRESENT 10~AND 20‘FOOT CONTOURS \ I . DATUM IS MEAN SEA LEVEL \I / ' 1 IV," , fl 3 116°02’30” _ . . Locations of tectonic fractures chiefly from plate 1 ; secondary tectonic fractures north- east and southeast of the main trace mapped by MAP SHOWING DISTRIBUTION OF EFFECTS OF THE BORREGO MOUNTAIN 3:3,;Sisézitifiar“:‘Sl‘fzfgf::migl.tflf;”e- EARTHQUAKE WITHIN THE EPICENTRAL AREA 0 476—246 0 - '72 (In pocket) No. 5 a) 7 7 DAY Wax Geology of the Negaunee Quadrangle, Marquette County, Michigan mi“ GEOLOGICAL SURVEY PROFESSIONAL PAPER 788 NC ‘ {ARV Prepared in cooperation with the Geoljogical Survey Division of the Michigan Department of Natural Resources ~ j DOCUMENTS DEPARTE‘JEF‘H’ \i‘fii gunner OCT 2 81975 3m 5% 7" “ MSZSE@- Geology of the Negaunee Quadrangle, Marquette County, Michigan By WILLARD P. PUFFETT GEOLOGICAL SURVEY PROFESSIONAL PAPER 788 Prepared in cooperation with the Geological Survey Division of the Michigan Department of Natural Resom‘rces Description of an area underlain by Precambrian rocks that contain important sedimentary iron-ore deposits UNITED STATES GOVERNMENT PRINTING OFFICE, WASHINGTON : 1974 UNITED STATES DEPARTMENT OF THE INTERIOR ROGERS C. B. MORTON, Secretary GEOLOGICAL SURVEY V. E. McKelvey, Director Library of Congress catalog-card No. 73-600? For sale by the Superintendent of Documents, US. Government Printing Office Washington, D.C. 20402 — Price $4.05 (paper cover) Stock Number 2401—02494 CONTENTS Abstract Introduction .............................. Location and accessibility .................................................. Work methods and acknowledgments .............................. Glacial features .......... ..... Previous work ...................................................................... Geologic setting ..... Stratigraphy Metamorphism ................................................ Lower Precambrian rocks Kitchi Schist ........................................................................ Definition, distribution, thickness, and description ........................................................ Relation to other formations ...................................... Relative age of the Kitchi Schist .............................. Mona Schist .............. Definition, distribution, and thickness .................... Lower member ..... Nealy Creek Member of the Mona Schist .............. Sheared rhyolite tuff member of the Mona Schist Lighthouse Point Member of the Mona Schist ...... Definition, distribution, and description .......... Felsic augen zone ............................................. Undifferentiated greenstone .. Compeau Creek Gneiss ......................................... Definition, distribution, and description ................ Composition ..... Silicified zones ............................................................ Pyrite-rich zones .................................. Structure .. Felsic metavolcanic rocks .................................................. Dead River pluton Definition, distribution, and age .............................. Granodiorite porphyry ................... Hornblende diorite ...................................................... Porphyritic syenite .................................................... Middle Precambrian rocks ..... Chocolay Group .................................................................. Reany Creek Formation ............................................ Enchantment Lake Formation Mesnard Quartzite ...................................................... Kona Dolomite Definition and distribution ................................. Eastern area ............................................. 4 .......... Western area ........................................................ Chemical composition .. Wewe Slate .................................................................. Menominee Group .............................................................. Ajibik Quartzite ....................................................... Definition, distribution, and description .......... Granule marker bed .......................................... Basal Ajibik ............................... Relation to other formations .............................. Page Middle Precambrian rocks—Continued Menominee Group—Continued Siamo Slate .................................................................. 31 Clastic dikes .................. 32 N egaunee Iron-Formation .. 32 Baraga Group .................. 34 Michigamme Slate ...................................................... 34 Distribution, general description, structure, and thickness .................................................... 34 Basal quartzite .................................................... 36 Thin-bedded graywacke, metasiltite, and slate ............................................................ 36 Carbonaceous graywacke .......................... 37 Magnetite-rich argillite ..... 37 Iron-formation .................................................... 37 Chert conglomerate ............................................ 38 Mafic intrusive rocks .................................................................. 39 Metagabbro ..... .. 40 Metadiabase 40 Diabase ..... 41 Structure ................. 41 Folds ...................................................................................... 41 Eagle Mills syncline ............................................ 41 North limb of the Eagle Mills syncline ........... 41 Syncline in the Lighthouse Point Member of the Mona Schist ...................................................... 43 Shear zones and faults ........................................................ 43 Definition 43 Carp River Falls shear zone ...................................... 43 Dead River shear zone ........... 43 Other shear zones ........... 45 Normal faults 45 Reverse faults 45 Geologic history .......................................................................... 45 Magnetic surveys ................ 46 Aeromagnetic survey .......................................................... 46 Ground magnetic surveys .................................... 46 Goose Lake Member of the Siamo Slate ................ 46 Zone of magnetic argillite unit in Michigamme Slate .......................................................................... 46 Anomaly north of Dead River .............. 47 Economic geology .............. 47 Iron .................. 47 History .......................................................................... 47 Types of ore ..... 47 Ore bodies .................................................................... 48 Origin ............................... 48 Base-metal sulfide deposits ................................................ 49 Vein deposits ................................................................ 49 Sulfides in the Chert-magnetite iron-formation 50 Base metals in Carp River Falls shear zone .......... 50 Gold .............................. 50 References cited ..... .. 50 Index .................................. 53 III IV PLATE FIGURE TABLE CONTENTS ILLUSTRAHTONS [Plates are in pocket] Bedrock geologic map and sections of the Negaunee quadrangle. Geologic map and sections of the north limb of the Eagle Mills syncline. Ground-magnetic map of part of the area north of the Dead River basin, Negaunee quadrangle. Isometric diagram of “sof ” iron-ore bodies in the Negaunee Iron-Formation. Page Geologic sketch map showing western Upper Peninsula of Michigan and location of Negaunee quadrangle ........ 3 Chart showing development of stratigraphic nomenclature in the Negaunee quadrangle .......................................... 7 Photomicrograph of lithic crystal tuft of Kitchi Schist 8 Photomicrographs of felsic crystal tufi of Kitchi Schist 8 Photograph of agglomerate in Kitchi Schist 8 Photomicrograph of quartz porphyry in Kitchi Schist 10 Photograph of pillow lavas in Mona Schist 12 Photomicrograph of schist 1n Nealy Creek Member of Mona Schist 13 Photograph of sheared rhyolite tuff member of Mona Schist 14 Sketch of lenticular structures in Lighthouse Point Member of Mona Schist 16 Photograph of felsic augen in Lighthouse Point Member of Mona Schist 16 Sketch of irregular contact between Compeau Creek Gneiss and felsic porphyry intrusive ...................................... 19 Photograph of porphyritic syenite, Dead River pluton .. 22 Photomicrographs of Kona Dolomite 27 Photograph of folded and broken thin chert beds, chert breccia unit, Kona Dolomite ................................................ 27 Photomicrograph of chert-granule marker bed in Ajibik Quartzite 30 Photograph of Siamo Slate ........ 31 Sketch of sandstone dikes in Siamo Slate 33 Sketch of diamond-drill core of Negaunee Iron-Formation 34 . Generalized columnar section of Michigamme Slate, Dead River basin .. 35 . Geologic map showing details of Michigamme Slate near center of sec. 15, T. 48 N., R. 26 W ................................ 39 . Photomicrographs of iron-formation in Michigamme Slate ’ 39 . Map showing major structures and intrusive igneous masses in Negaunee quadrangle .............................................. 42 . Photograph of phyllonjtized greenstone in the Carp River Falls shear zone 43 TABLES Page Chemical and semiquantitative spectrographic analyses of Kitchi Schist in the Negaunee quadrangle .................. 9 Chemical and semiquantitative spectrographic analyses of Mona Schist .. 11 Chemical analyses and norms of Compeau Creek Gneiss in the Negaunee quadrangle .............................................. 18 Chemical and semiquantitative spectrographic analyses of felsic rock (intrusive?) in Lighthouse Point Member of the Mona Schist 20 Chemical and semiquantitative spectrographic analyses and norms of the Dead River pluton ................................ 21 Chemical analysis of slate in Enchantment Lake Formation 24 Comparison of stratigraphic divisions of the middle Precambrian rocks in the hills east of Teal Lake .................... 25 Chemical and semiquantitative spectrogr aphic analyses of slates and chert breccia in Kona Dolomite .................. 29 Carbon content of graywacke from Michigamme Slate 37 Chemical and semiquantitative spectrographic analyses of magnetite-rich argillite in the Michigamme Slate ...... 38 Chemical and semiquantitative spectrographic analyses of mafic intrusive rocks 40 Chemical, semiquantitative spectrographic, and heavy—metal analyses of rocks in the Carp River Falls shear zone 44 Summary of iron ore production in the Negaunee quadrangle 48 Sulfide-bearing quartz or quartz-carbonate veins in the Negaunee quadrangle 49 GEOLOGY OF THE NEGAUNEE QUADRANGLE, MARQUETTE COUNTY, MICHIGAN By WILLARD P. PUFFETT ABSTRACT The Negaunee quadrangle covers about 52 square miles of the east-central part of Marquette County, Mich. Most of the area is of low relief, drainage is not integrated, and glacial deposits cover more than half the area. Rocks of Precambrian age underlie the entire quadrangle. Lower Precambrian granitic and mafic metavolcanic rocks border two areas of downfolded middle Precambrian meta- sedimentary rocks of the Marquette Range Supergroup. The downfolded rocks form the Eagle Mills syncline on the north limb of the Marquette synclinorium in the southern part of the quadrangle and the eastern part of the Dead River basin in the west-central part of the quadrangle. The middle Pre- cambrian rocks are unconformable against the older rocks, but faulting has juxtaposed rocks of different ages. Keweenawan diabase dikes cut most of the older Precambrian rocks. Lower and middle Precambrian rocks are metamorphosed. Early metamorphism converted the volcanic rocks to amphibo- lite near the Compeau Creek Gneiss; later, lower grade meta- morphism converted most of the other rocks to the chlorite grade. The Keweenawan diabase dikes are not metamorphosed. Lower Precambrian rocks are mafic to intermediate meta- volcanic rocks and felsic intrusive rocks. The metavolcanic rocks include the Kitchi Schist and the Mona Schist. The felsic rocks include the Compeau Creek Gneiss and Dead River pluton, which have intruded the Mona Schist. The Kitchi Schist, the oldest formation in the quadrangle, is gray-green pyroclastic rock ranging from lapilli tuff to agglomerate with clasts as large as 5 inches. Chemically, the rocks range from dacite to rhyodacite but are dominantly dacitic. The formation is nearly equivalent in age to sub- aqueous pillow lavas in the lower member of the Mona Schist. Maximum thickness in the quadrangle is about 4,500 feet. The Mona Schist consists mainly of metabasalt and layered amphibolite and is here divided into four members: lower member, Nealy Creek Member, sheared rhyolite tuif member, and Lighthouse Point Member. Undifferentiated greenstone is mapped in two areas. Total thickness of the formation is about 23,000 feet. The lower member of the Mona Schist, about 10,000 feet thick, is dark-green fine-grained massive metabasalt character- ized in many outcrops by large pillow structures. Layers of pillows dip steeply, and the top direction is to the north. Small, widely separated quartz and quartz-carbonate veins contain, sparse copper minerals. Axinite occurs as large crystals in a quartz-calcite vein. The Nealy Creek Member of the Mona Schist, 2,000—3,000 feet thick, is greenish gray quartz-sericite-chlorite schist, near rhyodacite in composition, and originally was an air-fall tufl". Cataclastic deformation is common. The sheared rhyolite tufl member, 1,300—3,000 feet thick, is pink to greenish gray, felsic and quartz rich, and contains tabular fragments 0.5—5 inches thick. It is bounded on the north and south by faults of large displacement. The Lighthouse Point Member of the Mona Schist, at least 8,500 feet thick, is principally dark-green fine-grained amphib- olite—originally basaltic tuff—in layers 1—6 inches thick. A zone of felsic augen, saussuritized feldspar porphyroblasts, 700— 1,000 feet wide is parallel to and 700-3,000 feet from the contact of the member with the Compeau Creek Gneiss. The Compeau Creek Gneiss, light-colored foliated medium- grained gneiss of granodioritic composition, has intruded Mona Schist in the northwestern part of the quadrangle and locally reversed the dip of layering in the Lighthouse Point Member of the Mona Schist. The Dead River pluton, nonfoliated porphyritic felsic rock, has intruded the Nealy Creek and lower members of the Mona Schist in the central part of the quadrangle. Rock adjacent to the lower member is a border phase of hornblende diorite. Broad bands of coarsely porphyritic hornblende-biotite syenite cross the pluton. The main body of the pluton has a composition near that of granodiorite. Felsic metavolcanic rocks, cataclastically deformed and in part mylonitized and commonly containing glassy quartz phenocrysts, form thin tabular bodies and irregular-shaped masses, probably early Precambrian in age. Some are metatufis and some possibly are intrusive. The middle Precambrian metasedimentary rocks underlie two widely separated areas in the northern and west-central parts of the quadrangle and are assigned to the Marquette Range Supergroup, which includes the Chocolay, Menominee, and Baraga Groups. In the southern part of the quadrangle, they include, from the base upward, the Enchantment Lake Formation, Mesnard Quartzite, Kona Dolomite, Wewe Slate, Ajibik Quartzite, Siamo Slate, and Negaunee Iron-Formation. In the west-central part of the quadrangle, they include the Michigarnme Slate south of Dead River and the Reany Creek Formation north of Dead River. The Enchantment Lake Formation, a lenticular unit less than 150 feet thick, consists of basal arkosic conglomerate with vein quartz pebbles overlain by sericitic slate, sericitic quartzite, and arkose. The Reany Creek Formation, 1,500—3,500 feet thick, depos- ited in a glacial environment, is divided into three units: (1) coarse basal conglomerate, (2) a medial unit of chloritic slate containing widely scattered granitic boulders and arkose intra- clasts, and (3) an upper unit of interbedded conglomerate, slaty graywacke, and arkose and quartzite lenses. It may be a time equivalent of the Enchantment Lake Formation, but could be older. 1 2 NEGAUNEE QUADRANGLE, MARQUETTE COUNTY, MICHIGAN The Mesnard Quartzite is massive or thick-bedded, ripple- marked, and crossbedded quartzite truncated by younger meta- sedimentary rocks near the west edge of the quadrangle. Maxi- mum thickness is 200 feet. The Kona Dolomite crops out in widely separated eastern and western areas on the north limb of the Eagle Mills syn- cline. The eastern area consists of fine- to medium-grained pinkish-gray crystalline dolomite and thin beds of purplish- pink dolomite and slate. Chert is conspicuous and algal struc- tures are abundant. The western area consists of sericitic quartzite, purple ferruginous slate, vitreous quartzite overlain by chert breccia and algal structures, and silty and ferruginous slate with thin quartzite beds, truncated by younger metasedi- mentary rocks. Thickness in the eastern area is 900—1200 feet; maximum thickness in the western area, 800—900 feet. Wewe Slate is not exposed but probably underlies an area near the east edge of the quadrangle and is truncated by an unconformity at the base of the Ajibik Quartzite. In the adja- cent Marquette quadrangle, the Wewe consists of gray and thick-bedded and green and white thinly laminated slate. The Ajibik Quartzite, averaging 150 feet thick, consists of white to purplish-gray, ripple-marked, and crossbedded quartz- ite underlain by sericitic slate. An important marker zone in the lower part of the formation is a chert granule bed overlain by uniquely crossbedded quartzite. The formation truncates older metasedimentary rocks and rests on lower Precambrian rocks at the west edge of the quadrangle. . Siamo Slate, 1,500—2,000 feet thick, consists of dark-gray sericitic and chloritic slate with thin to thick beds of graywacke and, rarely, conglomerate. Carbonate cement is common in some beds. Clastic dikes subparallel to slaty cleavage are con- spicuous locally. The Goose Lake Member, as shown by mag- netic surveys, is 500—1,000 feet above the base of the slate but does not crop out. Negaunee Iron-Formation, 2,000—3,000 feet thick, consists of chert and iron-rich beds. The iron minerals, originally siderite, are hematite, goethite, and magnetite. Chert and chert-rich layers are paper thin to 12 millimeters thick; iron-rich beds range from discontinuous wisps to layers 10 mm thick. Jaspilite is common in the upper part of the formation. Important iron deposits formed by redistribution of the iron minerals and leaching of chert by downward circulating waters. Michigamme Slate consists of basal quartzite, thin-bedded graywacke with magnetite-rich and carbon-rich beds, chert- goethite-hematite iron-formation, conglomerate with large chert clasts, and pyritic and carbonaceous slate and metasiltite. It crops out in the southern part of the Dead River basin and is unconformable against the Nealy Creek Member of the Mona Schist and the Dead River pluton. Estimated thickness is 5,000—7,000 feet. The top of formation is not determinable within the quadrangle; the upper limit is in part a fault. Mafic intrusive rocks include metadiabase, metagabbro, and diabase. Metadiabase dikes trend eastward, cut all lower Pre- cambrian rocks and the Negaunee Iron-Formation, are as much as 100 feet thick, and are probably of more than one age. Meta- diabase sills, as much as 1,500 feet thick, parallel foliation in the Lighthouse Point Member of the Mona Schist. Two small masses of metagabbro in the Lighthouse Point Member prob- ably are of early Precambrian age. Diabase dikes trend east- ward to northward, are as much as 100 feet thick, and cut nearly all units. Large folds include the west-plunging Eagle Mills syncline in the southern part of the quadrangle, a part of the Marquette synclinorium; a west- to northwest-plunging syncline in Dead River basin; and a southeast plunging syncline adjacent to the Compeau Creek Gneiss in the northeastern part of the area. Smaller southeast-plunging folds are on the north limb of the Eagle Mills syncline. Broad shear zones that might, in part, control location of the downfolded areas trend west to northwest across the quad- rangle. The Carp River Falls shear zone, in the southern part of the map area, in part forms the boundary between lower and middle Precambrian rocks. The Dead River shear zone crosses the central part of the area and includes a horst of lower Pre- cambrian rocks bounded by faults along which middle Precam- brian rocks have been downfaulted for hundreds to thousands of feet. . Horizontal displacement along several northwestward- trending high-angle faults is as much as 300 feet. Displacements along the two high-angle reverse faults have repeated, in out- crop, parts of middle Precambrian formations. Magnetic anomalies are caused by the Goose Lake Member of the Siamo Slate, a magnetite-rich argillite in the Michi- gamme Slate, and a chert-magnetite zone in lower Precambrian rocks north of Dead River. Magnetic values are relatively low over the hematite-goethite ore bodies and the oxidized Ne- gaunee Iron—Formation. More than 62 million tons of “soft” iron ore has been pro- duced from the lower part of the Negaunee Iron-Formation. The ore is localized near the contact with Siamo Slate and is thicker in synclinal folds or where the contact flattens. Iron content of “soft” ore is 50—55 percent. Production in 1968 was only from the Mather mine B. ‘ Widely separated, narrow quartz and quartz-carbonate veins contain sparse amounts of chalcopyrite and, rarely, tetrahedrite but no gold. No mines have been developed on these veins. INTRODUCTION LOCATION AND ACCESSIBILITY The Negaunee 71/2-minute quadrangle1 occupies an area of about 52 square miles between lat 46°30’N. and 46°37’30”N. and long 87°30’VV. and 87°37’30”W. in east-central Marquette County, Mich. (fig. 1; pl. 1). It includes all but the easternmost half-mile of T. 48 N., R. 26 W., the southern part of T. 49 N., R. 26 W., and narrow strips of adjoining townships on the west and on the south. The northeastern part of the city of Negaunee is in the southwest corner of the quadrangle; the Marquette city limits are about 5 miles east of the quadrangle. The main industry in the quadrangle is mining and processing of iron ore by the Cleveland-Cliffs Iron 00., which employed about 1,220 men in the spring of 1968. The Mather mine B, in the southwest corner of the quadrangle, is reportedly the world’s largest and deep- est underground iron mine.0re from this mine is proc- essed at the Pioneer ore improvement plant and pellet plant in the southeastern part of the quadrangle. Ore from the Republic mine, now inactive, about 9 miles southwest of the quadrangle, was pelletized at the Eagle Mills plant in the NE 14 sec. 36, T. 48 N., R. 26 W. 1 As distinguished from Negaunee 15-minute quadrangle, the southeast quarter of which is the Negaunee 7%-minute quadrangle. INTRODUCTION 3 WORK METHODS AND ACKNOWLEDGMENTS The Negaunee quadrangle was mapped as part of the restudy by the US. Geological Survey, made in cooperation with the Geological Survey Division of the Michigan Department of Natural Resources, of a much larger area that includes the Marquette synclinorium. C. E. Fritts mapped the southern part of the quad- rangle in 1962 and 1963 and identified the Carp River Falls shear zone. Jack Hallberg mapped part of sec. 19, T. 48 N., R. 26 W., in July 1965. The data were plotted directly on an enlargement of the topographic sheet. Pace and compass traverses were on north and south bearings at 300-foot intervals. A sun compass was used in areas of large magnetic decli- nation. Areas between traverses were searched sys- 90° 89° tematically for outcrops. Ground surveys with a magnetometer delineated magnetic anomalies. All outcrops are shown on the geologic map (pl. 1); in some places small, closely spaced outcrops are shown as a single exposure. At map scale, small, isolated out- crops are necessarily shown larger than they actually are. All diabase and metadiabase dikes are shown in their true orientation, but as many of them are less than 100 feet thick, their mapped thickness is not to scale. Rocks were identified megascopically and in thin section. Rapid-rock chemical analyses were made by the US. Geological Survey. Many of the photomicro- graphs were prepared by J. A. Denson, US. Geological Survey. l l EXPLANATION Paleozoic rocks Upper Precambrian (Keweenawan) rocks Includes some probable Cambrian rocks V V k 7 % Middle Precambrian rocks 47o _ Iran-formation shown solid black ‘6» \ oGEBlC 46°- 10 o l J 10 20 MILES a ta 2 o a ... z m rri ‘ \\\\\\\ \ FIGURE 1.—Geology of western Upper Peninsula of Michigan and location of the Negaunee quadrangle. 4 NEGAUNEE QUADRANGLE, MARQUETTE COUNTY, MICHIGAN The Cleveland-Cliffs Iron Co. freely permitted use of company mine maps and diamond-drill logs and inspec- tion of diamond-drill cores. GLACIAL FEATURES Striated bedrock surfaces are common and glacial deposits are Widespread. Glacial striae show that the glaciers moved toward the s0uthwest, probably modi- fying existing east-west topographic elongation by deposits that were left. Outwash aprons, kame terraces, and ground moraine cover much of the bedrock and have disrupted preexisting drainage. Outwash aprons and kame terraces form relatively conspicuous flat areas which decrease in altitude from sOuthwest to northeast and commonly have a steep slope along their northeast margins. In the southeast- ern part of the quadrangle, however, in the N 1/2 sec. 34, T. 48 N., R. 26 W., a lobate outwash apron has steep south slopes. An outwash terrace extending north- westward from the east boundary of the quadrangle in sec. 24 to see. 4, T. 48 N., R. 26 W., is pitted with kettlelike depressions 150—700 feet in diameter and 40—80 feet deep. Outwash material in a borrow pit in the N1/2 sec. 84, T. 48 N., R. 26 W., is fairly well sorted crossbedded sand. Ground moraine mantles much of the area, in many places only as a thin veneer. Areas of outcrop shown on the map (pl. 1) commonly include patches of thin ground moraine. The moraine is a source of gravel—the largest pit is south of county road 510 in sec. 15, T. 48 N., R. 26 W.—and contains blocks as much as 10 feet across, predominately of igneous or metamorphic rock; in the hills north of Dead River are large blocks of friable red sandstone and rounded boulders of fossilifer- ous limestone. A block of friable sandstone greater than 10 feet in diameter was found at an altitude of about 1,450 feet on the divide between the branches of Reany Creek near the east boundary of sec. 5, T. 48 N., R. 26 W. The sandstone probably is J acobsville of Early Cambrian age that crops out along Lake Superior about 2 miles northeast of the quadrangle. The limestone contains corals of Paleozoic age that must have origi- nated either in Michigan east of the quadrangle or near Hudson Bay. Lacustrine-type sediments were not seen but have been reported in drill holes. Stuart, Brown, and Rhode- hamel (1954, table 10) reported the following section of unconsolidated material in the Maas mine area in sec. 31, T. 48 N., R. 26 W: Thickness (feet) Outwash—sand and gravel .............................................. 60—130 Lacustrine deposits—red clay, gravel, and ‘hardpan layers; finer grained than the usual till ...... 70—120 Till—red clay, sand and gravel, boulders, and hardpan .................................................... 10-40 PREVIOUS WORK The geology in and adjacent to the Negaunee quad- rangle has been the subject of many reports since the discovery of iron ore near Negaunee in 1844. Van Hise and Bayley (1897 ) summarized the reports published prior to 1895, and Gair and Thaden (1968) the changes in stratigraphy, nomenclature, and concepts of geology published since 1895. No detailed geologic map of the entire quadrangle has been published; but detailed geologic maps of parts of the quadrangle are included in reports by Rominger (1881), Irving (in Williams, 1890), Van Hise and Bayley (1897), and Seaman (in Van Hise and Leith, 1911, pls. XIX, XX). Descriptions of early exploration for iron ore and development of the early iron mines and other resources are detailed by Brooks (1873). The development of stratigraphic nomenclature in the quadrangle is shown in figure 2. GEOLOGIC SETTING Parts of several large-scale structures extend into the quadrangle from adjacent areas. The north limb of the Marquette syncline and the subsidiary Eagle Mills syn- cline cross the southern part of the quadrangle. The east end of the Dead River basin, an area filled with metasedimentary rocks, is in the central part of the quadrangle; only a small part of the much larger mass of the Compeau Creek Gneiss is in the northeastern part of the area. A broad shear zone, subparallel to the course of Dead River, extends across the central part of the quadrangle but was not mapped in the Mar- quette quadrangle to the east. STRATIGRAPHY All the bedrock units in the Negaunee quadrangle are of Precambrian age. Lower Precambrian rocks include the Kitchi and Mona Schists, Compeau Creek Gneiss, and Dead River pluton. The Kitchi Schist, a fragmental volcanic rock of latite to dacite composition, is approximately the same age as the lower member of the Mona Schist. The Mona Schist is a thick series of predominantly mafic extrusive volcanic rocks but includes a large body of sheared rhyolite tufi. The for- mation has been divided into four members: lower member, Nealy Creek Member, sheared rhyolite tufi member, and Lighthouse Point Member. Undifferen- tiated greenstone, a part of the Mona Schist, is mapped in two areas. Compeau Creek Gneiss is predominantly granodioritic and has intruded the Mona Schist in the northeastern part of the quadrangle. The Dead River pluton, a composite body of syenite, diorite, and grano- diorite, has intruded the Mona Schist extensively south of and locally north of the Dead River. Rocks of middle Precambrian age are unconformable GEOIJOGIC against the Mona Schist on the north limb of the Eagle Mills syncline in the southern part of the quadrangle, the Dead River pluton on the south side of the Dead River basin, and the Mona Schist on the north side of the Dead River basin. Middle Precambrian rocks are, from the base upward, the Enchantment Lake Formation, Mesnard Quartzite, Kona Dolomite, Wewe Slate (not exposed but believed to be present), Ajibik Quartzite, Siamo Slate, and Negaunee Iron-Formation, all on the north limb of the Eagle Mills syncline. These formations belong to the Marquette Range Supergroup as defined by Cannon and Gair (1970). The Michigamme Slate, the youngest formation in the area, unconformably overlies the Dead River pluton in the Dead River basin. The Reany Creek Formation is of uncertain age, but probably is strati- graphically at the base of the Marquette Range Super- group north and northeast of the Dead River basin. Metamorphosed mafic and felsic dikes are common in the Mona Schist and Compeau Creek Gneiss; the felsic dikes are probably all of early Precambrian age; the mafic dikes might be either early or middle Pre- cambrian in age. Unmetamorphosed Keweenawan dia- base dikes have intruded all rock units in the area. METAMORPHISM Metamorphism has affected all lower and middle Precambrian rocks in the quadrangle but is reflected mainly in the mineral assemblages of the mafic rocks. The quadrangle is entirely within the chlorite zone of the last regional metamorphism of pre-Keweenawan age (James, 1955, pl. 1). The characteristic mineral assemblage of the mafic rocks in this zone includes sodic plagioclase, epidote minerals, sericite, quartz, and pale-green actinolite-tremolite. Gair and Thaden (1968, p. 17) found evidence in the Lighthouse Point Member of the Mona Schist for an earlier, higher grade metamorphism. They concluded that well-alined blue- green hornblende developed prior to the intrusion of metadiabase, probably the result of igneous activity connected with the formation of the Compeau Creek Gneiss. The Lighthouse Point Member is of higher metamorphic grade than the rest of the Mona Schist in the quadrangle. It is fine- to medium-grained amphib- olite and is coarser grained near the Compeau Creek Gneiss. The sedimentary rocks in the quadrangle are only weakly metamorphosed; the main evidence for meta- morphism is chlorite and sericite and the slaty cleavage in the finer grained rocks. The metamorphism has not obliterated primary sedimentary features except in the quartzites, where silicification locally has obscured the bedding. SETTING 5 LOWER PRECAMBRIAN ROCKS Rocks of early Precambrian age underlie more than two-thirds of the Negaunee quadrangle. They are divided into two groups, mafic to intermediate volcanic rocks and felsic coarse-grained intrusive rocks. The volcanic rocks have been assigned to the Kitchi Schist and Mona Schist in earlier reports, and the names are retained in this report. Felsic intrusive rocks are the Compeau Creek Gneiss, a foliated granodiorite in the northeastern part of the quadrangle, and the Dead River pluton, a nonfoliated, generally porphyritic syenite-diorite-granodiorite in the east-central part of the quadrangle. A few small intrusive bodies, similar to rocks of the pluton, crop out north of the Dead River along the south edge of sec. 5, T. 48 N., R. 26 W., and might be outliers of the Dead River pluton. Felsic dikes intrude all lower Precambrian rocks and probably are of early Precambrian age. Some meta- morphosed mafic dikes, particularly north of the Dead River basin, have been truncated by the unconformity between lower and middle Precambrian rocks and probably are of early Precambrian age. KITCHI SCHIST DEFINITION, DISTRIBUTION, THICKNESS, AND DESCRIPTION The Kitchi Schist is foliated greenstone containing pebbles and boulders of pyroclastic rock named by Van Hise and Bayley (1895, p.496) for the Kitchi Hills in sec. 33, T. 48 N., R. 27 W., about 3 miles west of the Negaunee quadrangle, not named on modern topo- graphic maps. The maximum mapped width of the formation is 4,800 feet, along the west edge of the quadrangle. The Kitchi Schist is mainly a pyroclastic rock, rang- ing from coarse agglomerate with accessory lapilli tuff to relatively fine grained crystal tuif and crystal-lithic tufi. Fragments range from less than 1 inch to more than 5 inches in diameter. Van Hise and Bayley (1897, p. 167) reported boulders as much as 2 feet in diameter; none this large were seen in the N egaunee quadrangle. Some large fragments are well rounded, but many are angular. As a result, the rock in some outcrops re- sembles a conglomerate, and in others a breccia. The larger fragments are generally similar to the enclosing rock, whereas the smaller fragments are mainly single crystals or aggregates of crystals; some are small lithic fragments (fig. 3). Where fragments consist of pieces of single crystals, the rock resembles a porphyry, but its pyroclastic origin seems certain (fig. 4). The fragmental character of the rock is apparent on weathered surfaces but generally obscure on smoothly NEGAUNEE QUADRANGLE, MARQUETTE COUNTY, MICHIGAN 5E8 <20: mmamm w memEom <20: 2:353: o 52 EBE M 5E8 E9: m xmdzoo 2596 N 5:456 ezmzmmg z 6%: 3455 D24 Emmi Z<> :me MHOZHSOZ LOWER PRECAMBRIAN ROCKS .ofiwqanwgu $58me on» E chub—«6:25: cannuumfiwbm mo enmfimofignld “EDGE emEom ESE 55% $8: zz30 5538 <23 .85 w m m 53m 553 s a m H0 3 m m a w u. m H 3 O m a m 55:50 E82 a W M D I 325 a m 53w 2.25 may: .5sz zeniionwzofi @2382 E55 whim 0.25052: «932m mboxcmmmbm mwz Co center sec. 15 // / / / // -. 04,, T. 48 N., R. 26 w. // ,_ \ \ // / / / . y Conglomerate v // / R \ / / I . / $040 (chert slabs in / x\ V // , ¢ _/ / 5 graywacke matrix) / \gofl / / _. x / / / l0 \— ///h \\_/ / / /. / / / / / \__ / 52 W / 7O 0/ / / \«563%/ i... ./ 3 / // // ,o / \ I {T / Massive graywacke '. (weakly ferruginous; contains 5 l 65 // slate beds) / 2-foot/ .. \ _/ bed of ' / / carbonaceous -— / graywacke O 100 200 300 F EET 3.. 5/ / /' / ./ / / FIGURE 21.—Details of Michigamme Slate near center of sec. 15, T. 48 N., R. 26 W. MAFIC INTRUSIVE ROCKS Metadiabase, metagabbro, and diabase form dikes and sills and a few irregular-shaped bodies in the Negaunee quadrangle identical to such rock in the Marquette and Sands quadrangles (Gair and Thaden, 1968, p. 51—52, 57—59). Metagabbro is probably early Precambrian in age, metadiabase is of several ages, some definitely early Precambrian, and diabase is of Keweenawan age. IVIETAGABBRO Two masses of metagabbro intrude the Lighthouse Point Member of the Mona Schist in the northwestern part of the quadrangle and probably are related to metadiabase in the same area. The rock is massive, nonfoliated, dark greenish gray, medium grained, and without diabasic texture. It is composed largely of blue- green raggedly terminated hornblende with sievelike texture and saussuritized feldspar. Minor minerals are carbonate, quartz, magnetite, and leucoxene. The chemical composition (table 11) is nearly iden- tical to that of layered amphibolite of the Lighthouse Point Member of the Mona Schist (table 2) and dia- base and compares closely with that of metagabbro in the Marquette quadrangle (Gair and Thaden, 1968, p. 50, table 17). METADIABASE Metadiabase dikes and sills(?) intrude all units of early Precambrian age, Siamo Slate, and the Negaunee 40 NEGAUNEE QUADRANGLE, MARQUETTE COUNTY, MICHIGAN 1...: at. FIGURE 22.—Photomicrographs of iron-formation in the Michi- gamme Slate in roadcut near center of sec. 15, T. 48 N., R. 26 W. Light areas are chert; opaque grains are hematite and goethite; note gradational boundary between chert and iron- rich layer. Globules are light yellowish gray, noncrystalline, generally bordered by dark rims, and may have been primi- tive organisms. Planelpolarized light. Iron-Formation. Most of the metadiabase is dark gray to gray green, fine to medium grained, and massive. Some in the Mona Schist is very coarse grained to porphyritic, containing dark-green phenocrysts as much as 5 mm long. Inclusions of greenstone in the Dead River pluton in sec. 14, T. 48 N., R. 26 W., closely resemble the metadiabase of the dikes. As metadiabase is less easily recognized in the Mona Schist than in the Compeau Creek Gneiss or the Dead River pluton, undoubtedly there are more metadiabase bodies in the Mona greenstone than are shown on plate 1. The metadiabase is composed of cloudy sericitic plagioclase studded with epidote minerals, raggedly TABLE 11.—Chemical and semiquantitative spectrographic analyses of mafic intrusive rocks [Rapid-rock analyses by P. L. D. Elmore, Lowell Artis, S. D. Botts, Gillison Chloe, J. H. Glenn, Hezekiah Smith, and James Kelsey, U.S. Geological Survey. Semiquantltative spectrographic analyses for 1 by J. L. Harris and for 2 by W. B. Crandell, both U.S. Geological Survey. Results are reported to the nearest number in the series 1, 0.7, 0.5, 0.3, 0.2, 0.15, 0.1, ........ , which represent approximate midpoints of interval data on a geometric scale. The assigned interval for about 30 percent of the semiquantitative results will include the quantitative value. Elements looked for but not detected in the semiquantitatlve spectrographic analyses: As, Au, B, Be, Cd, Ce, Eu, Ge. Hf, Hg, In, La, Li, Nb, Pd, Pt, Re, Sb, Sn, Ta, Th, T], U, W, Zn. Dash leaders ( ........ ) indicate element looked for but not detected] 1 2 Chemical analyses (weight percent) 51.0 13.5 1.7 10.1 8.9 8.1 2.5 .74 .04 1.9 .56 .06 .25 .06 99 Semiquantitative spectrographic analyses (weight percent) ............ <0.0001 0.015 .007 .005 .003 .01 .005 .02 .02 .0015 .0015 .0003 ............ .01 .005 .0007 ............ .007 .005 .02 ............ .05 .02 .003 .001 .0003 .0001 .005 .003 1. Diabase. P—228—66. Lab. No. W167844. 3,150 ft north, 2,950 ft west of SE. cor. sec. 35, T. 49 N., R. 26 W. 2. Metagabbro. P736—67. Lab. No. W169301. 4,000 ft north, 2,400 ft west of SE. cor. sec. 25, T. 49 N., R. 27 W. terminated dark-green hornblende, patches of carbon- ate, and mixtures of pyrite, leucoxene, and hematite from original ilmenite or titaniferous magnetite. Dia- basic texture can be recognized in much of the rock and is conspicuous on some weathered surfaces. The original texture has been obliterated in some of the rock because metamorphism recrystallized pyroxene to hornblende and patches of carbonate. Some metadia- base composed mainly of stubby hornblende crystals is probably metapyroxenite. Most dikes trend eastward to northeastward. Excep- tions are the northwest-trending dikes in and south of the Dead River pluton west of sec. 20, T. 48 N., R. 26 W. Dikes are less than 100 feet to more than 300 feet thick. Elongate bodies of coarse-grained metadiabase, 200— 1,500 feet thick, intrude the Lighthouse Point Member of the Mona Schist north of Dead River and are parallel to regional foliation. Some of these are trun- STRUCTURE 41 cated by the unconformity at the base of the Reany Creek Formation. Reddish-brown weakly magnetic altered granophyre 2,000 feet north and 1,500 feet east of the SW. cor. sec. 10, T. 48 N., R. 26 W., is mapped as metadiabase. The mafic minerals are altered to chlorite, feldspars are clouded with sericite, and hematite stains the nonopaque minerals. Two carbonatized greenish-gray dikes are in the bottom of Dead River Canyon about 800 feet west of the east edge of sec. 14, T. 48 N., R. 26 W. The dikes are 18-30 inches thick, strike N. 20°—45° W. and are nearly vertical. Megascopically they appear porphyritic because of light-gray patches resembling feldspar phenocrysts. Examination of the rock in thin section shows it to be composed of abundant tiny leucoxene grains in a matrix of carbonate and minor chert. Apparent phenocrysts are areas of carbonate devoid of leucoxene. A dike of similar rock occurs in the Enchantment Lake Formation in the Marquette quad- rangle (Gair and Thaden, 1968, p. 51). DIABASE Diabase dikes have intruded nearly all rock units in the quadrangle, but none is known on the north limb of the Eagle Mills syncline. The diabase is dark brown to black, fine to medium grained, and weathers rusty brown. Diabase can be distinguished from metadia- base in the field, as small fragments of the former are attracted to a magnet, whereas fragments of most of the metadiabase are not. The dikes range in thickness from a few inches to nearly 100 feet and are as much as several hundred feet to more than 2 miles long. The presence of dikes is indicated by narrow negative magnetic anomalies. Many diabase dikes shown on the geologic map (pl. 1) were extended across areas of no outcrop by such magnetic data. The diabase consists mainly of labradorite and augite with minor amounts of magnetite. Olivine is present in some dikes and is commonly altered to iddingsite or serpentine minerals. Quartz occurs in some dikes as granophyric intergrowths with feldspar. Commonly the dikes containing quartz are reddish brown, but not all reddish dikes in the area contain granophyric quartz. Diabase dikes trend eastward or a few degrees north of east except for one that trends northeastward across secs. 5 and 6, T. 48 N., R. 26 W. STRUCTURE Layered rocks of middle Precambrian age are down- folded into, and separated by unconformities from, lower Precambrian rocks. Vertical displacement along shear zones has been important locally in juxtaposing rocks of different age. Anticlines cannot be mapped between the downfolded areas; this suggests that most of the synclines were developed essentially as a result of relative uplift of the lower Precambrian rocks rather than by regional compression. The main structural features in the quadrangle are synclines, shear zones, normal faults, and at least two reverse faults (fig. 23). FOLDS Three large synclines are present in the quadrangle. The Eagle Mills syncline plunges gently west across the southern part of the area on the north limb of the Marquette synclinorium, most of which is outside the quadrangle. Michigamme Slate occupies a northwest- plunging syncline near the west border of sec. 14, T. 48 N., R. 26 W., but with truncation of the north limb by the fault along the Dead River shear zone, only the south limb continues across the quadrangle west of sec. 14. The south-dipping rocks in the Reany Creek Formation probably correspond to the north limb of the syncline but are separated from the south limb by the Dead River shear zone. A syncline in the layered amphibolite of the Lighthouse Point Member of the Mona Schist that plunges gently east-southeast south of the Compeau Creek Gneiss apparently was formed where uplift of the gneiss reversed the generally north- eastward dip of the amphibolite. Several smaller folds are present in the area. A south- east-plunging syncline and smaller associated anti- clines that possibly formed by drag along faults, all overturned in part, complicate the north limb of the Eagle Mills syncline in the NE% sec. 32 (pl. 2). Anti- clines in the Michigamme Slate are indicated by oppos- ing dips in the NWM; sec. 16, T. 48 N., R. 26 W., but the magnitude of the folding is indeterminate because of poor exposures. EAGLE MILLS SYNCLINE The Eagle Mills syncline is well delineated by the bedding attitudes in the Siamo Slate. South of Carp River, the beds dip 35°—60°, generally northeast; north of Carp River, 20°—35°, generally south. The axis of the syncline follows approximately the course of the Carp River in the southeastern part of the quadrangle, but the fold plunges to the west, whereas the river flows to the east. The syncline passes through the Maas-Negaunee mine (pl. 4) and into the Mather mine B, where it is cut off by the Jackson fault. The probable eastern extension of the Eagle Mills syncline reaches the NEM; sec. 6, T. 47 N., R. 25 W. (Gair and Thaden, 1968, pl. 1), where it corresponds to the north limb of 42 NEGAUNEE QUADRANGLE, MARQUETTE COUNTY, MICHIGAN I? 3 3 3 ~ , ‘Qr'fl/ —/_\' »4\ -r\, ~-’ ‘\'\ 1'5 79/ Y 25// *W/xl\&>,x7\¢x\|\‘~/-\'\’/_\I'\/< < > < a 25 Q/uq /\‘,~/‘ \/\\\l/7,\/J(/“/_\“/_"':\ /_‘/\/_1 \l I-1/—“,1\,\I\I -,,,\I V,\ \I,‘\\,‘ A» v 0 \\”1"\i:“<"':"fil’ /,\'/|\',/|;‘\J:|’ (I, _ 50‘7/;/ 7 55 \ ~\V\“\l/\/“_ \" "II/V I I >,\\,\'/‘ . \ {a l/|‘/’_\/\I)/\/\Ll:.:_/ H I" “4 ‘7 /‘ Dead R1ver pluton \ ’19 '\\\‘§5-\—\7\"I\’/ I \_/ —,5_\ /\\\//;‘~\—\ \ 25\ \/'\/_ /\\ \—/I‘\,.T’L\\ :"JIC/ \—,|\/_ \ 45 / I- l; A ~/\ \ ‘ \— ‘V? [Dr‘s 50 \‘LX'AAZNQ Compeau Creek \ rm ‘1 Gneiss ’4‘“, wwww Zone of shearing ._L_ 'T.49N. D High-angle fault U ,upthrawn side; D , downth'rown side —l—> Syncline Showing trace ofarial plane T.48 N. f BE‘TADmfilVER-fl V‘vsmv‘wmv‘m W“ .a .fi J1“, ‘fiwwtwwuwM SHEAR ZON E " V. v—vwv‘v‘mw Luau,” and direction of plunge Unconformity at base of Reany Creek Formation D\4 Unconformity at base of Enchantment Lake Formation o X ~35 45 _ . —“—5 . x 4330 cClure Basm Strike and dip \\\ < 47“ N r 7 ofbeds 4 W x\ 7":I\ ""4X4VL<.74>4<‘. ‘43—0 “A < 4 ’r r“/\ < x— v :L Strike and dip 8?‘ \ \ o ,1 V P V i V4 of foliation \\ \ v ’o a ‘56) L \eb 30* will?“ SHEAR ZONE 70“ 85“ CARP RIVEB< f/LLLS SHEAR ZONE W-WM’ “;“:::..:~,va «~-w~«w--*~*w~“ wm“ '/ \ _/ o \72; " u + \ Teal ’0’.— MILLS SYNC e ”—— «» L’NE \gake & R //9 \ T. 48 N. T. 47 N. ACK 0 FAULT __‘l’)/— \ R27 W. R.26 W. the Marquettesynclinorium. The axis of the syncline arches broadly to the north on a north-south axis. NORTH LIMB OF THE EAGLE MILLS SYNCLINE l The complex structure of the Kona Dolomite and other formations in the hills east of Teal Lake was mapped at a scale of 126,000 (pl. 2). Beds in most of the area dip homoclinally to the south or southeast; FIGURE 23,—Major structures and intrusive igneous masses in Negaunee quadrangle. the upper quartzite of the Kona occupies a syncline about 1,000 feet northeast of the center of sec. 32, and the quartzite and adjacent slate are in an isoclinal syncline overturned to the south about’1,800 feet west- northwest of the center of sec. 32. The Kona Dolomite extends only a very short dis- tance west of sec. 32. Because the Kona is in a south- east-plunging syncline in sec. 32, only the lower part STRUCTURE 43 is exposed in that area; dolomitic beds from higher in the section do not appear in the axial part of the syn- cline. An east-trending fault along the steep cliffs at the west edge of sec. 32 juxtaposes Ajibik Quartzite and Kona Dolomite. West of there, the Ajibik truncates the Kona in part on an unconformity and in places along a fault that juxtaposes the Mesnard Quartzite. The east-plunging folds contrast with the westward plunging Eagle Mills syncline and are not readily explained. East-plunging folds in the Kona Dolomite, 6 miles east in the N1/2 sec. 32, T. 48 N., R. 25 W., are related to a large cross-fold reflecting an uplift of base- ment rocks south of the Marquette synclinorium (Gair and Thaden, 1968, p. 63, pls. 1, 4). SYNCLINE IN THE LIGHTHOUSE POINT MEMBER OF THE MONA SCHIST The regional northeastward dip of the Lighthouse Point Member changes near its contact with the Compeau Creek Gneiss such that dips are west along the west side of the gneiss and south along the south side. The resulting syncline plunges southeastward south of the gneiss and northwestward west of the gneiss. The amphibolite evidently was turned up near the margins of the gneiss, either at the time of intru- sion or later. SHEAR ZONES AND FAULTS DEFINITION Shear zones shown on the map (pl. 1) indicate faults or fault zones too wide to be shown by a single line; some indicate special areas of broken rock along which there is no evidence of measurable offset. CARP RIVER FALLS SHEAR ZONE The Carp River Falls shear zone (C. E. Fritts, written commun., 1965), exposed near rapids in Carp River in the SE. cor. sec. 29, T. 48 N., R. 26 W., is marked by altered phyllonitic metavolcanic rock that can be traced east and west to near the edges of the quadrangle (pl. 1). The zone is indicated on the map as a narrow, well-defined structure, but the symbol marks only the north limit of a broad zone of sheared and altered lower Precambrian rocks. The sheared rocks, mapped as undifferentiated greenstone, have their greatestmapped width about 1,500 feet, in the NEIA sec. 31, T. 48 N., R. 26 W. The Carp River Falls shear zone apparently is part of the deformation belt (Gair and Thaden, 1968, p. 60—61) along the north margin of the Marquette syn- clinorium. The full width of the zone of shearing is not known, as only moderately deformed middle Precam- brian rocks cover the southern part of the sheared older rocks. The sheared rocks have been intensely altered. Good exposures of the altered rock in the shear zone are on the north side of US. Highway 41 near the south boundary of sec. 29, T. 48 N., R. 26 W. (fig. 24). Seri- cite, chlorite, carbonate, and leucoxene are the most conspicuous alteration products. Weathered surfaces are commonly stained brown from oxidized iron min- erals. Quartz-carbonate veinlets form an anastomosing network in some rock. Copper minerals are present locally. Analytical data suggest that the altered rocks are enriched in CaO and CO2 (table 12). Along the Carp River Falls shear zone in the eastern part of the quadrangle, middle Precambrian rocks are juxtaposed with the oldest Precambrian rocks in the area. The vertical displacement is unknown but could be thousands of feet. Nearly vertical lineation on shear surfaces suggests that horizontal movement was slight or has been obscured by later downdip movement. In the east, the shear zone is along the contact between the Mona Schist and the Enchantment Lake Forma- tion. Near the north boundary of sec. 33, the trend of middle Precambrian rocks turns west-southwest, whereas the Carp River Falls shear zone strikes nearly due west. This divergence in strike between the shear zone and the middle Precambrian rocks suggests the younger rocks were folded prior to the latest move- ment along the shear zone. The shear zone might have been an important structure formed in early Precam- brian time, and movement along it could have been repeated through later Precambrian time. FIGURE 24.—-Phyllonitized greenstone in the Carp River Falls shear zone exposed on north side of US. Highway 41 near south border of sec. 29, T. 48 N., R. 26 W. (For analysis of rock see table 12, column 1.) 44 NEGAUNEE QUADRANGLE, MARQUETTE COUNTY, MICHIGAN TABLE 12.—Chemical, semiquantitative spectrographic, and heavy-metal analyses of rocks in the Carp River Falls shear zone [Rapid-er analyses by P. L. D. Elmore, S. D. Botts, Lowell Artis, Gillison Chloe, Hezekiah Smith. and Dennis Taylor, U.S. Geological Survey. Semiquantitative spectrographlc analyses by J. L. Harris, U.S. _Geo_logical Survey. Heavy-minerals determination by atomic absorption by Claude Huffman and J. D. Mensik. U.S. Geological Survey. Results of semiquaptltative spectrographic analyses are reported to the nearest number in the series 1, 0.7. 0.5, 0.3, 0.2, 0.15, 0.1, ........ , which. represent approximate midpomts of interval data on a geometric scale. The assigned interval for about 30 percent of the semiquantitative results will include the quantitative value. Elements looked for but not detected in the semiquantitative spectrographic analyses: Ag, As, Au. Be, Bi, Cd, Ce, Ge, Hf, Hg. In, La, L1, Nb, Pd, Pt, Re. Sh, Sn. Ta, Te, Th, Tl, U, W, Zn. Dash leaders ( ........ ) indicate element looked for but not detected] 1 2 3 4 5 6 7 Chemical analyses (weight percent) 42.1 47.7 66.2 58.9 42.8 62.7 62.8 10.9 15.3 14.4 15.8 10.9 13.9 14.2 2.1 .00 1.0 .91 .69 1.9 1.0 8.4 10.9 .92 7.9 9.1 4.2 4.7 4.6 6.3 1.9 5.8 8.4 2.7 2.9 10.6 6.1 3.3 .63 8.5 2.8 3.5 .26 .10 2.3 .76 .19 .59 .55 1.7 1.6 3.2 2.0 1.3 2.8 2.4 .18 .08 .14 .17 .11 .15 .08 3.2 5.4 1.4 4.7 3.5 2.7 2.3 . . .53 . . .52 .50 . .10 .16 .16 .13 3.9 4.8 99 100 0.03 0.03 .002 .0015 .015 .02 .003 .005 .001 .001 .0003 ............ .01 .007 .0007 ............ .001 .001 .01 .007 .01 .01 .0015 .001 00015 .0001 01 .007 Heavy-metal analyses (ppm) Copper ............................ 90 28 13 50 160 25 38 Zinc ....... _. 64 100 10 100 120 43 51 Lead ................................ (all samples less than 25) 1. P—125—65. Lab. No. 165793. 3,000 ft west, 150 ft north of SE. cor. sec. 29, T. 48 N., R. 26 W. Sericitic chloritic greenstone with knots of quartz-carbonate and limonite. Strongly sheared and highly altered. (See photograph, fig. 24.) 2. P—126—65. Lab. No. 165794. 400 ft east of P—125—65. Assumed to be unaltered equivalent of P—125—65. 3. P—127—65. Lab. No. 165795. 1,020 ft east, 350 ft north of SE. cor. sec. 30. T. 48 N., R. 26 W. Pinkish-gray sericitic carbonatized felsic porphyry with scattered pyrite. Along north margin of Carp River Falls shear zone. 4. P—128—65. Lab. No. 165796. 1,650 ft west, 850 ft south of NE. cor. sec. 86, T. 48 N., R. 27 W. Gray “unaltered” slate; sericitic, chloritic, few chert veinlets, 5. Pflg‘gisfiGBF’l'lgfi'No. 165797. 100 ft downslope north from P—128-65. Altered equivalent of P—128—65. Carbonatized chlofitic slate with chert and muscovite. Some limonite and chalcopyrite. 6. P—180—65. Lab. No. 165798. 1,700 ft west, 350 ft south of NE. cor. sec. 36, '1‘. 48 N., R. 27 W. Strongly sheared felsic tufE(?) in Carp River Falls shear zone. Quartz-sericitecarbonate chert rock; few angular and subrounded quartz grains. 7. P—181—65. Lab. No. 165799. 2,600 ft south, 400 ft west NE. cor. sec. 36, T. 48 N., R. 27 W. Identical to P—130-65. Along trace of Carp River Falls shear zone. DEAD RIVER SHEAR ZONE The Dead River shear zone lies within the sheared rhyolite tufi member of the Mona Schist between the northwestward trending faults that bound the tufi. For most of its extent in the Negaunee quadrangle, the zone thus is in a horst of lower Precambrian rocks bounded on the north and south by middle Precambrian rocks. The width of the shear zone ranges from about 800 feet near the west border of the quadrangle to about 3,400 feet in sec. 10, T. 48 N., R. 26 W. The southeastern extension of the shear zone is not well defined near the east border of the quadrangle and was not found in the Marquette quadrangle (Gair and Thaden, 1968, pl. 1). The amount of vertical ofiset on the faults bound- ing the shear zone is unknown but could be several thousand feet. On the scuth side of the shear zone where the sheared rhyolite tufl member and Michi- gamme Slate are juxtaposed, the ofiset is greater than the entire Marquette Range Supergroup below the Michigamme Slate plus an unknown thickness of lower Precambrian rock. On the north side of the shear zone, the sheared rhyolite tufi member and the stratigraphi~ cally equivalent Nealy Creek Member of the Mona are faulted against the Reany Creek Formation. The mini- mum vertical displacement on the north side of the shear zone exceeds the thickness of the Lighthouse Point Member of theMona Schist. GEOLOGIC HISTORY 45 The Dead River shear zone probably controls the north limits of the Dead River basin. The northwest projection of the shear zone lies along the northeast edge of the sedimentary rocks in the Dead River area (Seaman, in Van Hise and Leith, 1911, pl. XX). The projection of the shear zone to the southeast across the Marquette quadrangle meets a conspicuous reentrant in the shoreline of Lake Superior (Gair and Thaden, 1968, pl. 1). The most recent movement in the shear zone must have occurred in late middle Precambrian or early late Precambrian time after Michigamme Slate deposition and before Keweenawan diabase intrusion. A northwest- trending metadiabase dike crosses the south boundary of the shear zone without offset in the SE14 sec. 10, T. 48 N., R. 26 W., indicating that. some metamorphism occurred after the faulting. The Dead River shear zone is one of many northwest- trending structures in northern Michigan that have largely controlled the location and extent of basins of metasedimentary rocks. Older rocks have been rela- tively uplifted and younger rocks depressed, escaping the planation that removed younger rocks from other areas. OTHER SHEAR ZONES \ The Willow Creek shear zone in the northwest quarter of the quadrangle and the Picket Lake shear zone in the southwest quarter of the quadrangle are poorly defined. The Willow Creek shear zone is represented in sev- eral small but widely scattered outcrops. Irregular- shaped streaks of felsic rock in dark-gray-green amphib- olite about 150 feet south of the NW. cor. sec. 4, T. 48 N., R. 26 W., appear to be fragments of a once- continuous body of felsic rock. The streaks are 1—5 inches thick by 5—18 inches long and generally lens shaped. Fragments of felsic rock in amphibolite in the NV; SE 14 sec. 31, and 2,000 feet south and 500 feet east of the NW. cor. sec. 31, T. 49 N., R. 26 W., are similar to those near the NW. cor. sec. 4. The Picket Lake shear zone trends west and north- west from Picket Lake in the NW% sec. 30, T. 48 N., R. 26 W. Fragments of greenstone in quartz-carbonate vein are direct evidence for this shear zone. Steep cliffs that parallel the trend of the vein northwest of Picket Lake afford indirect evidence of faulting. NORMAL FAULTS Well-defined faults along which relative offsets can be measured are few in the quadrangle. Those that are most easily recognized strike at high angles to layering, but there probably are undetected faults that parallel layering in the greenstone or metasedimentary rocks. Many faults in the quadrangle have a conspicuous northwest trend, dip at a high angle or vertically, and are assumed to be normal faults. The faults in secs. 20, 25, and 26, T. 48 N., R. 26 W., are indicated in part by conspicuous declivities in the land surface, but their relative displacements are unknown. REVERSE FAULTS At least two high-angle reverse faults are known in the quadrangle: one in the NW1/1 sec. 12, the other east of Teal Lake in sec. 31, T. 38 N., R. 26 W. The fault in sec. 31 has repeated the Ajibik Quartzite; that in sec. 12 cuts the Reany Creek Formation and has repeated the contact between the Lighthouse Point Member of the Mona Schist and the Reany Creek. GEOLOGIC HISTORY The earliest recorded geologic event in the quad- rangle was the volcanism in early Precambrian time that gave rise to the Kitchi Schist and the Mona Schist. Subaerial volcanism produced the coarse agglomerate and lapilli tuff of the Kitchi Schist and subaqueous ex- trusions of basalt, the pillow structures in the Mona Schist. Basaltic ash falls and intermittent intrusions, extrusions, and ash falls of more felsic rock formed the Lighthouse Point Member of the Mona Schist. The vol- canic rocks were later tilted and foliated. The Compeau Creek Gneiss intruded the Lighthouse Point Member of the Mona Schist, metamorphosed it to amphibolite, and tilted or dragged the layered rocks upward around the margins of the gneiss. Probably in late early Precambrian time, felsic rocks of the Dead River pluton intruded the lower member and the Nealy Creek Member of the Mona Schist, and diabase dikes and sills intruded many of the lower Pre- cambrian rocks. Diastrophism closed early Precambrian time. The land was deeply eroded. Glaciers deepened valleys in the Lighthouse Point Member of the Mona Schist and deposited sediments of the Reany Creek Formation— the earliest deposits of the Marquette Range Super- group. The bevelled lower member of the Mona Schist was covered by mixtures of quartz and quartz-feldspar sand from a granitic terrane. Seas inundated the rocks, and the clean sands of the Mesnard Quartzite formed as lenticular shelf deposits. Carbonate rocks of the Kona Dolomite were deposited in shallow seas and algal col- onies — the earliest evidence of life on this planet — flourished. The seas deepened and the pelitic sediments of the Wewe Slate accumulated. Another period of erosion bevelled the Mesnard Quartzite, Kona Dolomite, and Wewe Slate. Clean 46 NEGAUNEE QUADRANGLE, MARQUETTE COUNTY, MICHIGAN sands of the Ajibik Quartzite formed another shelf deposit similar to the Mesnard Quartzite. Detritus from the Kona Dolomite accumulated in a granule marker bed in the lower part of the quartzite. The seas deepened and mud and mixtures of mud and coarser detritus were deposited as a delta. Flood-swollen streams carried coarse gravels into the upper part of the deposit. A long period of structurally stable conditions fol- lowed during which iron-rich and silica-rich sediments of the Negaunee Iron-Formation accumulated in shal- low troughs adjacent to long-exposed low-lying land masses (James, 1966, p. W50). Uplift and erosion followed the lithification of the Negaunee Iron-Formation. Blocks, cobbles, and peb- bles of iron-formation were deposited on the eroded surface of the iron-formation. At the time that the iron-rich rocks of the present Negaunee Iron-Forma- tion were being eroded, parts of the Dead River pluton were being exposed to erosion. During later incursions of the sea, the pyritic and carbonaceous pelitic sedi- ments of the present Michigamme Slate accumulated. Volcanism contributed detritus to a conglomerate over- lying iron-formation in the lower part of the Michi- gamme Slate. Diastrophism followed, marked by crustal displace- ments along northwest-trending zones and crumpling of rocks along west-to-northwest-trending axes. This interval of deformation marked the close of sedimenta- tion in middle Precambrian time. Low grade regional metamorphism converted mafic rocks to greenstone and pelitic rocks to chloritic-seri- citic slate. Later, in late Precambrian (Keweenawan) time, diabase dikes were intruded. There is little evidence in the quadrangle of the geologic history between the close of Precambrian time and the Pleistocene glaciation. Glacially transported blocks of sandstone believed to be of Early Cambrian age are widely distributed and might represent deposits that covered the area in early Paleozoic time. Some glacial erratics are of fossiliferous Paleozoic limestone. Undoubtedly, a vast amount of bedrock was eroded, but detritus from this erosion is not known in the quad- rangle. The soft iron ore bodies were formed during this long interval by the leaching of silica and redistribution of iron minerals in the Negaunee Iron-Formation. Pleistocene glaciation was the latest major geologic event in the area. Accumulations of debris as much as 200 feet thick were left by melting of the glaciers. Struc- tural rebound after disappearance of the ice interrupted preexisting drainage; the modern drainage is noninte- grated, and swamps and nongraded streams are com- mon. MAGNETIC SURVEYS AEROMAGNETIC SURVEY The aeromagnetic survey of an area that includes the N egaunee quadrangle shows a magnetic high related to the Dead River pluton, a magnetic high trending north- west from the SE. cor. sec. 5, T. 48 N., R. 26 W., and some east-west linear magnetic lows caused by Kewee- nawan diabase dikes (Case and Gair, 1965). A general decrease in magnetic values over the Negaunee Iron- Formation in the southwest quarter of the quadrangle is probably caused by oxidation of the iron-formation (Case and Gair, 1965, p. 5). GROUND MAGNETIC SURVEYS Ground magnetometer surveys were made during the winter months when lakes and swamps were frozen. Traverses were made at 300-foot intervals; readings were generally taken every 100 feet along traverses, but every 50 feet above steep magnetic gradients. Standard procdure was to establish a base station which was re- occupied at intervals of 2 hours or less and from which each day a correction graph was made for fluctuations in the earth’s magnetic field. Magnetic readings in a particular part of the area were corrected to a common base, but base stations in the several widely separated areas surveyed were not calibrated with one another or to a common base. Ground magnetic surveys were in four areas in the quadrangle, covering a total area of approximately 6 square miles. Three of the surveys are discussed subse- quently. The fourth sought unsuccessfully to determine if there is a marked magnetic gradient along the north and south margins of the sheared rhyolite tufi member of the Mona Schist and is not discussed further. GOOSE LAKE MEMBER OF THE SIAMO SLATE A narrow linear magnetic anomaly was traced from the edge of the quadrangle in sec. 36, T. 48 N., R. 26 W., northwest to near the center of sec. 34, T. 48 N., R. 26 W. (pl. 1). There are no outcrops of the magnetic bed, but its position within the Siamo Slate suggests that it is the Goose Lake Member (Tyler and Twenhofel, 1952, p. 118—125). The average width of the magnetic ridge is 200—250 feet. The total relief, at right angles to the trend of the anomaly, is 500—800 gammas. A narrow east-trending magnetic zone with relief of 350 gammas passing under Teal Lake in sec. 36, T. 48 N., R. 27 W., is in the correct position to be the Goose Lake Member. The anomaly could not be traced through the residential area east of the lake. The absence of a magnetic anomaly westward from sec. 34 toward Teal Lake might be due either to lensing ECONOMIC GEOLOGY 47 out of the Goose Lake Member or to its oxidation. The anomalies described evidently are too low in magnetic relief or too limited in extent to be indicated on the aeromagnetic map. ZONE OF MAGNETIC ARGILLITE UNIT IN MICHIGAMME SLATE The magnetic argillite in the Michigamme Slate south of Dead River can be traced by ground magnetic surveys from the west edge of the quadrangle to the SE14 sec. 15, T. 48 N., R. 26 W. The argillite is 100— 300 feet thick and affects the readings of only 1 or 2 stations in traverses across it. The magnetic relief, a few hundred gammas to as much as 5,000 gammas, is due to differences in the amount of oxidation of the slate and thickness of overburden. On the aeromag- netic map the anomaly evidently is masked by higher magnetic values over the Dead River pluton. The anomaly is discontinuous and poorly defined in sec. 16, T. 48 N., R. 26 W., and to the east. One or two linear anomalies of low intensity might have been caused by the argillite. As no anomaly could be traced around the east end of the syncline at the northeast shore of McClure Storage Basin, it is likely that the unit either has lensed out there or is oxidized. ANOMALY NORTH OF DEAD RIVER A ground magnetic survey (pl. 3) more closely defined the aeromagnetic anomaly (Case and Gair, 1965) ex- tending northwest from the SE. cor. sec. 5, T. 48 N., R. 26 W. The anomaly on the aeromagnetic map is as much as 2,000 feet wide (about 500 ft above the earth’s surface) ; however, the ground survey revealed that the anomaly is generally less than 300 feet wide at ground level. The anomaly source is in layered Chert-magnetite rock that is bordered by coarse-grained metadiabase. Chert layers range in thickness from less than one-half inch to more than 1 inch and are locally brecciated. Pyrite is common. Magnetite is concentrated along parting planes between layers. Fine-grained purplish- brown sphalerite is contained in a few exposures in the Nl/ZSEIA sec. 6, T. 48 N., R. 26 W. The full thickness of the Chert-magnetite rock is unknown; where exposed in the roadcut near the SE. cor. sec. 5, it is a few feet thick, but in sec. 6, T. 48 N., R. 26 W., it is about 180 feet thick. The Chert-magnetite rock is probably a metasedi- mentary unit of early Precambrian age. It extends for several miles northwest of the quadrangle but is limited on the southeast by the unconformity at the base of the Reany Creek Formation. The relation between the magnetic unit and bordering metadiabase is not known. Perhaps diabase sills were intruded into a sedimentary sequence and gave rise to the magnetite by contact thermal metamorphism or metasomatism. The presence of sphalerite suggests such contact effects of the meta- diabase. Exploration for base-metal sulfides, therefore, could be guided by magnetic surveying. ECONOMIC GEOLOGY Iron ore is the most important mineral product in the Negaunee quadrangle and has been mined since the late 1840’s. Widely scattered veins contain mainly cop- per minerals with lesser amounts of zinc and lead sul- fides; some of the veins have been explored in shallow shafts, but no production is known. A small amount of Kona Dolomite in the NE IA sec. 35, T. 48 N., R. 26 W., was quarried for flux in the old Morgan furnace that went into blast November 27, 1863 (Brooks, 1873, p. 35) and operated for several years. Greenstone was quarried north of the Morgan Heights Sanatorium for use in roofing slate. Rock from the Enchantment Lake Formation, quarried in 1849 and 1850 (Brooks, 1873, p. 21) near Teal Lake in the NW% sec. 31, T. 48 N., R. 26 W., was sawed into blocks for use as whetstones. Several pits were the source for gravel in 1968. IRON HISTORY Iron ore was discovered near Negaunee, Mich., in 1845 (Brooks, 1873, p. 14) a few hundred feet south- west of the N egaunee quadrangle. Some ore from the original location (Jackson) was mined in 1846 and reduced in a catalan forge adjacent to the Carp River near the NE. cor. sec. 32, T. 48 N., R. 26 W., on Febru- ary 10, 1848 (Brooks, 1873, p. 17). The Morgan fur- nace, in the NWIA sec. 36, T. 48 N., R. 26 W., produced 337 tons of iron in 1863. Remnants of kilns that sup- plied charcoal to Morgan furnace can be seen today in the SW14 sec. 12, SE14 sec. 17, T. 48 N., R. 26 W., and near the SW. cor. sec. 33, T. 49 N., R. 26 W. In 1872, 5,213 tons of steel were produced from ore that con- sisted of 75 percent specularite (hard ore) and 25 per- cent hematite (soft ore). Iron ore was shipped to Pennsylvania in 1850 and 1852; after the ship canal at Sault Ste. Marie, Mich., opened in 1856, regular shipments were made to the lower Great Lakes (Marsden, 1968, p. 492). Large-scale mining operations started about 1870 in secs. 6, 7, 8, and 18, T. 47 N., R. 26 W., (Brooks, 1873, p. 50). Several large underground mines, dating from 1887, are in the southwest quarter of the quadrangle along the limbs and trough of the Eagle Mills syncline, westward from near the east limits of the iron-forma- 48 NEGAUNEE QUADRANGLE, MARQUETTE COUNTY, MICHIGAN tion; from east to west, the mines are the Barassa, Negaunee, Adams, Maas, and Mather B. The Mather mine B includes old properties formerly known as Cleveland Hematite, East New York, Iron Center, and Ames, parts of which extend outside the Negaunee quadrangle. All except the Mather mine B were inactive in 1968. Production data are tabulated in table 13. TABLE 13.—Summary of iron ore production in the Negaunee quadrangle [Source: The Lake Superior Iron Ore Association and The Michigan Department of Conservation, Geological Survey Division] Total production (tons) Years of Type of ore production 1913-24 Name of Location mine Adams ............ T. 48 N., R. 26 W.: SW54SWV. sec. 32. T. 47 N., R. 26 W.: NW1/4NW14 sec. 5 and E%NE% sec. 6. .......... T. 48 N., R. 26 W.: SW%SEV4. sec. 32. Soft, red, nonbessemer 242,348 1903 (an old ex- ploration in 1903) l907~67 Barassa Mans ................ T. 47 N., R. 26 W.: NWMNWIA sec. 5 and Nl/zNI/é sec. 6. T. 48 N., R. 26 W.: E%E%SW% sec. 31 and Wl/QSWV‘ sec. 32. ...... T. 47 N., R. 26 W.: 1887—1949 Soft, red, nonbessemer ....22,735,479 Soft, red, bessemer and nonbessemer. 21,311,386 Negaunee 'swx/i sec. 32. Mather B ........ T. 47 N., R. 27 W.: 1950-67 ................................ 18,491,331 sec. 2 (except NIANEJANE’A) and sec. 1 (except N%N%NW%). Total 62,789,312 TYPES OF ORE According to Anderson (1968, p. 51), iron ore con- sists of (1) high-grade direct-shipping “soft” ore, (2) high-grade direct-shipping “hard” ore, (3) siliceous ore, and (4) concentrates and agglomerates (pellets) from low-grade iron-formation. “Soft” ore is porous, friable, earthy to semiplastic, and made up chiefly of hematite and martite with minor amounts of goethite, unreplaced chert, and silicates (mica, chlorite). “Hard” ore is hard, dense, and compact, has low porosity, and is composed of magnetite, martite, dense compact hematite, and specularite; this ore constitutes the lump ore. Marsden (1968, p. 464) gave the important compo- nents of ore from the Marquette Range as follows: Type of ore Fe SiOz P Mn Average nonbessemer ...... 52.40 4.47 0.532 0.27 .......... 7 > 5" O a. U) ' '0: Moisture Average pellets ...... .. 63.04 7.02 .038 .09 .86 .......... 1 20 Average siliceous o ....... 36.92 48.71 .028 .07 .58 .......... 2.29 Lump ore ....................... 61.77 6.46 .110 .28 1.85 0.005 .50 The average iron content of primary iron-formation approximates 26 percent (dried), and the average iron content of altered Negaunee Iron-Formation approxi- mates 31 percent (dried) (Anderson, 1968, p. 510). Production from mines in the quadrangle has been direct-shipping “soft” ore. No hard ore has been pro- duced. In recent years some of the “sof ” ore has been agglomerated into pellets. The bodies of “soft” ore contain more than 50 per- cent iron. Areas adjacent to the ore bodies might have been enriched to as much as 40—50 percent iron but generally are not minable for direct-shipping ore. The ore bodies discussed in this report are those containing direct-shipping ore; information about them has been obtained from the records of the mining companies. ORE BODIES The generalized shape of the ore bodies in the Negaunee Iron-Formation is shown on plate 4. The ore is in the lower part of the Negaunee Iron-Formation and is thickest and most continuous in troughlike structures or where the contact between the Siamo Slate and the Negaunee Iron-Formation flattens. Ore bodies are as much as 260 feet thick and 1,400 feet wide (measured at right angles to the general plunge of the troughlike structures). The ore is confined to the trough of the Eagle Mills syncline in the eastern part of the formation and can be traced southWestward into the Mather mine B, where it is downdropped to the south along the Jackson fault. The Cambria-Jackson ore body, shown between sea level and 500 feet in alti- tude on the 9,400-foot-west, 9,800-foot-west, and 10,200-foot-west fences, appears to have been localized adjacent to metadiabase dikes rather than to a fold. A zone of sheared rock that forms the north boundary of the ore in the Maas and Negaunee mines might repre- sent a fault that cannot be recognized in rocks at the surface. Metadiabase cuts the ore bodies in all the mines, and a thick metadiabase sill is above some of the ore in the Mather mine B, particularly on the downdropped side of the Jackson fault. Locally the ore is thicker on the updip side of dikes, but ore is not present above all dikes cutting the iron-formation or is not obviously thicker above dikes than elsewhere in an ore body. Some of the first mining of “soft” ore was above metadiabase sills (Anderson, 1968, p. 512), but no ore from that type of environment was being mined in 1968. The ore in general was high in iron and low in sulfur and phosphorus. The ore extends to the deepest part of the Negaunee Iron-Formation in the Negaunee quadrangle. Workings of the Negaunee mine extend from near the surface, at an altitude of about 1,000 feet, down the plunge of the structure to a point about 150 feet below sea level some 5,000 feet to the west. Deep workings in the Mather ECONOMIC mine B are 1,500 feet below sea level. The contact between the Siamo Slate and the Negaunee Iron-For- mation limits the bottom extent of the ore bodies. ORIGIN The “soft” iron ore bodies are enriched parts of the Negaunee Iron-Formation. Enrichment was the result of at least two processes: the oxidation and redistribu- tion of iron minerals and the leaching and removal of chert and other noniron minerals. It is assumed that the base of the iron-formation, where the ore bodies are now found, originally was compositionally similar to unoxidized, nonleached iron-formation. Downward percolating solutions were important in concentrating the ore minerals. Structural traps for the solutions are the troughlike features along the Negaunee-Siamo contact and intersections between dikes and bedding. Marsden (1968, p. 504) suggested that some magmatic or hydrothermal assistance to induce circulation is required, although most of the solutions were oxygenated surface waters that moved downward through the formation. The progressive enrichment in the iron-formation can be seen in drill core from the Mather mine B. The upper parts of the formation consist of alternations of chert and iron-rich laminae. With increasing depth, the GEOLOGY 49 chert beds become corroded and replaced by iron min- erals. Within the ore bodies, near the base of the iron- formation, the rock is a mass of layered hematite and goethite that contains only a few wisps of chert. Immense quantities of silica and other noniron min- erals were removed by the iron-enrichment process. There is no clue as to the final destination of these materials. BASE-NIETAL SULFIDE DEPOSITS Base-metal sulfides including chalcopyrite, galena, sphalerite, and, rarely, tetrahedrite occur in quartz and quartz-carbonate veins, in fractures in the Dead River pluton, in the chert-magnetite iron-formation north of Dead River, and in conglomerate near the base of the Reany Creek Formation (table 14). The little explora- tion work done suggests that these occurrences cannot be considered likely sources for base metals. VEIN DEPOSITS The veins that contain the base-metal sulfides may be all quartz with only scattered crystals of carbonate, mixtures of quartz and carbonate, or predominantly carbonate with minor amounts of quartz. The carbon- ate is light-tan to buff ferruginous dolomite or ankerite. Chlorite is common in some veins. The veins in sec. 26, T. 48 N., R. 26 W., contain pink to red albite. TABLE 14.—Sulfide-bearing quartz or quartz-carbonate veins in the N egaunee quadrangle [Distances measured from southeast corner of section] Sulfides Thickness (in addition Location Orientation (ft) to pyrite) Development 300 ft west, 4,200 ft north, N. 70" E., 5—6 Chalcopyrite ................ Short adit, shaft (depth?) . sec. 26, T. 48 N., R. 26 W. vertical. Foundations of old building. No sign of production. 2,250 ft west, 3,500 ft north, N. 55° E., .2—.8 ........ do ........................ None. Possibly this is an sec. 26, T. 48 N., R. 26 W. vertical. extension of that above. 950 ft west, 2,600 ft north, Due west, 2—4 ........ do ........................ None. sec. 26, T. 48 N., R. 26 W. vertical. 3,000 ft west, 2,650 ft north, N. 75° W., 1.5—2 Chalcopyrite, Shaft about 5 ft deep. sec. 28, T. 28 N., R. 26 W. 80° S. tetrahedrite. 4,200 ft west, 3,900 ft north, N. 80° W., 10 Chalcopyrite ................ Several shallow prospect pits. sec. 30, T. 48 N., R. 26 W. vertical. 1,600 ft west, 4,200 ft north, N. 85° W., .1—.2 ........ do ........................ None. sec. 36, T. 48 N., R. 27 W. . 85° S. 2,000 ft west, 4,400 ft north, (?) .3 Chalcopyrite, Do. sec. 22, T. 48 N., R. 26 W. galena. 700 ft west, 1,100 ft north, N. 50° E., 1—1.5 Chalcopyrite Do. sec. 14, T. 48 N., R. 26 W. 60° S. tetrahedrite(?) . 1,300 ft west, 1,000 ft north, N. 50° W., Very thin. Chalcopyrite ................ In face of quarry for railroad sec. 14, T. 48 N., R. 26 W. 80° N. Adjacent to thin ballast. . . mafic dike. 4,400 ft west, 1,300 ft north, N. 20" E , .8—1.2 Chalcopyrite, Several shallow prospect pits. sec. 4, T. 48 N., R. 26 W. 85° S sphalerite, galena. 1,400 ft west, 1,800 ft north, N. 30° E , 2 Chalcopyrite ................ None. sec. 33, T. 49 N., R. 26 W. 30° S 2,200 ft west, 550 ft north, N. 60° W , 1.8—2 ........ do ........................ None. Vein occurs within sec. 10, T. 48 N., R. 26 W vertical diabase. 1,200 ft west, 1,650 ft north, (?) 1—.2 ........ do ........................ None. sec. 3, T. 48 N., R. 26 W. 800 ft west, 2,650 ft north, (?) .1—.2 ........ do ........................ Do. sec. 3, T. 48 N., R, 26 W. 50 NEGAUNEE QUADRANGLE, MARQUETTE COUNTY, MICHIGAN The sulfides are randomly distributed and constitute no more than a few percent of the vein material. Chalco- pyrite grains and aggregates are as much as three- fourths inch wide. Sphalerite encloses shreds of, and appears to be younger than, the chalcopyrite. Galena is rare but forms cubes as much as one-fourth inch across, intergrown with sphalerite in the vein in sec. 4. Tiny grains of tetrahedrite occur in two of the veins but represent only a small percentage of the copper minerals. The thickest and most promising sulfide-bearing vein, in the NW% sec. 30, T. 48 N., R. 26 W., consists of several quartz stringers in a zone 10—15 feet thick, extending along strike for several hundred feet. Some tourmaline is in the vein. A shallow shaft explored part of the vein and found some Chalcopyrite-rich rock. Dump material appears to contain less than 1 percent copper. Veins appear to be randomly distributed, although the base-metal occurrences south of Dead River may be spatially related to diabase or metadiabase (Pufiett, 1966, p. 1311). Some are near magnetic lows that com- monly are crossed by diabase dikes. A chalcopyrite- bearing quartz vein, 2,200 feet west and 550 feet north of the SE. cor. sec. 10, T. 48 N., R. 26 W., is within a diabase dike. At 950 feet west and 2,600 feet north of the SE. cor. sec. 26, T. 48 N., R. 26 W., sulfides are adja- cent to metadiabase. The chert-magnetite iron-forma- tion north of Dead River contains scattered sulfides near its contact with coarse-grained diabase. Diabase or metadiabase is not exposed near the other base metal sulfide occurrences. There are miles of diabase or meta- diabase elsewhere in the quadrangle, but no base-metal sulfides are known to be associated with them. SULFIDES IN THE CHERT-MAGNETITE IRON-FORMATION Base-metal sulfides are known in widely separated areas in the chert-magnetite iron-formation bounded by coarse-grained metadiabase north of Dead River. After the iron-formation was delineated magnetically, soil samples were taken at 100-foot intervals along it and analyzed for copper, lead, and zinc. Generally, they contain less than 30 ppm (parts per million) copper, less than 25 ppm lead, and 25 ppm or less zinc. A sample from 1,100 feet west and 2,100 feet north of the SE. cor. sec. 6, T. 48 N ., R. 26 W. (pl. 3) contained 150 ppm copper, 125 ppm lead, and 500 ppm zinc; another sample from 1,850 feet west and 1,200 feet south of the NE. cor. sec. 1, T. 48 N., R. 27 W., contained 60 ppm copper, less than 25 ppm lead, and 125 ppm zinc. Blocks of iron-stained cherty rock containing much fine- grained purplish-brown sphalerite were found near shallow prospect pits about 100 feet upslope from the location of the sample containing high values in the SE 14 sec. 6. A sample of the mineralized rock contains 700 ppm copper and 10,000 ppm zinc as determined by semiquantitative spectrographic analysis. Soil samples from east and west of the sample with high values, in the same relative position to the prospect pits, did not contain anomalous amounts of metal. BASE METALS IN CARP RIVER FALLS SHEAR ZONE Copper and zinc are present in and near the Carp River Falls shear zone (table 12). Copper is most abundant in rocks richest in CO2. High values for zinc are not necessarily in the samples having high copper, but zinc values do appear to correlate with the FeO content. Chalcopyrite has been found in chloritic slate near the shear zone, but sphalerite has not been recognized. GOLD Samples from nine widely separated localities in the quadrangle were analyzed for gold. Three were of cop- per-bearing veins, two were of pyrite—rich gneiss, and four were of quartz veins or pyrite concentrations in the Lighthouse Point Member of the Mona Schist. A sample of an 8-inch-thick quartz vein from 350 feet west and 600 feet south of the NE. cor. sec. 32, T. 49 N., R. 25 W., contained 1.4 ppm gold. The other samples contained either less than 0.02 ounces gold per ton or less than 0.06 ppm gold. REFERENCES CITED Anderson, G. J ., 1968, The Marquette district, in Ore deposits in the United States, 1933—1967 (Graton-Sales volume): New York, Am. Inst. Mining Metal]. Petroleum Engineers, v. 1, p. 507—517. Bailey, S. W., and Tyler, S. A., 1960, Clay minerals associated with the Lake Superior iron ores: Econ. Geology, v. 55, p. 150—175. Bayley, R. W., Dutton, C. E., and Lamey, C. A., 1966, Geology of the Menominee iron-bearing district, Dickinson County, Michigan, and Florence and Marinette Counties, Wiscon- sin: U.S. Geol. Survey Prof. Paper 513, 96 p. Brooks, T. B., 1873, Iron-bearing rocks (economic), pt. 1, of Upper Peninsula of Michigan (1869—73): Michigan Geol. Survey, v. 1, 319 p. Cannon, W. F., and Gair, J. E., 1970, A revision of stratigraphic nomenclature for middle Precambrian rocks in northern Michigan: Geol. Soc. America Bull., v. 81, p. 2843—2846. Case, J. E., and Gair, J. E., 1965, Aeromagnetic map of parts of Marquette, Dickinson, Baraga, Alger, and Schoolcraft Counties, Michigan, and its geologic interpretation: U.S. Geol. Survey Geophys. Inv. Map GP—467, scale 1262,000. Dapples, E. C., 1959, The behavior of silica in diagenesis, in Ireland, H. A., ed., Silica in sediments—a symposium: Soc. Econ. Paleontologists and Mineralogists Spec. Pub. 7, p. 36—54. Engel, Theodore, J r., 1954, The stratigraphy and petrography of the Holyoke meta-sediments of the Dead River Basin, Marquette County, Michigan: Michigan State Coll, Agr. and App]. Sci., M.S. thesis. REFERENCES CITED 51 Fritts, C. E., 1964, Stratigraphy of Animikie (formerly Huron- ian) rocks east of Teal Lake, Michigan [abs.]: Inst. Lake Superior Geology, 10th Ann. Mtg., Ishpeming, Mich., p. 5—6. Gait, J. E., and Thaden, R. E., 1968, Geology of the Marquette and Sands quadrangles, Marquette County, Michigan: US. Geol. Survey Prof. Paper 397, 77 p. Gair, J. E., and Wier, K. L., 1956, Geology of the Kiernan quadrangle, Iron County, Michigan: US. Geol. Survey Bull. 1044, 88 p. Gruner, J. W., 1946, Dickite and chromium silicate in the iron ores of the Marquette and Gogebic ranges, Michigan [abs.]: Am. Mineralogist, v. 31, p. 195. Jackson, K. C., 1950, A heavy mineral study of the Ajibik and Mesnard quartzites of Marquette County, Michigan: Michigan Coll. Mining and Technology, M.S. thesis. James, H. L., 1955, Zones of regional metamorphism in the Precambrian of northern Michigan: Geol. Soc. America Bull., v. 66, no. 12, pt. 1, p. 1455—1488. 1958, Stratigraphy of pre-Keweenawan rocks in parts of northern Michigan: US. Geol. Survey Prof. Paper 314—C, p. 27—44. » 1966, Chemistry of the iron-rich sedimentary rocks, in Data of geochemistry [6th ed.]: U.S. Geol. Survey Prof. Paper 440—W, 61 p. McKee, E. E., and Weir, G. W., 1953, Terminology for stratifi- cation and cross-stratification in sedimentary rocks: Geol. Soc. America Bull., v. 64, p. 381—389. Marsden, R. W., 1968, Geology of the iron ores of the Lake Superior region in the United States, in Ore deposits in the United States, 1933—1967 (Graton-Sales volume) : New York, Am. Inst. Mining Metall. Petroleum Engineers, v. 1, p. 489—506. Maxwell, J. C., 1962, Origin of slaty and fracture cleavage in the Delaware Water Gap area, New Jersey and Pennsyl- vania, in Engel, A. E. J., James, H. L., and Leonard, B. F., eds., Petrologic studies—a volume in honor of A. F. Bud- dington: Geol. Soc. America, p. 281—311. Nanz, R. H., J r., 1953, Chemical composition of pre-Cambrian slates with notes on the geochemical evolution of lutites: J our. Geology, v. 61, no. 1, p. 51—64. Nockolds, S. R., 1954, Average chemical composition of some igneous rocks: Geol. Soc. America Bull., v. 65, no. 10, p. 1007—1032. Powell, C. McA., 1969, Intrusive sandstone dykes in the Siamo Slate near Negaunee, Michigan: Geol. Soc. America Bull., v. 80, no. 12, p. 2585—2594. Pufi'ett, W. P., 1966, Occurrences of base metals south of Dead River, Negaunee quadrangle, Marquette County, Michigan [abs.]: Econ. Geology, v. 61, no. 7, p. 1310—1311. 1969, The Reany Creek Formation Marquette County, Michigan: US. Geol. Survey Bull. 1274—F, 25 p. Rittmann, Alfred, 1952, Nomenclature of volcanic rocks: Bull. Volcanol., ser. 2, v. 12, p. 75—102. Rominger, C. L., 1881, Marquette iron region, Pt. 1, in Upper Peninsula 1878—1880: Michigan Geol. Survey [Repts.], v. 4, p. 1—154. Shrock, R. R., 1948, Sequence in layered rocks: New York, McGraw-Hill Book Co., 507 p. Stose, A. I. J ., and Stose, G. W., 1946, Geology of Carroll and Frederick Counties [Md.]: Maryland Dept. Geology, Mines, and Water Resources, Carroll and Frederick Coun- ties Rept., p. 11—131. Stuart, W. T., Brown, E. A., Rhodehamel, E. C., 1954, Ground- water investigations of the Marquette iron-mining district, Michigan: Michigan Geol. Survey Tech. Rept. 3, 92 p. Tyler, S. A., and Twenhofel, W. H., 1952, Sedimentation and stratigraphy of the Huronian of Upper Michigan, pts, 1 and 2: Am. Jour. Sci., v. 250, no. 1, p. 1—27, and no. 2, p. 118—151. Van Hise, C. R., and Bayley, W. S., 1895, Preliminary report on the Marquette iron-bearing district of Michigan: US. Geol. Survey, 15th Ann. Rept., p. 477—650. 1897, The Marquette iron-bearing district of Michigan: US. Geol. Survey Mon. 28, 608 p., atlas. Van Hise, C. R., and Leith, C. K., 1911, The geology of the Lake Superior region: U.S. Geol. Survey Mon. 52, 641 p. Williams, G. H., 1890, The greenstone schist areas of the Menominee and Marquette regions of Michigan: US. Geol. Survey Bull. 62, 238 p. ' Page Accessibility .................................................... 2 Adams mine ................ .. 48 Aeromagnetic survey 46 Agglomerate . 5 Ajibik Quartzite 24, 25, 28, 42. 45 Algal structures 28 Amphibolite . 15, 19, 20, 45 Argillite . .. .97, 47 Arkose .. 23, 24 Artis, Lowell, analyst .................. 11, 18, 20, 21, 29 38, 40, 44 Augen zone, felsic .............................. 15 Baraga Group .. 34 Barassa mine . 48 Barium ............. 18 Base-metal sulfides . .. 49 Bismark Creek ...... 15, 17, 19 Botts, S. D., analyst 11, 18, 20, 21, 29 38, 40, 44 Breccia .............................................................. 27 Cambria-Jackson ore body . 48 Carp River .............................. 9, 26, 32, 41, 47 Carp River Falls shear zone. 9, 10, 17, 19, 4.9, 50 Chart ............................................. 32. 33. 47 Chloe, Gillison, analyst ..... . 11, 18, 20, 21, 29 38, 40, 44 Chocolay Group .............................................. 22 Club Lake ..................... 15 . 4, 5, 11, 15. 17 40, 41, 43, 45 Compeau Creek Gneiss . Conglomerate ........ 23,24, 32. .78 Copper ........... 12, 18, 28, 43, 50 Corals ................................... 4 Crandell, W. B., analyst .. .. 11, 21, 29, 40 Dead River .......... Dead River basin Dead River Canyon .. Dead River pluton . 5, 13, 19, 47 4, 10 ...................... 13, 41 5, 10, 13, 20, 34, 36 40, 45, 46, 49 .................. 41, 4.9 14, 17, 23, 34. 36 Dead River shear zone ......... Dead River Storage Basin .. Diabase ............. 41 Diastrophism 45 Dickite .......... .. 34 Bikes . 5, 9. 18, 22, .92, 39, 41 Diorite . ........................ 5, 20, 21 Dolomite .......................................................... 25 Eagle Mills syncline .. . ............... 47 Economic geology ....... Elmore, P. L. D., analyst .. 11, 18, 20. 21, 29 38, 40. 44 Enchantment Lake Formation. 13, 17, £4, 41, 43 Faults ......... Felsic rock . Fennelly, E. J., analyst Folds ..... Foliation . . Frost, I. 0., analyst ........................................ 38 . 16. 18, 23, 36, 4S INDEX [Italic page numbers indicate major references] Page Geologic history .............................................. 45 Geologic setting 4 Glacial features 4, 46 Glenn, J. H., analyst 11, 18, 20, 21, 29, 38, 40 Gneiss . 17 Gold Goose Lake Member .1 Granodiorite ................ Granodiorite porphyry £0 Granule bed £9 Graywacke ...... 28, 31. .96, .97 Greenstone .................. 5, 10, 12, 17, 19, 40, 45, 46 Ground magnetic surveys ............................ 46 Harris, J. L., analyst .......... 11, 20, 21, 38, 40, 44 Hematite .................... 18, 19, 22, 25, 28, 29, 33, 37 Hemlock Formation 19 Hoist Dam .................. 35, 36 Hornblende diorite .............. 91 Huffman, Claude, analyst .. 44 Introduction .................................................... z Intrusive rocks . 5 Iron-formation .97, 46 Iron ore ........ 2. 33, 47 Jackson fault .................................................. 41, 48 Kaolinite ............................. 34 Kelsey, James, analyst 18, 20, 29, 40 Kitchi Schist ............................. 5, 17. 45 Kona Dolomite 25, 30, 42, 45 Kona Hills 25 Lead .................................................................. 18, 50 Lighthouse Point Member ................ 5, 10, 15, 39, 41, 44, 45 Lighthouse Point Member syncline .......... 4.? Location ............................................ 9 Lower Member of Mona Schist .. 11, 17, 45 Maas-Negaunee mine ....... 41 McClure Storage Basin 34, 36, 47 Mafic intrusive rocks . 89 Magnetic surveys .. .. 46 Magnetite ..... 32, 33, 35, 37, 47 Manganese Marquette Range Supergroup . Marquette syncline ......... Marquette synclinorium .. Maas mine ......... Mather mine B .. Menominee Group ........ Mensik. J. D., analyst .. 44 Mesnard Quartzite ....... . 24, 30, 45 Metadiabase .1 .. .79 Metagabbro .. .. .99 Metamorphism 5, 8, 11, 17, 33, 46 Metasiltite .......................... 86 Michigamme Slate 13, 20, .94, 41, 44 Midway Creek ................................................ 21 Page Mona Schist 10, 15, 19, 34, 40, 43, 45 Morgan furnace . ................................. 25, 47 Nealy Creek Member ...... 10, 18, 15, 20, 35, 44, 45 Negaunee Cemetery ................... 17, 24, 25 Negaunee Iron-Formation .92, 40, 46. 48 Negaunee mine .. 48 Normal faults ..... 45 Picket Lake shear zone .. 45 Pillows ..... 11, 12, 15, 17 Porphyritic syenlte 22 Precambrian rocks, lower . 5 middle ......... 2.9 Previous work 4 Pyrite ................. 9, 13. 15, 18. 32, 47 Pyroclastic rock ......... 5 Quartz porphyry . ....................................... 9 Quartzite ............... 24, 25, 26, 28, 29, 31, 32, .96 Quinnesec Formation 19, 20 Reany Creek .................................................... 4 Reany Creek Formation 20, 22, 41, 44, 45, 47, 49 Republic mine . 2 Reverse faults .. . 45 Rhyolite tufl' member, sheared 11, 14, 34, 44, 46 Schist ..... . 8, 10, 13 Sericite .. 9 Shear zones . 4.! Sheared rhyolite tutf member 11, 14, 34, 44, 46 Siamo Slate ............................ 30, .91, 40, 41, 46, 48 Siderite 33 Silicification 5, 17 Sills ............... 15, 39 Slate .............. 13, 23, 24, 25, 2’7, 31, 33, 35, .96, 46 Smith, Hezekiah, analyst 11, 18, 21, 29, 38, 40,44 Stratigraphic section, Enchantment Lake Formation 24 Kona Dolomite .............. 25 Maas mine area .. 4 Stratigraphy 4 Structure Sulfides Syenite Taylor, Dennis, analyst .......................... 11, 38, 44 Teal Lake . 24, 25, 28, 30, 42. 45, 46 Tooker, E. W., analyst . Tufl‘ .................................... Volcanic rocks Volcanism Wewe Slate ........................................ 25, 28, 31, 45 Willow Creek shear zone 45 Work methods .................... 3 Zinc .................................................................... 50 53 ‘ik’ US. GOVERNMENT PRINTING OFFICE: 1974 0—535-181 UNITED STATES DEPARTMENT OF THE INTERIOR GEOLOGICAL SURVEY PREPARED IN COOPERATION WITI—I "TI-IE GEOLOGICAL SURVEY DIVISION OF THE MICHIGAN DEPARTMENT OF NATURAL RESOURCES PROFESSIONAL PAPER 788 PLATE 5 35 32 33 87° ’ 87°37’30” 36 31 31 32 I 5 4 1 6 5 Soil sample Cu Pb Zn 60 ppm <25 ppm 125 ppm EXPLANATION MAGNETIC INTENSITY, IN GAMMAS """ <1 ,000 LOGO—2,000 2,000—5 ,000 5 ,000—10,000 >10,000 ' IIII MAGNETOMETER STATION — Values Soil sample given for maximum and minimum ....... — Cu Pb Zn stations only ..':-:-I':':":': 150ppm 125ppm 500ppm ' T.48 N. ----- 46°35’ — HI L9 ‘11 Z 0 < z 01 D < [I < o 3 < O :3 g 0 U) LIJ LL] LLI w E Z < 3 o 5.) Lu LIJ Z Z 87°37’ 30” T.48 N. GROUND-MAGNETIC MAP 1000 O 1 I | I | I 500 I I OF PART 0 1000 2000 3000 FEET I L J 0 500 METERS J F AREA NORTH OF DEAD RIVER BASIN, NEGAUNEE QUADRANGLE, MARQUETTE COUNTY, MICHIGAN IVagnetometer survey by W. P. Puffett, 1966—67 PROFESSIONAL PAPER 788 PREPARED IN COOPERATION WITH PLATE 2 THE GEOLOGICAL SURVEY DIVISION OF THE MICHIGAN DEPARTMENT OF NATURAL RESOURCES 1/ UNITED STATES DEPARTMENT OF THE INTERIOR GEOLOGICAL SURVEY ,,M"“‘“””mm m, made” N» VMJMNWMMWM / Wmmwwn WW, wwwmw‘ _ 5' ”K W W /:V‘ MW} xMW, ”Mk”; ' ' 3 ,.M~w_mw www- i} ,WMWMM ,» V‘M M mkx ’ ; 5a Maw» \ f k. f," g i 3 f fi & E 3 fl W3 fl flfiflfi aw iflfiflm a l Geology mapped 1965—68 Base from 0.8. Geological Survey 124,000, 1952 O _1_000 2000_/ 3000 FEET 500 METERS 5W CORRELATION 0F MAP UNITS DESCRIPTION OF MAP UNITS ms I ms $AMOSLATE -m® MONASCHBT ma . ma ‘ AJIBIK QUARTZITE ‘ OUTCROP OR OUTCROP AREA 1500' 1500 rims _ mas 1 SERICITIC SLATE AND GRAYWACKE Unconformity A Contact Unconformity U 815 . D Fault, showing dip # Dashed where approximately located; queried h . i . . . hr . QUARTZITE, WHITE VITREOUS w 'ere uncertain U, upthrown Side, D, downt . own Side 1000' 1000, —-+—- Syncline — Showmg trace of ax1al plane and bearing and plunge of K D l ‘t L FERRUGINOUS SLATE, GRAY AND PURPLE — Contains thin axis. Dashed where approximately located ona O omi e PRECAMBRIAN quartzite beds ._ ———9——— Overturned syncline - Showing trace of axial plane mkx CHERT BRECCIA # Locally dark purple, ferruginous . . . - Strike and dip of beds * QUARTZITE, WHITE VITREOUS ~ Ripple marked and cross- __L6_O Inclined, degree of dip given where known , bedded Inm2f SLATE,PURPLHHLGRAY,FERRIKHNOUS T$' ovanunai 1500 150d mm . . . mkl QUARTZITE, REDDISH—GRAY, SERICITIC — Grades upward Mm“, f9” Showmg plunge me] into slate -—)-O45 Anticline Unconformity mm MESNARD QUARTZITE {—940 Syncline mos mel ENCHANTMENT LAKE FORMATION x Prospect [SI Shaft at surface 1000' 1000' Unconformity GEOLOGIC MAP AND SECTIONS OF THE NORTH LIMB OF THE EAGLE MILLS SYNCLINE. NEGAUNEE QUADRANGLE, MARQUETTE COUNTY, MICHIGAN PREPARED IN COOPERATION WITH UNITED STATES DEPARTMENT OF THE INTERIOR THE GEOLOGICAL SURVEY DIVISION OF THE GEOLOGICAL SURVEY MICHIGAN DEPARTMENT OF NATURAL RESOURCES 8733730" 87°30 4 6’37’30" 4633780" 4.73 w R. 27 w, R. 26 w. ASAMMIE. . . . . , 1 ,. . l4“) 5163000m.|\: “7162 ”mm T 49 N ’ T, 48 N. 3437.7 H SEN (NECAUNE;E “ {MAQQUETTEJ 3577 III SW ‘ r / ‘\ , \ ._\ \ \ MAROU ELIE CO \ l4/5." , ’+ . AIRPORT MARQUETTE 5.5 MI. BM 1410)» BROOKTON CORNERS 2.3 MI. ores oe Lakéx" \ MAIRQUETTE 5.8 MI, TI m 4Icz.mmseupe; ’12 MI. . ‘ISHPEMING 3 Mi. ' "Pioneer: \- . i'PEHe-t Plaht .. 50 3 _‘ T.48N. ' . . 7" ‘ ‘ " ” ,' . , '5 ’ ,. ,“ . . ‘-. ». ' ,' ‘. ‘w ~,..A 3T.48N. TAMI. ~ Q i/ Edgy/V * , - , .. / j , . — ' m , " ; ' ' * . . ' ‘ x x,. 7.47m. _. 3500mm.” I 3 9,; 7.5.1: . ,. /’ ,/'!m5f‘*‘{ 46°30” 46°30’ 7 » CORRELATION OF MAP UNITS Keweenawan Series Baraga ’ Group Menominee ms I Group T .33; >Marquette Range ma Supergroup Unconformity >Cgocolay mm roup mel mrc Unconformity ccg Mona Schist m1 msg ’ chl DESCRIPTION OF MAP UNITS DIABASE —- Massive black dike rock with diabasic texture; weath- ers to dull brown. Not metamorphosed. Can be traced between outcrops in many areas, as it creates a sharp negative magnetic anomaly MICHIGAMME SLATE — Thin-bedded graywacke with local carbon- rich beds, and sericitic, chloritic, carbonaceous, and pyritic meta- siltite and slate. Unconformable against Dead River pluton and Nealy Creek Member of Mona Schist. Thin magnetite argillite unit causes magnetic anomaly in lower part of formation. Greater than 5,000 feet thick Conglomerate composed of angular chert clasts in coarse- grained graywacke matrix Chert-goethite-hematite iron-formation Conglomeratic, sericitic, and arkosic quartzite at base of for- mation NEGAUNEE IRON-FORMATION — Dark-brown thin—bedded chert-hematite-goethite iron-formation. Red and gray jaspi- lite near west border of quadrangle. Poorly exposed. Contai’ns important iron ore deposits. Mined areas shown by ruled pat— tern. Greater than 2,000 feet thick SIAMO SLATE -— Dark-gray thin-bedded slate, weathers gray to brown. Beds of massive graywacke; some containing abundant mud chips; as much as 10 feet thick in some exposures. Rusty- brown-weathering seams mark carbonate-rich beds. Magnetic unit in lower part (Goose Lake Member?) not exposed but indicated by magnetic anomaly. Conglomerate beds in upper part. Clastic dikes conspicuous in some exposures. Upper con- tact gradational. 1,500 to 2,000 feet thick AJIBIK QUARTZITE — Gray to white vitreous quartzite; seri- citic slate and graywacke near base. Locally iron stained pink to purple. Ripple marks and crossbeds common. Marker bed containing pink chert granules near base. Overlaps older forma- tions near Teal Lake. Averaging 150 feet thick Unconformity mk mel mrc mnc ml ,-_,n,_g‘“j ch- Unconformity WEWE SLATE — Not exposed in quadrangle; projected from Mar- quette quadrangle, where it is gray and greenish-gray laminated and massive sericitic-chlorite-quartz slate KONA DOLOMITE — Pinkish-gray fine- to medium-grained crys- talline dolomite, locally cherty, and thin beds of purplish-gray argillite and argillaceous dolomite near Morgan Pond. Light- gray to white quartzite, interbedded with purplish-gray ferru- ginous, sericitic slate and chert breccia west of Negaunee Cem- etery. 800 to 1,200 feet thick MESNARD QUARTZITE — Gray to pink vitreous thick-bedded quartzite; lenticular; ripple marks common. Maximum thick- ness 200 feet ENCHANTMENT LAKE FORMATION — Conglomerate, gray- wacke, arkose, sericitic quartzite, and sericitic slate. Uncon- formably overlies Mona Schist on north limb of Eagle Mills syncline. Lenticular; maximum thickness 150 feet REANY CREEK FORMATION — Conglomerate, chloritic slate, some containing widely dispersed boulders of granite and intra- clasts of arkose, fine-grained gray feldspathic quartzite, and pink arkose. In part of glacial origin. Unconformably overlies Mona Schist north of Dead River. Age in doubt, might be pre- Marquette Range Supergroup. 1,500 to 3,500 feet thick FELSIC METAVOLCANIC ROCK — Gray porphyritic felsic rock, quartz phenocrysts common; amphibole and pyroxene phe— nocrystsrare.Weathers light gray to near white. Cataclastically deformed; locally mylonitized. Mainly volcanic rock, but in- cludes some possible intrusive rock. Some crystal tuff. Occurs as dikes or sills(?) and as irregular-shaped bodies of uncertain orientation. Common along boundaries of thick coarse-grained metadiabase. Indicated only by (f) where outcrops are too small to map. Hornblende-syenite shown by (hs) METAGABBRO — Dark-gray to green medium-grained rock. Min- eralogically identical to the metadiabase (md). Forms two in- trusive bodies in northwest part of quadrangle DEAD RIVER PLUTON —- Massive nonfoliated porphyritic rock; pink to red on weathered surface. Age uncertain, possibly late early Precambrian Hornblende diorite Porphyritic syenite containing large and abundant perthite phe- nocrysts Granodiorite porphyry COMPEAU CREEK GNEISS — Light-gray to light-pink foliated medium- to coarse-grained tonalitic gneiss. Stippling indicates zone of intense silicification LIGHTHOUSE POINT MEMBER — Dark-green fine-grained lay- ered amphibolite; near contact with Compeau Creek Gneiss less distinctly layered and streaked with light-gray felsic material. At least 8,500 feet thick Felsic augen zone. SHEARED RHYOLITE TUFF MEMBER — Pink to greenish-gray strongly sheared rhyolitic rock; glassy quartz phenocrysts com- mon. Bright green wafers of clay, probably altered lapilli, are conspicuous locally. 1,300 to 3,000 feet thick NEALY CREEK MEMBER — Dark-gray to gray-green quartz- feldspar-sericite-chlorite schist. Biotite present near Dead River pluton. Possibly a metamorphosed tuff of intermediate compo- sition. 2,000 to 3,000 feet thick LOWER MEMBER — Dark-green massive metabasalt. Large pillow structures common. About 10,000 feet thick UNDIFFERENTIATED GREENSTONE -— South of the Carp Riv- er Falls shear zone unit generally sheared and includes sheared felsic rock; possibly Kitchi Schist in part. North of Dead River unit mainly fine-grained massive metabasalt containing abun- dant intrusions of felsic rock and coamegrained metadiabase. Chert beds, locally contain magnetite, sphalerite, and chaloopy- rite } Upper Precambrian k; > Middle Precambrian > Lower Precambrian PROFESSIONAL PAPER 788 PLATE 1 rPRECAMBRIAN J Age relations and stratigraphic posrtion uncertain ?_ __I__. Contact — Queried where uncertain; dotted where concealed. Out- crops indicate degree of accuracy of location Fault ~ Dashed where approximately located; queried where uncertain; dotted where concealed. U, upthrown side; D, downthrown side Shear zone ~— Not shown where a known fault is shown Vein — Commonly quartz-carbonate Syncline — Showing trace of axial plane and direction of plunge. Dashed where approximately located; dottcd where concealed -—U— Overturned syncline — Showing trace of axial plane 20 Bearing and direction of plunge of minor anticline — Value of plunge given where known Strike and dip of beds —(p)indicates strike and dip of pillow struc- tures in greenstone 80 “‘— Inclined — Value of dip given where known .22; Overturned Strike and dip of foliation — Includes layering in Lighthouse Point 75 Member of Mona Schist 4— Inclined — Value of dip given where known —9— Vertical Strike and dip of cleavage £1 Inclined »—+ Vertical —/-5— Strike and dip of joint — Value given where known Bearing and plunge of lineation 19* Plunging o Vertical X Prospect pit Gravel pit X Active X Inactive n Shaft o Diamond-drill hole Py Abundant pyrite Pb, Zn, Cu Areas containing lead, zinc, or copper sulfides ._ Crest of magnetic anomaly from ground-magnetometer survey 89 88 90° 1 87° 48° I | LE \ —’ _48 I EOYALES 0 \L \ 86° / f ‘4 1) € 4> \ 85” 47° v z o \\ 84° W 4, ' 47° ‘- ” \9 . % ii, arquette \ Sm1 It Ste Maine Ironvmod 32 Area of 83° 46, ‘ this report {W46 x) U “x 90. IronsglIountain hebpygan\ \ L6 . 1. . 4 ¢~ \ M229“ oi... A .\ W» . k; y" c: °Cad1 lac ‘3’ .27 A m 44.. if 044. O J N Saginaw‘ Z M - k 0 . ’3’ 43, fl ‘5 eg n nFlint I 43. Grand - 5 1:“, Port Huron R: LANSI c g I DETROIT , a vazoo° . V :I A“ “am" i ~ we---_1s_I_g’% 87° 8 6° 85" 84° 83" INDEX MAP OF MICHIGAN <0! . 40 so MILES L4—J—l—I_L_|_I_a _ . .I... 1 .. . » ~ ' .. ' ’ I ‘ «,3 «A,» 11,. E 43' A 2 HARVEY 13 MI. I o . ,. ,, R, 27 w_ <, PALMER {VIA MICH 35) 4.9 MI. ,_ (ggé/Idfighm ...OU o‘ea . EEI' I459 MARauer 1h ,7 ML dmemE 87°30’ 87 37 30 . VGW’NN (WA ”"7” J5) ’2 M" ' Geology by C E. Fritts 1962—63 KITCHI SCHIST —— Dark-green fragmental metavolcanic rocks of Base from U-S- Geo'og'cal Survey. 1953 and W. p, Puffett 19621—68 latite to rhyodacite composition. Large, often rounded, pyro- * SCALE 1'24 000 ' . . . . MN 1 1 O. 1 MILE clasts m tuffaceous matrix. Crystal tuff and crystal-lithic tuff ' ‘GN :4 . 7 . - . common. Possibly contemporaneous with lower member of 1029I H b3 1090 2090 30_00 4090 5000 609° 70100 FEET Mona Schist. Maximum thickness is 4,500 feet —M|.L—S 0.25, La }__' ,_,5 }_I '_i 0 } K'LOMETER I [ma METADIABASE — Greenish-gray to dark-green rock. Fine grained 9 7 MILS ' to coarsely porphyritic. Diabasic texture conspicuous on weath- g ered surfaces in some exposures. Forms dikes and sill-like bodies. 2' In Some are pre-Marquette Range Supergroup in age 2 U, um GRID AND 1953 MAGNETIC :CEJETIH :9: fl Outerop or outcrop area jl-IJ DECLINATION AT CENTER OF S LIJ & z 32 2% E a a g 0: d “w e e e as m a 52 a I e . s .. A' A 50: L» ma ‘3‘") E- md 3&3 msg ch f 1. 53 E 1600' 51“- EU) 6 md L8) ml md mnc/ / ,\ pg] PS IS mrc kd mrc kd ‘\ »\. I . m m0 mlp 1600: ~ . / \ q . . , / . , ‘ . . ,, I; Q, I I ,, 1200' 1200 ‘4 II , x i p5 / [I I \ \ 800’ :2 800’ 400' 400' SEA LEVEL SEA LEVEL 400, 400' e s W Q) a w a .9 a w 0 3 °> : , B 934 w a ma § B 1600, EAGLE MILLSg m1 Pd (3» E 9% § :1600' SYNC LINE 3 m‘” 3 U3 9: mlp CCS 1200' .. —1200' , . — 800’ 800 I 400, I II : 400' SEA LEVEL I ‘SEA LEVEL l 400' 400' BEDROCK GEOLOGIC MAP AND SECTIONS OF THE NEGAUNEE QUADRANGLE, MARQUETTE COUNTY, MICHIGAN UNITED STATES DEPARTMENT OF THE INTERIOR GEOLOGICAL SURVEY PREPARED IN COOPERATION WITH PROFESSIONAL PAPER 788 THE GEOLOEICAL SURVEY DIVISION OF THE PLATE 4 MICHIGAN DEPXRTMENT OF NATURAL RESOURCES EXPLANATION MAFIC DIKES AND SILLS BOO FEET “SOFT” IRON-ORE BODY NEGAUNEE IRON-FORMATION SIAMO SLATE FAULT — Showing relative movement. Dashed where approximately located 2;.“ «“5 Compiled in 1967 from the Cleveland-Cliffs Iron Company records. Geologic details shown only near ore bodies. Negaunee-Siamo contact generalized in Maas and Negaunee mines Amount of offset on Jackson fault in Negaunee mine is unknown a». w: ISOMETRIC DIAGRAM OF “SOFT” IRON-ORE BODIES IN NE GAUNEE IRON-FORMATION, NE GAUNEE QUADRANGLE, MARQUETTE COUNTY, MICHIGAN "7,7 1.. '. ,y Geochemistry of Lower Eocene Sandstones in the Rocky Mountain Region RSITY 0 %\\\\V F 041400 ,9 4i 7 (o "W8 7973 19 \ SCIENCE um“ Geochemistry of Lower Eocene Sandstones in the Rocky Mountain Region By JAMES D. VINE and ELIZABETH B. TOURTELOT With a section on DIRECT-READER SPECTROMETRIC ANALYSES By RAYMOND G. HAVENS and ALFRED T. MYERS GEOLOGICAL SURVEY PROFESSIONAL PAPER 789 A regional study of element distribution, petrology, and diagenesis among samples of fluvial sandstone from basins of Tertiary age in the Rocky Mountain region UNITED STATES GOVERNMENT PRINTING OFFICE, WASHINGTON : 1973 UNITED STATES DEPARTMENT OF THE INTERIOR ROGERS C. B. MORTON, Secretary GEOLOGICAL SURVEY V. E. McKelvey, Director Library of Congress catalog-card No. 72-600317 For sale by the Superintendent of Documents, US. Government Printing Office, Washington, DC. 20402 Price (paper cover) 75 cents domestic postpaid or 55 cents GPO Bookstore Stock Number 2401~00275 CONTENTS Page Page Abstract ..................................................................................... 1 Petrology and mineralogy—Continued Introduction and acknowledgments ................ 1 Diagenesis and metamorphism ...................................... 18 Geologic setting ----------------------------------------------------- 2 Chemical composition of sandstones ..................................... 20 Distribution and paleogeography .................................. 2 Statistical methods ___________________________________________________________ 20 Stratigraphic relations ................................................... 3 Means ________________________________________________________________________________ 20 Sampling plan """" , """""""""" 6 Geochemical relations ______________________________________________________ 25 Methods of analySIS ....................................................... 12 A 1 sis of variance 25 Direct-reader spectrometric analyses, by Raymond na y , """"""""""""""""""""""""" G. Havens and Alfred T. Myers ................................ 13 00mm?” “313’s“ ------------------------------------------------- 27 X_ray diffraction analyses ______________________________________________ 13 Geochemistry of sample areas ............................... 29 Petrology and mineralogy ...................................................... 14 Discussion ---------------------------------------------------------------------------------- 32 Mineral composition and classification of samples... 14 Summary .................................................................................... 33 Cement and matrix .......................................................... 17 References cited ........................................................................ 33 FIGURE TABLE ILLUSTRATIONS Pa e 1. Map showing basins and major uplifts in early Eocene time, present outcrop pattern of lower Eocene g sedimentary rocks, and sample localities ............................................................................................................... 4 2. Generalized cross section of an Eocene basin ............................................................................................................. 5 3. Triangular diagram showing average composition of lower Eocene sandstone in 36 areas in the Rocky Mountain region ......................................................................................................................................................... 14 4. Photomicrograph of quartz-chert sandstone from the Willwood Formation, Bighorn Basin ............................ 14 5. Map showing distribution of lower Eocene sandstone types, and colors of the sandstone in powdered sam- ples in 36 areas in the Rocky Mountain reglon ..................... 15 6—14. Photomicrographs: 6. Arkosic sandstone from the Wind River Formation, Wind River Basin .................................................. 16 7. Intermediate sandstone, Great Divide Basin ................................................... 16 8. Lithic sandstone from the Wind River Formation, Wind River Basin... ...... 17 9. Etched quartz grain from the Wasatch Formation, Wasatch Range ............ 17 10. Altered feldspar grain from the Hanna Formation, northern Hanna basin ............................................. 17 11. Sandstone, showing a uniformly textured clay cement from the Willwood Formation, west flank of the Bighorn Basin ............................... 18 12. Sandstone, showing a montmorillonite matrix from the Wasatch Formation, Sand Wash basin ........ 19 13. Laumontite-cemented sandstone from the Cuchara Formation, Raton basin .......................................... 19 14. Feldspar grain partly replaced by epidote in arkosic sandstone from the Wasatch Formation, Great Divide Basin ..................................................................................................................................................... 19 15. Diagram showing concentration range of major constituents in 216 samples of lower Eocene sandstone in the Rocky Mountain Region ..................................................................................................................................... 24 16. Diagram showing concentration range of minor elements in 216 samples of lower Eocene sandstone in the Rocky Mountain region ............................................................................................................................................. 25 17. Vector diagram for 216 samples of lower Eocene sandstone .................................................................................. 26 18. Vector diagram for color subsets of the sandstone samples ................................. , .................................................. 30 TABLES Page 1. Sample localities, stratigraphic assignments, and petrographic descriptions of 216 samples of lower Eocene sandstone ______________________________________________________________________________________________________________________________________________________________________ 7 2. Arithmetic mean, standard deviation, geometric mean, and geometric deviation of the analysis of 10 splits from one sample _________________________________________________________________________________________________________________________________________________________ 12 III IV TABLE CONTENTS . Analytical conditions for the direct—reader spectrometric analysis of sandstone ............................................... . Spectral lines used for the elements reported and the concentration ranges covered ....................................... Comparison of direct-reader spectrometric analyses with recommended values ......................... v. ........................ . Limits of detection and values used to replace indeterminate analyses .................................................................. . Distribution of constituents in 216 samples of lower Eocene sandstone and in 36 areas and four color subsets of the samples ............................................................................................................................................... . Variance components ......................................................................................................................................................... . Areas where any constituent mean value for six samples is greater than one geometric mean times the geometric deviation of all samples or is less than one geometric mean divided by the geometric devia- tion of all samples ..................................................................................................................................................... Page 13 13 13 20 21 27 29 GEOCHEMISTRY OF LOWER EOCENE SANDSTONES IN THE ROCKY MOUNTAIN REGION By JAMES D. VINE and ELIZABETH B. TOURTELOT ABSTRACT The composition of lower Eocene fluvial sandstones in the Rocky Mountain region was studied to aid in interpretation of the regional geochemical environment of uranium and other mineral deposits in rocks of early Eocene age. The study was based on 216 samples, from 18 basins, collected “according to a hierarchial plan that involved the random selection of three pairs of beds in each of 36 sample areas. The geometric means of the constituents (in percent) and of the minor elements (in parts per million) for these 216 samples are as follows (geometric deviation in parentheses) : Sl02, 72.9' (1.2); A1303, 6.6 (1.9); total iron as Fe203, 1.2 (2.1); MgO, 0.6 (3.0); CaO, 2.4 (4.3) ; Na20, 0.5 (4.6); K20, 1.4 (2.3); H20+, 1.4 (1.6) ; T102, 0.2 (1.9); P205, 0.05 (2.6); MnO, 0.06 (2.1); C03, 0.7 (12.6); B, 13 (2); Ba, 415 (2); Co, 4 (2); Cr, 13 (3); Cu, 8 (2); Ga, 9 (2); La, 12 (2); Ni, 9 (2); Pb, 8 (2); Sr, 138 (3); V, 24 (2); Y, 10 (2); Zr, 113 (2); U, 2.4 (2); and Th, 4.3 (2). The sampled rocks include feldspar-rich sandstones, which were derived from the erosion of crystalline rocks, and quartz- rich sandstones and carbonate-rich sandstones, which were derived from reworked older sedimentary strata. Rocks rich in both carbonate and feldspar are classed as intermediate. Feldspar-rich sandstones are characteristic of a broad area of central Wyoming, much of Colorado, and northern New Mexico. Carbonate—rich sandstones are characteristic of a narrow belt through central Utah with extensions in the adjacent States. Intermediate rocks lie between carbonate- rich and feldspar—rich sandstones in western Wyoming and northeastern Utah. Quartz-rich rock was found in one area of the Bighorn Basin of Wyoming. The data from chemical, X-ray diffraction, and petro- graphic analyses were analyzed statistically. Principal—com- ponent analysis of the correlation matrix was used to identify five geochemical groups—a quartz group, a feldspar group, a carbonate group, and two minor-element groups between the feldspar and carbonate groups. Boron, titanium, and zir- conium occur in resistate minerals and correlate poorly with everything else except possibly with quartz. Cobalt, chromium, copper, nickel, and vanadium tend to be associated with the intermediate rocks. The enrichment of uranium tends to cor- relate with feldspar-rich sandstones. Statistical analysis of dominantly red, orange, yellow, or green subsets of the samples shows that the dominantly green samples differ significantly from the other subsets and that the greatest difference is between the green and the red sub- sets. The subset of 11 green sandstones has a greater concen- tration of mica, cobalt, chromium, copper, nickel, strontium, vanadium, gallium, total iron, titanium, and the feldspars, especially plagioclase, than the average of all samples. Early diagenetic alteration is evidenced by widespread authigenic clay, carbonate, and zeolite minerals, the etching of quartz, the presumed dissolution of mafic silicate minerals, and the enrichment of minor elements in areas of intermedi— ate rock rather than in areas rich only in feldspar, quartz, or carbonate. In many areas, red sandstone probably represent rocks that were originally greenish gray and that were altered at a later stage. Although rock color can be the result of many processes, the red alteration may have been caused by the introduction of oxygenated meteoric ground waters and the accompanied leaching of iron and many of the more mo- bile elements. INTRODUCTION AND ACKNOWLEDGMENTS Conventional petrologic techniques are combined with the statistical analysis of chemical and miner- alogic data in this study to interpret the geochem- ical history of lower Eocene fluvial sandstones in the Rocky Mountain region. Rocks of Eocene age, includ- ing the sandstones which intertongue with or grade into other facies of Eocene sedimentary rocks, con- tain many diverse and valuable mineral deposits in the same or neighboring basins such as uranium, oil shale, trona, nahcolite, dawsonite, petroleum, phos- phate, and gold. Genetic relations between such deposits might be brought out by studying the de- posits within the larger framework of the regional geochemical setting. Uranium deposits occur in the Wind River, Shirley, Great Divide, and Powder River Basins (Harshman, 1969) ; oil-shale deposits, in the Green River, Piceance Creek, and Uinta Ba- sins (and lesser deposits in the Great Divide, Wash— akie, Fossil, Sand Wash, and Bighorn Basins and the Gunnison Plateau) (Bradley, 1964; Donnell, 1961a; Cashion, 1957, 1967; Culbertson, 1964); trona, in the Green River Basin (Bradley and Eugster, 1969) ; dawsonite and nahcolite, in the Piceance Creek Basin (Smith and Milton, 1966 ; Hite and Dyni, 1967) ; hydrocarbons, including petroleum gilsonite, and other solid bitumens, in parts of the Uinta, Piceance Creek, Green River, and Washakie Basins (Bell and Hunt, 1963; Oil and Gas Journal, 1970); gold, in parts of the Sand Wash, Hoback, and western Wind River Basins (Theobald, 1970; Antweiler and Love, 1 2 GEOCHEMISTRY OF LOWER EOCENE SANDSTONES IN THE ROCKY MOUNTAIN REGION 1967); and uranium-bearing phosphorites, in the Green River Basin (Love, 1964). The origin and distribution of most of the Eocene uranium deposits are thought to be directly related to the geochemical history of lower Eocene sand- stones. The chemical properties of mineralizing fluids trapped within a wedge of sediments were probably affected by different lithologies, from clas- tic detritus to evaporites. Migration patterns of the fluids were locally complex, and possibly fluids mi; grated between basins. Therefore, a thorough under- standing of the distribution of uranium and other epigenetic mineral deposits depends on some knowl- edge of the geochemical history of the entire wedge of sediment in each basin. This report deals with only one lithologic facies, the fluvial sandstones, and only the lower part of the total wedge of Eocene sediments. Thus, any conclusions reached are neces- sarily limited in time and scope, and much addi- tional work remains to be done. Emphasis has been placed on the compilation of geochemical data for determining the history of alteration and for studying the formation of authi- genic minerals and the mobility of minor elements. Evidence for widespread formation of authigenic minerals by either digenesis or hydrothermal altera- tion was presented in a preliminary report (Vine and Tourtelot, 1970a). Some of the same evidence will be reviewed here, and a more thorough statis- tical analysis of the chemical and mineralogic data will be given. To plan and execute a regional project as exten- sive as this required the advice, stimulation, and direct assistance of many colleagues. We thank A1- fred T. Miesch for his help in planning and teaching us the use of the statistical methods that we applied. Geologists of the US. Geological Survey who con- tributed information about the areas to be sampled included William B. Cashion, William C. Culbertson, John R. Donnell, James R. Gill, E. N. Harshman, Ross B. Johnson, William R. Keefer, Thomas E. Mul- lens, Willis L. Rohrer, William N. Sharp, Paul K. Theobold, Harry A. Tourtelot, and Howard D. Zeller. Richard S. Zeman assisted the senior author in field- work during the summer of 1968. We thank Leonard G. Schultz for his advice on making and interpreting the X-ray diffraction analyses. We used a pelletized- sample technique designed by Omer B. Raup for the X-ray analyses. Richard A. Sheppard helped identify laumontite in the X-ray traces and in thin section. Melvin E. Johnson made special plastic-impregnated thin sections of the many friable-sandstone samples. The work of John Moreland in preparing, grinding, and splitting samples and renumbering all sam- ples in a random sequence for submission to the lab— oratories is much appreciated. Louise S. Hedricks and Richard B. Taylor aided us in preparing the photomicrographs of thin sections. George Van Trump, Robert Terrazas, and others in the computer facility helped us use the computer. Chemical and spectrometric analyses were per- formed in the laboratories of the US. Geological Survey. Direct-reader spectrometric analyses were done in the Denver laboratory by Nancy Conklin, R. G. Havens, and Lorraine Lee, analysts, under the supervision of A. T. Myers. Total-carbon analyses were made in the Denver laboratory by I. C. Frost, Elsie Rowe, G. D. Shipley, and Winona Wright. The rapid rock analyses were made in the Washington laboratory by Lowell Artis, S. D. Botts, G. W. Chloe, Paul Elmore, J. S. Glenn, James Kelsey, and H. Smith, analysts, under the supervision of Leonard Shapiro. Carl M. Bunker and Charles A. Bush per- formed the gamma-ray analyses for uranium and thorium. The chemical and mineralogical data were placed on punch cards at the geochemical data stor- age and retrieval facilities, operated by Raoul V. Mendez and Josephine Boerngen. GEOLOGIC SETTING DISTRIBUTION AND PALEOGEOGRAPHY Eocene sedimentary rocks in the Rocky Mountain region were deposited in intermontane basins sepa- rated by tectonically active uplifts resulting from orogenic forces which began in Late Cretaceous time. Potassium—argon radioactive age dates indicate a probable interval of about 5 my (million years) from 54 to 49 my. RP. (before present) for depos- its of early Eocene age (Evernden and others, 1964). The composition, distribution, and coarseness of sed- iments in each basin reflect the composition of rocks in the source terrane and the degree of tectonic activity in the adjacent ranges as well as the tectonic activity and environment of deposition within the basin. Some uplifts, such as the fold belt in western Wyoming and adjacent parts of Utah and Idaho, consisted chiefly of Paleozoic and Mesozoic sedimen- tary rocks. Detritus shed from these ranges included the disintegration products of limestone, dolomite, chert, shale, and sandstone. Precambrian crystalline rocks were exposed in the Bighorn Mountains, Wind River Range, and Laramie Mountains of Wyoming and the Medicine Bow Mountains and Park Range of Colorado. These ranges contributed much arkosic material beginning at about the start of the Eocene. The Precambrian core at the eastern end of the Uinta Mountains was probably also exposed by earliest Eocene time (Culbertson, 1969), but there the Precambrian consisted of quartz-rich sedimen— tary rocks. Some basins were partly surrounded by GEOLOGIC SETTING 3 mountain ranges that consisted of rocks of widely different compositions, and so a great variety of detritus was deposited simultaneously in the same basin. Figure 1 shows the distribution of deposi- tional basins in the Rocky Mountain region in early Eocene time and the major uplifts that were impor- tant contributors of sediment to the basins. Low interbasin divides that contributed little or no sedi- ment are not shown. Some other basins, such as North Park, 0010., may have been receiving sedi- ments at that time, but since then any such sedi- ments either have been eroded away or have been completely buried by younger sediments. Fanglomerates and diamictite facies (Tracey and others, 1961) formed adjacent to rapidly rising mountains. Basinward, the coarse facies grade into fluvial sandstones and mudstones and, locally, clay- stones, algal limestones, carbonaceous shales, and coal beds. Coal beds are numerous in the Powder River, Great Divide, and Hanna basins and are sparse to absent in the other basins. Swamps and lakes began to form in the central parts of some of the basins early in Eocene time. The resultant palu- dal and lacustrine sediments interfinger with the fluvial lower Eocene sediments (Bradley, 1964; Roehler, 1965) of this study. A warm, mild climate is inferred from numerous studies of fossil plants and vertebrates (Barghoorn, 1953; Bradley, 1963; Van Houten, 1948; Soister, 1968, p. A44). About middle Eocene time the basins became filled with sediment. Ponds and lakes expanded, and some former low divides were buried by sediment deposited in giant lakes such as Lake Gosiute (Brad- ley, 1964, fig. 10, p. A36) in southwestern Wyoming. Elsewhere, low divides were buried by fluvial sedi- ment spreading from adjacent basins. Probably, interfingering fluvial and lacustrine sediment in the Hoback, Green River, Fossil, Great Divide, Washa- kie, and Sand Wash basins formed a continuous deposit. Middle Eocene volcanic sediments occur in both the lake-bed and fluvial sediments. Toward late Eocene time the lakes grew smaller and disappeared, and the fluvial sediments of that time contain much volcanic debris. Periodic minor downwarp and uplift reestablished some low divides. In some basins, folding and faulting occurred at the end of early Eocene time and was renewed peri- odically during later Tertiary time, but in the Wind River Basin, as noted by Keefer (1970, p. D9) , “with few exceptions these movements were of minor con- sequence and did not greatly modify the structural patterns that had already been established.” Many of the basins continued to fill throughout most of Tertiary time (Love, 1970, p. Clll—C123) until only a few of the highest mountain peaks extended above the general basin fill. Sedimentation ceased at differ- ent times from basin to basin, but in general the present major cycle of exhumation and downcutting probably began in Miocene or Pliocene time. Early Eocene sedimentation rates were moderately rapid to rapid in most basins of the Rocky Mountain region, ranging from about 500 to 5,000 feet in 5 my These compare with Late Cretaceous sedimen- tation rates of from 60 feet per my in central North and South Dakota to as much as 2,500 feet per my locally in western Wyoming, according to J. R. Gill (oral commun, 1971). The Eocene sedimentation was sufficiently rapid that many unstable minerals, including feldspars, clay minerals, and mafic miner- als, probably arrived at the site of deposition without much chemical alteration. The resulting sediments include mineral constituents that have not had time to come into chemical equilibrium with each other or with the enclosing pore waters. The lower Eocene sedimentary rocks were exposed (fig. 1) by late Cenozoic time at the margins of most basins or areas of preservation. The largest volumes of lower Eocene strata are preserved in the large basins such as the Green River, Wind River, and Powder River basins of Wyoming, but smaller de- posits of lower Eocene strata are also preserved in other structural settings. The Wasatch Range and Wasatch Plateau areas of Utah contain Eocene strata preserved in uplifted mountain blocks, and the Galisteo basin of New Mexico contains Eocene rocks preserved in several isolated downfaulted blocks. The paleogeographic setting is most easily recon- structed for the basin areas and is difficult or im- possible to reconstruct for the other areas. Although our original intention was to study sandstones representing fluvial deposition, we were unable to distinguish among stream-channel sand- stones, flood plains, alluvial fans, pediments, debris flows, and perhaps even paludal, dune, and beach deposits. Truly lacustrine sandstones, as recognized by horizontal lamination, much fossil shell debris, or well-rounded and well-sorted grains, were excluded. STRATIGRAPHIC RELATIONS Though different stratigraphic names have been applied to lower Eocene sedimentary rocks in differ- ent basins, the history of sedimentation is very simi- lar throughout the region. In most basin areas, especially in Wyoming, sedimentation was continu- ous from latest Cretaceous to middle or late Tertiary time except for local interruptions in late Eocene time (Love and others, 1963). Unconformities occur locally at or near the base of Eocene strata along 110” I GEOCHEMISTRY 0F LOWER EOCENE SANDSTONES IN THE ROCKY MOUNTAIN REGION 105° 45° ——~__ 41" Wasatch Range UTAH EXPLANATION Principal basin urea Outcrop of lower Eocene sedimentary rocks 37. Uplifted area exposing sedimentary rocks U Uplifted area exposing crystalline rocks M Uplif ted fold belt @—@ Areas sampled Each circle represents the approximate lo- cality of three pairs of sandstone samples selected at random from the lower Eocene sequence; 216 samples, total. Paired areas are linked by lines I MONTANA Sand Wash ‘ Basin Piceance Creek _ Basin San Juan Mountains 9 fl Gaisteo Basin COLO RADO ! 4 NEBRASKA i i. i. i. __ _ _ _ _ __L iOKLAHOMA I NEW MEXICO | ‘ TEXAS I I O 100 MILES L—g FIGURE 1.— Basins and major uplifts in early Eocene time, present outcrop pattern of lower Eocene sedimentary rocks, and sample localities. Interbasin divides that did not contribute much debris are not shown. Basins and major uplifts modified from King (1959, figs. 66, 69). Present outcrop pattern compiled from various sources, including State geologic maps. GEOLOGIC SETTING 5 Sample localities shown in figure 1 1. Enos Creek 10. Firehole Basin 19'. Huerfano Park 28. Patmos Head 2. Hillberry Rim 11. Difficulty 20. West Spanish Peak 29. Selina 3. Evanston 12. Carbon basin 21. Little Snake River 30. Gunnison Plateau 4. Fossil 13. Dubois 22. Elkhead Mountains 31. Woodrufl 5. Cerrillos 14. Hoback basin 23. Cuba 32. Morgan 6. Hagen 15. Rangely 24. Gobernador 33. Baggs 7. Oregon Buttes 16. Battlement Mesa 25. Bates Hole 34. Erickson—Kent Ranch 8. Crooks Gap—Wamsutter 17. Pine Ridge—Pumpkin Buttes 26. Chalk Hills 35. Lost Cabin 9. La Barge 18. Horse Creek—Crazy Woman Creek 27. Soldier Summit 36. Wallace Creek the basin margins, but deposition was probably con- than Paleocene rocks; red, green, and purple mud— tinuous near the center of each basin, perhaps only stones are characteristic of the lower Eocene strata, a few miles distant (fig. 2). whereas drab shades of brown and gray characterize In parts of southwestern Wyoming, abundant mudstonw in the Paleocene strata. The lowest arko- mammal fossils allow precise dating of strata (Gazin, sic sandstones form a mappable boundary for the 1962). However, in many other areas fossils are :base of the Eocene in some parts of Wyoming lacking or are not diagnostic. In mapping, an arbi- , according to N. M. Denson (oral commun., 1970). trary boundary between Paleocene and Eocene strata Color, texture, and rock composition are all unreli- is commonly defined on the basis of subtle changes: in able indicators of age, however, and within the color, texture, or composition of the. rocks. In many Rocky Mountain region the exact placement of the areas, Eocene rocks tend to be more brightly colored boundary between Paleocene and Eocene strata in 'I \\ I/ \‘l u _\ (\l\ 4' /_’ I!” I: :I? \‘ _ (5/: / .. ,\/ 1/\\_~_ / ~ / ‘3‘ I‘J .‘x' ’ 1 \ \ / / _ ”L‘ 1/,w x, \ \ \ \\ / / ’(J: :/_::,:,;\:,.::P’I \ \ \ \\ \\ // // // / (No l r \1 ' ’ \~ / ~ / ',~ (A \‘~\ I \ \ ./ / / / / ,,//:|\’//l/:l,\ \ \\ \\ \\ // / {/ / "-1 C _\/._\ ”~V: ’l>’/\/\'\” \ ‘;¥--—-’/ // / / -\—,\'1“/\I /_\’l \"-/’7rl‘1 ">\‘ \ \\ \\ / / / "\l\'\— ’\\-S \I:\’ ,p /_\\.‘,—-—I ,,\ \ \\ // / / \\“\\,\\|_ __| \/\I /I /z,l \\ / / / / ,/l// \, ”(gs/"‘1 ’\|\\-|’ \ \\ ~———.—’ / / \ I _l//, /’\\ ‘1)" ,_ z-‘/_’I/I\;\:\’/\//’ \ \\\ // / ' \ "’>l\‘/\ ” l/‘T/‘7 [<7 l -/<’\\’.)’_4l,v‘\~I \‘L / \ “_——-’—’/ // -~\‘ "T‘i'rvi'h’x? is“; ’1'7 ’ row v-—“.’\ “I" \ ,lqvf/ ‘ ”-/.’ , - n ~/\m ‘ EXPLANATION Upper and middle Eocene fluvial sediments . , , Pre-Eocene sedimentary rocks Lacustnne and evaponte sediments Ix ‘7 \‘I/ o . I, ‘ \ 7 Lower Eocene fluvial mudstone and Precambrian “WEBB rocks sandstone _—_____ ° ..° . ‘ Geologic boundary ‘ Q ~ Dashed: whlere gradatioml _._L Lower Eocene conglomerates and diamictites Fault. Amw showsvathrawn side FIGURE 2. ~— Generalized cross section of an Eocene basin at the end of the Eocene Epoch. Modified from Bradley (1964). 490-247 0 - 73 - 2 6 GEOCHEMISTRY OF LOWER EOCENE SANDSTONES IN THE ROCKY MOUNTAIN REGION any specific area is often disputed among paleo— botanists, vertebrate paleontologists, invertebrate paleontologists, and physical stratigraphers. Not un- commonly the boundary based on paleontological cri- teria is within a lithostratigraphic unit. Whereas the lower boundary of the lower Eocene strata is locally unconformable and is generally based on paleontologic evidence, the upper boundary is nearly everywhere conformable and is based on a lithologic change supplemented by meager fossil data. Throughout western Wyoming, northwestern Colo- rado, and northeastern Utah, the lower Eocene flu- vial strata intertongue with and are overlain by lacustrine strata. The upper boundary of the strata included in this study is not a time-stratigraphic boundary but is a boundary between fluvial and lacustrine rocks in the Uinta, Piceance Creek, Green River, Sand Wash, Fossil, Washakie, Great Divide, and central Bighorn Basins and in the Wasatch Plateau. A lithologic change from dominantly detri- tal to dominantly volcanogenic sediments marks ap- proximately the same stratigraphic positions in the western Bighorn, Wind River, and Shirley Basins. Other lithologic changes are recognized in the north- ern Raton and Hoback basins. Middle or upper Eocene rocks have not been reported from the Hanna, San Juan, or Powder River Basins. The boundary between lower and middle Eocene rocks was arbi- trarily placed on the basis of meager fossil evidence, as one-third of the distance above the base of the Eocene rocks in the Galisteo basin. No stages are recognized in the Wasatch Range. We could not find thoroughly documented or de- scribed stratigraphic sections of the rocks to be sampled for each basin. We relied as much as pos- sible on published reports of the presence of early Eocene fossils or on physical evidence that the rocks sampled could be correlated with strata that were reported to contain early Eocene fossils. Table 1 lists the areas where sandstone was sampled, principal references to a stratigraphic section described in each area, and petrographic descriptions. Most stratigraphic thicknesses in the table are quoted directly from the cited reference, but we estimated the thickness at some places, as indicated in the table. Where a range in thickness is given, the stratigraphic unit was generally sampled near the area where it is thickest. Where a minimum thickness is given, the complete thickness may be unknown because of poor exposures or because the top has been eroded away. Where two thicknesses are separated by a comma, the fluvial sequence is divided into two tongues which are separated by an unspecified thick- ness of lacustrine strata. Plant microfossils recently described by Gill, Merewether, and Cobban (1970, p. 47—48) suggest that the lowermost samples we collected from the Carbon basin may be late Paleo- cene rather than early Eocene in age. SAMPLING PLAN The sample population for this study consists of those accessible lower Eocene fluvial sandstones ex- posed in the Rocky Mountain region (fig. 1) that are not conspicuously weathered, mineralized, or meta- morphosed. We recognize that all sandstone included in the sample population have been weathered to some degree and have been diagenetically altered during their 50 my. history; these processes have contributed to their present geochemical character. The object of the sampling plan was to obtain unbiased geochemical information about the sample population. The sampling plan was designed so that statistical techniques could be used to look for geo- graphic patterns in the distribution of individual constituents and to determine the characteristics of the target population. We hope that the sampling plan would be such that the characteristics of the lower Eocene fluvial sandstones could be determined and that the effect of microenvironments within the fluvial sedimentation regime could be minimized. Sampling was also limited by the amounts of time and money available for field and analytical work as well as by natural conditions such as the accessibility and abundance of suitable outcrops. We designed a four-level hierarchial sampling plan similar to the plans used by Olson and Potter (1954) and Krumbein and Slack (1956), who described in detail the implications, applications, and theory of hierarchial sampling plans. One major difference be- tween the cited sampling plans and the one used in this study is that the cited plans were based on pre— determined geographic spacing between samples, whereas this plan is based on the structural pattern of the region sampled, and the geographic distance between sample localities varies greatly (fig. 1). The design of the sampling plan used in this study is as follows: Level 4: The 18 major basins of lower Eocene deposition within the Rocky Mountain region ;1 Two areas, or stratigraphic sections, within each major basin of deposition; A set consisting of three pairs of sand- stone beds within each stratigraphic section; and A pair of sandstone beds, each repre— sented by one sample. Level 3 : Level 2 : Level 1 : 1The original sampling plan included two areas (12 samples) from the Bearpaw Mountains of north-central Montana. Analysis of these samples showed that some contained authigenic amphibole, indicating a higher degree of metamorphism than we thought acceptable for this study. Data from these samples were included, however, in a preliminary report (Vine and Tourtelot, 1970a). SAMPLING PLAN 7 TABLE 1.—Sample localities, stratigraphic assignments, and petrographic descriptions of 216 samples of lower Eocene sandstone [An asterisk (‘) preceding the thickness indicates Color Chart” of Goddard and others (1948): IOYR 8/2, 10Y and N 9; and green, 5Y 7/1, 5Y 6/1, 5Y 5/2, 5 . . supplemented by field descriptions, X-ray diffraction data, and a few heavy mineral studies] our estimate. The four color subsets represent a compilation of the following colors from the “Rock- Red, 10R 6/2, 10R 6/6, 10R 7/4, 7R 7/4, 10R 5/4, BYR 8/1, 5YR 7/2, 5YR 8/4, and 5R 8/1; orange, R 7/4, 10YR 8/6, IOYR 7/6, IOYR 6/4, 5Y1? 4/1, 5YR 5/6, and 5YR 6/4; yellow, 5Y 7/2, 5Y 8/1, 5Y 8/4, _5YR 6/1, N 8, GY 7/2, BGY 8/1, 5G 8/1, and NY 8/2. Petrographic descriptions are based on thin-section study, Locality No. (figs. 1, 5) Area and reference Enos Creek (Rohrer, 1966). Hillberry Rim (Van Houten, 1944). Stratigraphic unit ; thickness (ft) Willwood Formation ; 3 4 0 . Willwood Formation ; ’600. Field sample No. BH07— BH12 BH01— BH06 Locality Sec., T. and R. Bighorn Basin, Wyo. NE‘A 28 46 N., 100 W. Number of samples in specified color subset 5 orange, 1 yellow. 4 orange, 2 yellow. Petrographic description Very poorly to poorly sorted, very fine to coarse- grained. Quartz, microcline, fresh to altered plagioclase, chert, and composite grains in clay or carbonate matrix; some iron stain. Accessory minerals: white mica, fresh to very altered green and brown biotite, chloritoidlike minerals, zir- con, garnet, tourmaline, glauconite, and opaque minerals. Poorly sorted, very fine to medium-grained. Quartz, chert, potassium feldspar, sparse plagioclase, and clay closely packed in clay matrix; some iron stain. Accessory minerals: fresh to altered biotite, zircon, tourmaline, hypersthene, glauco- nite, and opaque minerals. Fossil Basin, Wyo. Evanston (Oriel and Tracey, 1970). Fossil (Oriel and Tracey, 1970). Wasatch Formation ; 1 ,000—1,500. Wasatch Formation ; 1.000—1,500. F001— F006 F007— F012 22 15 N., 119 W. SE14 26 22 N., 118 W. 5 yellow, 1 orange. 4 yellow, 2 orange. Poorly sorted, very fine to very coarse grained. Quartz, chert, fresh to altered feldspars (chiefly potassium feldspars), composite grains, schist, sedimentary rock fragments, and calcite and dolomite grains in carbonate matrix. Accessory minerals: white mica, green biotite, zircon, gar- net, epidote, chlorite, and opaque minerals. Very poorly to poorly sorted, medium to very coarse grained. Quartz, chert, dolomite rhombs, and sparse feldspars in carbonate and clay ma- trix; some iron stain. Accessory minerals: zir- con; few accessory minerals present. Galisteo Basin, N. Mex. Cerrillos (Robinson, 1957) . Hagen (Stearns, 1943) . Galisteo Formation, lower part; 930. Galisteo Formation, lower part; 1,270. GA01— GA06 GA07— GA12 SE34 16 14 N., 8 E. 27 13 N., 6 E. 4 red. 1 orange, 1 yellow. 2 red. 2 orange, 2 yellow. Very poorly to poorly sorted, medium to very coarse grained. Quartz. chert, generally fresh feldspars, composite grains, and carbonate grains in clay or calcite matrix. Accessory minerals: white mica, zircon, tourmaline, garnet, opaque minerals; few accessory minerals present. Poorly to moderately well sorted, fine- to coarse- grained. Quartz, fresh to completely altered feld- spars, chert, and schist closely packed to floating in clay matrix, sparse calcite matrix; some iron stain. Accessory minerals: fresh to altered brown and green biotite, epidote, garnet, zircon, clino- zoisite, and opaque minerals. Great Divide Basin, Wyo. Oregon Buttes (Zeller and Stephens, 1969). Crooks Gap— Wamsutter (Stephens, 1964; Masursky, 1962). Wasatch Formation, main body; 3,000. Wasatch Formation, Cathedral Bluffs Tongue; 160—200. Battle Spring Formation ; 500— 2,500. Wasatch Formation ; SOD—3,500. GDOI- GD02 GD03— GD06 GD07— GD08 GD09— GDlO GDll— GD12 9 24 N., 99 W. 51/2 21 27 N., 101 W. NW% 19 27 N., 92 W. 2 26 N., 93 W. 22 22 N., 94 W. 2 yellow. 2 green, 2 yellow. 2 yellow. do. 1 yellow, 1 green. Poorly sorted, fine- to medium-grained. Quartz, chert, fresh to altered feldspars, dolomite, and dctrital calcite in clayey calcite matrix. Acces- sory minerals: fresh to altered biotite, chlorite, garnet, zircon, tourmaline, epidote, zeolite, and opaque minerals. Very poorly sorted, granular to conglomeratic. GD03 and GD04 too friable for thin section. Quartz, fresh to completely altered feldspars, schist, and sandstone fragments in calcite and chlorite matrix. Accessory minerals: abundant muscovite and biotite, authigenic epidote, chlo- rite, garnet, zircon, and opaque minerals. Very poorly sorted, granular. GD07 too friable for thin section. Microcline, fresh to altered plagio- clase, and quartz in sparse clay matrix. Acces- sory minerals: altered green and brown biotite; few accessory minerals present. Poorly sorted, coarse-grained. Etched quartz and altered feldspars in clay or calcite matrix. Ac- cessory minerals: altered biotite; very few acces- sory minerals. Poorly to moderately well sorted, very fine to medium-grained. Quartz, fresh to altered feld- spars, detrital dolomite, limestone, and shale grains in iron-stained clay matrix. Accessory minerals: altered biotite, white mica, garnet, zircon, tourmaline, epidote, chlorite, and opaque minerals. Green River Basin, Wyo. 9 La Barge (Oriel, 1961, 1962; Privrasky, 1963). Wasatch Formation, £153“, Fork Tongue; B307- B310 9 27 N., 113 W. 4 green. Poorly sorted, very fine to medium-grained. BBO7— BB09: Quartz, carbonate grains, fresh to altered feldspars, and analcime tightly packed in very little clay matrix. Accessory minerals: abundant green, 'red, and brown biotite, garnet, zircon, and opaque minerals. B810: strongly etched quartz and feldspars in carbonate matrix. Ac- cessory minerals: biotite, garnet, zircon, tourma. line, anhydrite, and opaque minerals. 8 GEOCHEMISTRY OF LOWER EOCENE SANDSTONES IN THE ROCKY MOUNTAIN REGION TABLE 1. —Sample localities, stratigraphic assignments, and petrog'raphic descriptions of 216 samples of 'lower Eocene sand- stone—Continued Locality No. (figs. 1, 5) Area and Stratigraphic unit ; reference thickness (ft) Field sample Locality Sec., T. and R. Number of samples in specified color subset Green River Basin, Wyo. —— Continued Petrographic description 10 Wasatch Formation, main body; ‘1,500. Firehole Basin Wasatch Formation, (Culbertson, 1965). main body; 2,400. Wasatch Formation, Niland Tongue ; 450. BB11— B812 BBOI— BB04 BB05— BB06 9 27 N., 113 W. 34 17 N., 106 W. SW14 34 17 N., 106 W. 2 orange. 2 yellow, 1 red, 1 orange. 2 yellow. Poorly sorted, fine- to coarse—grained. Quartz, fresh to altered feldspars, chert, and mica schist in iron-stained carbonate matrix. Accessory min— erals: abundant biotite, muscovite, clay, dolo- mite, garnet, tourmaline, epidote, and zircon. Poorly to moderately well sorted, very fine to medium-grained. BB01—BBO2: quartz, fresh to altered feldspars, chert, and dolomite in iron- stained clayey carbonate matrix. Accessory min- erals: fresh to altered green and brown biotite, white mica, garnet, zircon, tourmaline, and chloritoidlike minerals. BB03—BB04: quartz, feld- spars which are altered to sericite or analcime, dolomite grains with iron-stained rims, chert, and schist in clay matrix. Accessory minerals: fresh to altered biotite, chlorite, abundant garnet (as large as 5mm), white mica, epidote, blue and green tourmaline, and opaque minerals. Poorly to moderately well sorted, fine- to medium. grained. Etched quartz, fresh to altered feldspars, schist, and dolomite closely packed to floating in calcite and clay matrix. Accessory minerals: red, green, and brown biotite, chlorite, white mica, coarse garnets, epidote, tourmaline, and opaque minerals. Hanna Basin, Wyo. 11 12 Difficulty (Knight, Unnamed Eocene 1951; Gill, 1970, rocks; 1,500. p. 47—48) . Carbon basin Unnamed Eocene (Dobbin and rocks and rocks of others, 1929). probable Paleocene age mapped as Hanna Formation; 7.000. HNOI— HN04 HNOS— HN06 CA01— CA04 CA05 CA06 NW% 26 24 N., 81 W. SW14 23 24 N., 81 W. SE14 20 21 N., 79 W. SW14 24 21 N., 79 W. NWIA 25 21 N., 79 W. 3 yellow, 1 orange. 1 yellow, 1 orange. 2 orange, 1 red. 1 yellow. 1 yellow. 1 yellow. Hoback—western Wind River Basins, Wyo. Very poorly to poorly sorted, fine-grained to con- glomeratic. HN01 too friable for thin section. Quartz, fresh to altered feldspars, composite grains, and clay aggregates in clay matrix; some iron stain. Accessory minerals: fresh to altered biotite, chloritoidlike minerals, zircon, tourma- line, epidote, and opaque minerals; accessory minerals sparse. Very poorly to poorly sorted, medium-grained to granular. Composite grains of quartz and feld- spar, quartz, and fresh to altered feldspars in clay and calcite matrix. Accessory minerals: fresh to altered green and brown biotite, white mica, zircon, nephrite, chlorite, and opaque min- erals; accessory minerals sparse. % Poorly sorted, fine-grained. Quartz, chert, feldspars, schist, and dolomite rhombs densely packed to floating in clay and carbonate matrix; some iron stain. Accessory minerals: altered biotite, white mica, chlorite, zircon, garnet, tourmaline, and opaque minerals. Poorly sorted, medium-grained. Quartz, fresh to altered feldspars, and chert in calcite matrix. Accessory minerals: fresh to altered biotite, gar- net, tourmaline, zircon, rutile, epidote, and opaque minerals. Poorly sorted, coarse-grained. Etched quartz, al- tered feldspars, and chert floating in calcite matrix. *Accessory minerals: abundant altered biotite, authigenic epidote, garnet, zircon, tour- maline, and opaque minerals; accessory minerals abundant. 14 Dubois (Keefer, Indian Meadows and 1957). Wind River Formations, undivided; 0—1,800. Hoback basin (Dorr, Hoback Formation, 1952) . upper part; 3,000. WHO 1— WR06 H001— H002 H003— H004 H005— H006 42 N., 106 W 28 38 N., 113 W. NWV; 34 38 N., 113 W. 19 37 N., 113 W. 6 yellow. H orange, 1 yellow. 1 green. 1 yellow. ,4 yellow, 1 orange. Piceance Creek basin, Colo. Very poorly to poorly sorted, fine to very coarse grained. WR04—WR06 too friable for thin sec- tion. Quartz, fresh to altered feldspars, and car- bonate grains in carbonate and clay matrix. Accessory minerals: fresh to very altered green and brown biotite, chloritoidlike minerals, gar- net, zircon, tourmaline. glauconite, and opaque minerals. Poorly sorted, fine- to medium-grained. Quartz, chert, detrital calcite and dolomite, and minor feldspar in carbonate and clay matrix. Accessory minerals: altered biotite, tourmaline, zircon, and opaque minerals; accessory minerals sparse. Poorly sorted, very fine to coarse-grained. Quartz, chert, fresh to altered feldspars, and detrital carbonate grains closely packed in clay and car- bonate matrix. Accessory minerals: abundant biotite, garnet, zircon, white mica, and opaque minerals. H004 contains several pollen grains or spores. Poorly to moderately well sorted, medium- to coarse-grained. Quartz, fresh to altered feldspars, and chert in closely packed clay (H005) and calcite (H006) matrix. Accessory minerals: fresh to altered biotite, garnet, sphene, epidote, monazite, zircon, zoisite, and opaque minerals; accessory minerals abundant. 15 Rangely (Donnell, Wasatch Formation; 1961a). 375—580. P001— P004 10 1 N., 100 W. 3 orange, 1 yellow. Poorly sorted, medium-grained. Quartz, generally altered feldspars, chert, dolomite grains, and 5. SAMPLING PLAN 9 TABLE 1. ——Sample localities, stratigraphic assignments, and petrograph’lc descriptions of 216 samples of lower Eocene sand- stone—Continued Locality N. (figs. 1.5) Area and reference Stratigraphic unit ; thickness (ft) Number Locality 0f Field samples sample in No. specified color subset Piceance Creek basin, Colo. —— Continued Sec., T. and R. Petrographic description 16 Battlement Mesa (Donnell, 1961b) . Wasatch Formation : 2,1 P005— P006 N 10 orange. 1 N., 100 W. P007— P010 SWV; 30 2 orange. 9 S., 96 W. 2 yellow. P011— P012 SEM?)7 25 1 red, 9 S.,1 yellow. Powder River Basin, Wyo. kaolinite closely packed in clay and calcite ma— trix. Accessory minerals: fresh to altered biotite, garnet, zircon, glauconite, tourmaline, and al- tered mafic minerals. Poorly to moderately well sorted, fine-grained. Quartz, feldspars, limestone, and dolomite grains closely packed in calcite matrix; PC06 very iron stained. Accessory minerals: zircon, glauconite, chloritoidlike minerals and opaque minerals; ac- cessory minerals sparse. Poorly sorted, fine— to coarse—grained. Quartz, fresh to completely altered feldspars, kaolinite, chert, and limestone grains in iron-stained clay and sparse calcite matrix: birefringent rim of clay around many grains. Accessory minerals: very altered biotite, white mica, zircon, chlorite, and opaque minerals; accessory minerals sparse. Very poorly to poorly sorted, fine- to coarse- grained. Quartz, fresh to altered feldspar, kaolin- ite, and chert in clay matrix: birefringent rim of clay around many grains. Accessory minerals: very altered biotite, white mica, zircon, tourma- line, and opaque minerals; accessory minerals sparse. 17 Pine RidgehPumpkin Buttes (Sharp and others, 1964). Horse Creek—Crazy Woman Creek (Olive, 1957; Hose, 1955). Wasatch Formation ; 1,575 . Wasatch Formation ; 150, 200 PROI— EV 12 2 red, PR04 2 yellow. SW14 13 1 orange, . 1 yellow. orange, 1 yellow. '11 5'6 0 4 I ... "U ’(‘U 9 <0 I a; 7 yellow. 49 N.. 79 W. 4 1 49 N.. 80 W. Raton basin, Coio. Very poorly to moderately well sorted, very fine to medium-grained. Quartz, chert, and sparse potassium feldspar in sparse clay matrix. Acces- sory minerals: white mica, fresh green and blue hornblende, pyroxenes, zircon, tourmaline, epi- dote, and opaque minerals. Poorly sorted, coarse-grained. Too friable for thin section. PR05 contains heulandite. Poorly sorted, fine-grained. Deeply etched quartz and sparse feldspar and dolomite rhombs floating in calcite matrix. Accessory minerals: bright- green and brown biotite, white mica, zircon, garnet, pyroxene, and altered mafic grains. Poorly sorted, very fine to medium-grained. Etched quartz, chert, and sparse feldspar in abundant calcite matrix. Accessory minerals: brown, yel- low, and green biotite, white mica, zircon, gar- net, epidote, glauconite, tourmaline, and opaque minerals. 19 20 21 22 Huerfano Park (Johnson, 1959). West Spanish Peak (Johnson and others, 1958). Little Snake River (Sears and Bradley, 1924; Theobald, 1970). Elkhead Mountains (McKenna, 1960; Theobald, 1970). Cuchara Formation ; Huerfano Formation ; Huerfano Formation ; 2,000 Cuchara Formation ; 400. Wasatch Formation ; >1,500 Wasatch Formation ; , RA07- 7 1 red, RAOS 26 S., 70 W. 1 yellow. RAOQ— SW1/4r 17 and 31 4 red. RA12 26 S., 70 W. RAOI— 21 4 red. RA04 31 S., 68 W. RA05- SW14 23 RA06 31 S., 68 W. N orange. Sand Wash basin, Colo. orange, 1 yellow. yellow. E O 0‘ l is .. at at. N “O HND—l yellow, SW07— SW08 SWV; 8 10 N., 88 W. N yellow. SWO9— 26 2 yellow. SW10 11 N.. 90 W. 1 orange. Very poorly sorted, coarse-grained to granular. Quartz, feldspars, and schist grains closely packed in sparse calcite matrix. Accessory minerals: altered biotite and opaque minerals; accessory minerals sparse. Poorly sorted, coarse—grained to granular. Quartz, feldspars (abundant microcline), mica schist, and chert in clay matrix. Accessory minerals: brolwn biotite, garnet, epidote, and opaque min- eras. Poorly sorted, medium to very coarse grained. Quartz and fresh to altered feldspars in clay and laumontite matrix; laumontite replaces some of matrix and some plagioclase. Accessory minerals: authigenic epidote, garnet, zircon, and opaque grains. Poorly sorted, medium to very coarse grained. Quartz and feldspars in clay and laumontite matrix; laumontite replaces some plagioclase. Accessory minerals: altered biotite, authigenic epidote, garnet, and opaque minerals. Poorly to moderately well sorted, fine- to coarse- grained. SW02-SW06 too friable for thin section. SW01: Quartz and fresh to altered feldspars in clay and iron oxide matrix. Accessory minerals: fresh to altered biotite and white mica, epidote, zircon, garnet and opaque grains. SW06: Quartz, fresh to altered feldspars and composite grains floating in calcite matrix. Accessory grains: biotite, white mica, and opaque min- erals: other accessory minerals contained in framework grains. Poorly sorted, medium— to coarse-grained. SW07 too friable for thin section. Quartz, fresh to altered feldspars, chert, and shale in clay ma- trix; rims of clay around grains. Accessory minerals: white mica, zircon, and opaque min- eras. Poorly to moderately well sorted, fine- to coarse- grained. Quartz, chert, potassium feldspar, and shale in clay matrix; rims of clay around many grains. Accessory minerals: green biotite, white mica, zircon. chloritoidlike minerals, and opaque minerals. 10 GEOCHEMISTRY 0F LOWER EOCENE SANDSTONES IN THE ROCKY MOUNTAIN REGION TABLE 1. —Sample localities, stratigraphic assignments, and petrographic descriptions of 216 samples of lower Eocene sand- stone—Continued Locality No (figs. 1, 5) Area and reference Stratigraphic unit ; thickness (ft) Locality Field sample Sec., T. and R. Number of samples in specified color subset Sand Wash basin. Colo. -— Continued Petrographic description SW11— 7 SW12 11 N., 90 W. 2 orange. Poorly sorted, fine-grained to conglomeratic. SW12 too friable for thin section. Quartz, fresh to altered feldspars, shale, and chert in sparse clay matrix. SW12 contains gypsum. AcceSsory min- erals: altered biotite, authigenic epidote, garnet, zircon, and opaque minerals. San Juan Basin, N. Mex. 23 Cuba. (Baltz. 1967). 24 Gobernador (Simpson, 1948) . San Jose Formation, Llaves Member; 1,300. San Jose Formation, Cuba Mesa Member. upper tongue; 70. San Jose Formation. Regina Member; San Jose Formation, Cuba Mesa Member, upper tongue: 52. San Jose Formation; SJOl— NWKSEV; 36 SJOZ 24 N., 1 W. SJ03 SW54 3]. 25 N.. 1 E. SJ04 SW14 31 25 N., 1 E. SJ 05— SJ 06 SW54 1 21 N., 2 W. SJ07— SJ12 SE14 8 29 N., 4 W. 1 orange, 1 red. 1 yellow. 1 orange. 2 orange. 5 orange, 1 yellow. Very poorly to poorly sorted, medium- to coarse- grained. Many composite grains, quartz. chert, fresh to altered feldspars, and schist in clay matrix. Accessory minerals: white mica, biotite, tourmaline, zircon, garnet, apatite. and opaque minerals. Poorly sorted, medium-grained. Quartz, feldspars, and garnet schist in clay matrix. Accessory minerals: biotite, muscovite, garnet, zircon, chloritoidlike minerals, and opaque minerals. Moderately well sorted, coarse-grained. Quartz, potassium feldspar, chert, and rare plagioclase closely packed in clay matrix. Accessory min- erals: zircon and opaque minerals. Poorly sorted, coarse-grained. Quartz, fresh to altered feldspars, chert, schist, and sandstone and shale fragments in sparse clay matrix. Accessory minerals: muscovite, fresh to altered biotite, zircon, chlorite, and opaque minerals. Very poorly to moderately well sorted, medium to very coarse grained. Quartz, fresh to altered feldspars, and fragments of igneous, metamor- phic, and sedimentary rocks in clay and sparse calcite matrix; some iron stain. Accessory min- erals: altered biotite, zircon, white mica, epidote, abundant garnet (especially in SJ12). and opaque minerals. Shirley Basin, Wyo. 25 Bates Hole (Harshman, 19 68) . 26 Chalk Hills (Harshman. 1968) . Wind River Fggmation ; 300— Wind River Formation ; 300— $1101— SH02 NW1/; 3 27 N., 80 W. SH03— SHOG SHO’l— SH12 28 N., 80 W. Wl/é 16; SE14 17 27 N., 77 W. 2 yellow. 4 yellow. 6 yellow. Very poorly sorted, medium- to coarse-grained. SHOZ too friable for thin section. Coarse quartz and composite grains and fine feldspar grains loosely packed in iron-stained clay matrix. Ac- cessory minerals: altered green biotite, zircon, epidote, wollastonite, monazite, and opaque min- erals: accessory minerals abundant. Very poorly sorted, granular to conglomeratic. Too friable for thin section. Very poorly sorted, very fine to medium-grained. Very friable; thin sections for SHO'T and SH08 only: Quartz. fresh to altered feldspars, and composite grains in clay matrix; plagioclase altered to clay. Accessory minerals: white mica, zircon in quartz ”grains, and opaque minerals; accessory minerals sparse. Uinta Basin, Utah 27 Soldier Summit (Henderson, 1958) . 28 Patmos Head ( Fisher and others, 1960). Colton Formation ; Wasatch Formation ; about ‘500. UIOI— 25 U106 11 S., 8 E. SW1/4 23; SW14 22 10 S. 7 E. 1 UIO7— Nl/2 1 16 s., 14 E. 5 red, 1 yellow. 3 orange, 2 red, 1 yellow. Very poorly to poorly sorted, very fine to medium- grained. Quartz, altered feldspars, kaolinite. and carbonate grains in clay and carbonate matrix: many grains altered to clay, sericite, and iron oxides. Accessory minerals: white mica, zircon, tourmaline, etched garnet, and opaque minerals. Very poorly to moderately well sorted, very fine to coarse-grained. Quartz. fresh to altered feld- spars, chert, and carbonate grains in iron- stained carbonate matrix; sparse clay matrix in U107. Analcime forms part of matrix and re- places some feldspar grains in all samples except U107. Accessory minerals: white mica, garnet. zircon, tourmaline, altered mafic minerals, and opaque minerals. Wasatch Plateau, Utah 29 Salina (McGookey, 1960) . 30 Gunnison Plateau (Hardy and Zeller, 1953). Colton Formation ; 530. Colton Formation ; 600. WP07— 24 WP08 22 S.. l E. WP09— WP12 SE14 25 22 S.. 1 E. WP01— 12 WP06 16 S., l E. 2 orange. 2 orange. 2 red. 3 orange. 2 yellow, 1 green. Poorly to moderately poorly sorted, fine- to coarse- grained. Quartz, dolomite, limestone, chert, and kaolinite in carbonate matrix. Accessory min- erals: white mica, zircon, and opaque minerals: accessory minerals sparse. Very poorly to poorly sorted, fine- to medium- grained. Quartz, fresh feldspars, dolomite, and limestone in carbonate matrix; patches of celes- tite matrix in WP09 and WPIO. Accessory min- erals: chert, zircon, tourmaline, garnet. white 'mica, sparse biotite, and opaque minerals. Poorly to moderately well sorted, very fine to coarse-grained. Quartz, chert, and fresh to al- tered feldspars closely packed in clay matrix; WP06 has carbonate matrix. Accessory minerals: abundant altered biotite, garnet, zircon, tourma- line, epidote, chlorite, white mica, heulandite in WP03, analcime in WP05, and opaque minerals; accessory minerals abundant except in WP06. SAMPLING PLAN 11 TABLE 1. —Sample localities, stratigraphic assignments, and petrogmphic descriptions of 216 samples of lower Eocene sand- stone—Continued Locality No. Area and (figs. reference 1, 5) Stratigraphic unit ; thickness (ft) Locality Field sample Sec., '1‘. and R. Number of samples in specified color L A. Petrographic description 31 Woodrufi (Stokes and Madsen, 1961 ) . 32 Morgan (Mullens, 1971, p. D18). Wasatch Formation ; ‘0—1,000. Wasatch Formation ; 5,0 Wasatch Range. Utah WMOl- 23 WM02 9 N., 5 E. 8 WM03— 1 9 N., 6 E. WM04 WM05— WM06 21 9 N., 6 E. WM07— WM08 SE14 7 3 N., 3 E. WM09— WM12 81/29 3N.,3 E. 2 red. 2 red. 2 orange. 2 red. Very poorly sorted, very fine grained to granular. Chert, quartz, and detrital limestone in calcite and hematite matrix. Accessory minerals: tour- maline, garnet, and zircon. Poorly sorted, fine- to coarse-grained. Quartz and chert in hematite and kaolinite matrix. Acces- sory minerals: zircon, tourmaline, white mica, and opaque minerals (mostly in WM04). Poorly to moderately well sorted, very fine to coarse-grained. Quartz, chert, detrital limestone and dolomite in calcite matrix; calcite replaces some framework grains. Accessory minerals: zircon, tourmaline, white mica, and opaque min- erals; biotite has been replaced by calcite (most accessory minerals in WMOB). Poorly sorted, coarse— to fine-grained. Quartz. chert, dolomite, limestone, kaolinite, and shale in calcite matrix. Accessory minerals: fresh to altered biotite, zircon, tourmaline, and opaque minerals; all mostly altered to calcite in WM07. Very poorly to moderately well sorted, fine-grained to granular. Quartz, chert, dolomite, feldspars (except WMll), limestone, and shale in calcite and hematite matrix. Accessory minerals: zircon, tourmaline, altered biotite, and opaque minerals (very few accessory minerals in WMll—WM12). Washakie Basin. Wyo. 33 Baggs (Bradley, 1964 ; Olson, 1959) . 34 Erickson-Kent Ranch (Bradley, 1964, p. A22; Roehler. 1970). Wasatch Formation, main body: >500. Wasatch Formation, Cathedral Bluffs Tongue; 1,250. Wasatch Formation, main body; 1,700. Wasatch Formation, Cathedral Bluffs Tongue: 850. WK01— E1/2 31 WK04 13 N., 91 w. WK05— 6 WK06 14 N., 92 W. WK07— WKOS NE1/4 13 14 N., 102 W. WK09— SW14 19 WK10 14 N.. 101 W. WKll— 18 WK12 14 N., 99 W. 4 yellow. 1 yellow, 1 orange. 1 yellow, 1 orange. 2 yellow. 2 yellow. Very poorly to moderately well sorted, fine- to coarse-grained. Quartz, generally fresh feldspars, chert, and shale in dirty clay matrix: KYDsum matrix for WK01. Accessory minerals: fresh to altered green and brown biotite, white mica, zircon, garnet, tourmaline, chlorite, epidote, barite, vesuvianite, and opaque minerals. Very poorly to poorly sorted, coarse-grained. Fresh to altered feldspars, quartz, chert, schist and quartzite in calcite and clay matrix. Accessory minerals: altered biotite, white mica, epidote. garnet, zircon, tourmaline, and opaque minerals. Moderately well sorted, fine- to medium-grained. Quartz, altered feldspars, and clay (dolomite and limestone in WKOS) closely packed in clay matrix with minor iron oxides. Accessory min- erals: abundant biotite and white mica that are oriented parallel to the bedding, zircon, tourma- line, garnet, chloritoidlike minerals, and opaque minerals. Moderately well sorted, fine-grained. Quartz, fresh to altered feldspars, detrital dolomite, and chert in brown-stained clay and calcite matrix. Acces- sory minerals: white mica, altered brown biotite, chlorite, abundant garnets, zircon, tourmaline, and opaque minerals. Poorly sorted, fine-grained. Quartz, fresh to altered feldspars, and chert floating in calcite matrix. Micas occur along bedding planes, and biotite is in all stages of alteration. Accessory min- erals: irregularly shaped garnets, tourmaline, chloritoidlike minerals, zircon, and opaque min- era 5. Wind River Basin, Wyo. 35 Lost Cabin (Tourtelot. 1946 ; Keefer, 1965). 36 Wallace Creek (Rich, 1962). Wind River Formation, Lysite Member; 254. Wind River Formation, Lost Cabin Member, 173. Wind River Formation, upper coarse facies; 850. Wind River Formation, lower variegated facies: 275. Wind River Formation, lower fineLgrained facies; 500. WR07- SE14 15 WRIO 39 N., 90 W. WRll— WR12 NElA 22 38 N., 89 W. WR13— WR14 NWI/1 21 32 N., 84 W. WRIS— WR16 NE% 2 34 N., 87 W. WR17— WR18 SW1/4 19 35 N., 86 W. 3 orange. 1 yellow. 2 green. 2 yellow. 1 yellow, 1 orange. 1 yellow, 1 orange. Very poorly to moderately well sorted, very fine to medium—grained. Quartz, silicified mudstone, dolomite, chert, and sandstone closely packed to floating in iron-rich carbonate matrix. Accessory minerals: biotite, White mica, glauconite, and zircon; framework grains contain a variety of accessory minerals. Very poorly to poorly sorted, fine- to coarse-grained. Quartz, feldspars. chert, and limestone floating in iron-rich clay matrix. Accessory minerals: olive, bright-green, and bright-red biotite, white mica, glauconite, garnet, zircon, tourmaline, epi- dote, zoisite, and opaque minerals. Very poorly sorted, granular to conglomeratic. Too friable for thin section. Poorly sorted, coarse-grained to granular. WRlG too friable for thin section. WR15: Quartz and fresh to altered feldspars in composite grains closely packed in clay and biotite matrix. Acces- sory minerals: mostly zircon in composite grains. Very poorly sorted, medium-grained to granular. WR17 too friable for thin section. WRIS: Quartz and fresh to altered feldspars in composite grains in clay matrix. Accessory minerals: brown biotite, white mica, glauconite, chlorite, zircon, and opaque minerals. 12 GEOCHEMISTRY OF LOWER EOCENE SANDSTONES IN THE ROCKY MOUNTAIN REGION The areas to be sampled were selected on the basis of published descriptions of a stratigraphic section in the areas or on the basis of personal knowledge Of suitable outcrops. The location of the sampled areas are shown in figure 1, and the sample localities are listed in table 1. Because the Hoback basin is too small to warrant sampling at more than one place, the paired section was chosen from the westernmost end of the Wind River Basin; we thought that the geographic settings of the sequences in the two areas were sufficiently similar to negate the effects of devi- ating from the sample plan. Where possible, sand- stone was sampled where a complete sequence of lower Eocene strata is exposed within a short hori- zontal distance. In some basins, roadside outcrops extending for several miles were sampled owing to the lack of better outcrops. Within the stratigraphic section in each area, three stratigraphic horizons were chosen at random. For example, the section might be divided into 50 or 100 equal parts and the three sample horizons picked from a table of random numbers from 1 to 100. If only a few sandstone beds were exposed within the section, the beds were num- bered, and three beds were selected at random. After a sandstone bed near the randomly selected horizon was sampled, a paired sample was collected from the next higher or lower sandstone bed as determined by a toss of a coin. Each selected sandstone bed was divided into equal units 1—2 feet thick, and the exact sample site Within the bed was chosen at random. The samples averaged 5—10 pounds in weight, depending on the grain size of the sandstone. Weath- ering rinds, lichens, plant roots, concretions, iron-stained zones, large fossil fragments, and other obviously unrepresentative matter were looked for and were excluded where noticed. Every effort was made to sample fresh, unweathered outcrops. After we selected the sample horizon, we generally moved along the strike of the beds to find the least weathered location. If there were no unweathered outcrops, an alternate site was chosen. We also at- tempted to dig in beyond the most weathered part of the outcrop. Where sandstone and mudstone beds alternate, picking the next higher or lower sandstone bed was no problem. In thick sandstone sequences having no obvious bedding, the paired sample was collected at least 20 feet above or below the first sample. Sets of cross strata were regarded as one bed. In conglom- eratic sequences, samples were collected from the sandstone bed nearest the randomly selected locality or from the matrix of the conglomerate. In the field, recognition and distinction of sandstone from sandy varieties of other rock types generally was not diffi- cult. Beds of sandy and conglomeratic mudstone were included in our definition of sandstone although their classification as such may be questionable. A more difficult distinction to make was between calcareous sandstone and sandy limestone. Although we at- tempted to exclude limestones, two samples collected from the Salina area of the Wasatch Plateau con— tain slightly more than 50 percent carbonate min- erals. METHODS OF ANALYSIS One chip was removed from each sample for a thin section, and the rest was crushed and split for rapid rock analysis of major elements, direct-reader spectrometric analysis of minor elements, and X-ray- diffraction analysis of major minerals. Color was described from the ground and pelletized samples, using the color chart of Goddard and others (1948) for uniformity. Samples submitted for chemical and spectrometric analyses were arranged in a random sequence so that any potential analytical drift would be converted to random error. To estimate the pre- cision of the analytical methods, 10 splits of one composite sample were included with the other sam- ples without the analysts’ knowledge. Results of the replicate analyses of the 10 hidden splits are shown in table 2. Good precision is indicated by the rela- tively small geometric deviations. The precision of analysis for the 216 samples is probably not this good, because the precision for each constituent varies somewhat with the amount of the constituent present. Rapid rock analyses were performed in the Washington laboratory of the US. Geological Survey TABLE 2.-——Am’thmetic mean, standard deviation, geometric mean, and geometrw dematwn of the analysis of 10 splits from one sample [Major-constituent analysis, rapid rock method, by Lowell Artis, S. D. Botts, G. W. Chloe, Paul Elmore, J. S. Glenn, James Kelsey, and H. Smith; minor-element analysis, direct-reader spectrometric method, by N. M. Conklin, R. G. Havens, and L. M. Lee] Standard Arithmetic Geometric Geometric mean deviation mean deviation Major constituents (in percent) Si02 54.83 0.38 54.81 1.01 A1203 11.20 .07 11.20 1.01 1Fe203 3.77 .05 3.77 1.02 M30 4.28 .14 4.28 1.03 0:10 7.41 .68 7.38 1.09 Na20 1.40 .08 1.40 1.06 K20 3.84 .14 3.84 1.04 1120+ 3.17 .46 3.14 1.15 T102 .39 .03 .39 1.07 P205 .14 .04 .14 1.41 MnO .08 .04 .07 2.2 002 7.48 .21 7.48 1.03 Minor elements (in parts per million) B 41 16 38 1.6 Ba 2,350 412 2,318 1.8 Co 14 2 14 1.1 Cr 49 6 48 1.1 Cu 25 1 25 1.05 Ga 19 2 19 1.1 N1 23 3 23 1.0 Pb 213 29 210 22.1 Sr 845 236 817 1.3 V 84 15 83 1.2 Zr 104 20 103 1.2 1Total iron as Fe203. _ 2Includes 5 samples reported to contain less than 20 ppm for which an arbitrary value of 5 ppm was assigned. METHODS OF ANALYSIS 13 using the single solution wet chemical method de- scribed by Shapiro (1967). Direct-reader spectro- metric analyses were performed in the Denver laboratory of the US. Geological Survey. A descrip— tion of the direct-reader spectrometric method fol- lows. The precision of the analytical results as it affects our interpretations of the geochemistry of lower Eocene fluvial sandstones is further discussed in the section on “Analysis of Variance.” DIRECT-READER SPECTROMETRIC ANALYSES By RAYMOND G. HAVENS and ALFRED T. MYERS Direct-reader spectrometric analysis of sandstone has evolved in the US. Geological Survey from a three-step semiquantitative method of analysis de- scribed by Myers, Havens, and Dunton (1961) and a later six-step semiquantitative method used with conventional spectrographs. When the direct-reading spectrometer is used, photoelectric recording and measurement of spectral line intensity replace the photographic recording of conventional spectro- graphs. The precision and accuracy of the direct- reader method as outlined here apply specifically to the sandstone samples in this report. Synthetic standards are prepared from high-pur- ity chemicals containing known concentrations of each element. By successive dilutions with a common matrix, a series of standards is obtained that covers a wide concentration range (Myers and others, 1961, p. 209—210, 225—228). Working curves covering the desired ranges of concentration for each element are established from these prepared standards. Samples to be analyzed are ground to about 100 mesh or finer. A 10-mg (milligrams) portion of the sample is weighed and is thoroughly mixed with 20 mg of pure graphite powder. The sample-and-graph- ite mixture is then tamped into the cavity of a graphite electrode and is excited in a d-c arc. Table 3 gives the analytical conditions for the direct-reader spectrometric analysis of sandstone. During the excitation period of the sample, radia- tion from the excited sample passes through the optics of the spectrometer and strikes the photo- multiplier tubes which convert the radiation energy into electric current. By the end of the exposure period, the current from a photomultiplier has charged a capacitor to a specific voltage which is proportional to the integrated intensity of the asso- ciated spectral line and, therefore, to the concentra- tion of the particular element in the sample. The spectral lines used for the elements reported are shown in table 4, along with the width of the exit slits and the concentration range covered for each element. Table 2 indicates the precision obtained from 10 hidden splits that were submitted at the 4907247 0 - 73 , 3 time of analysis. In table 5 the results obtained on the US. Geological Survey standard rock samples G—1 and W—1 are compared with the recommended values (Fleischer, 1969) for the standard rocks. TABLE 3,—Analytical conditions for the direct-reader spectro- metric analysis of sandstone Spectrometer ..................................... Consolidated Electrodynamics Corp. 3-meter, type 22-101. Grating ............................................... 21,000 lines per inch. Dispersion... ..3.9I A/mm. Spectral regi 2100—5200 A. Slit width (entra Excitation source... Voltage (open circuit). Current (short circuit) Analytical gap ............ Exposure at 6 amp Exposure at 16 am Exposure total... Sample electrode Counter electrodes. 20 sec. Ultra Carbon Corp., type 3170—U2. ...Made from Ultra Carbon Corp. high-purity l45-inch graphite rod. Graphite powder ............................... Ultra Carbon Corp. UCP—2, conducting. TABLE 4. —Spectral lines used for the elements reported and the concentration ranges covered - Width of Concentration Element ?E:Ctsiilmlr:§f exit slit range covered g (microns) (parts per million) 2,496.78 50 30— 1,000 4,554.03 50 10—10,000 3,453.50 100 8— 1,000 4,274.80 100 5— 1.000 3,247.54 100 1— 1.000 2,943.64 50 10— 1,000 3,414.76 50 7— 1,000 2,833.06 50 20- 1,000 4,607.33 50 10— 5,000 4,379.24 100 10— 1,000 3,279.26 100 20— 1,000 TABLE 5,—Comparison of direct-reader spectrometric analyses with recommended values [All values are in parts per million] El t Sample G—1 Sample W—l emen Direct reader Recommended1 Direct reader Recommended1 B ............................ <30 1.5 <30 15 Ba 1,100 1,200 200 180 2.4 42 50 16 22 150 120 15 13 100 110 22 18 24 16 <7 1—2 76 78 43 49 <20 8 330 250 200 180 21 16 270 240 Zr... 180 210 100 100 1All values, except those for B, are from Fleischer (1969, table 5), who listed recommended values for most elements based on determinations by various methods from different laboratories. For B, the above figures are listed as suggested magnitudes. X-RAY DIFFRACTION ANALYSES The relative mineral composition of samples was determined by X-ray diffraction analysis of Whole rock splits. These data are expressed as inches of deflection of the X-ray diffractometer recording needle (peak heights) above background. This method does not determine absolute quantities of minerals detected but does permit comparison of approximate mineral contents between samples. Hand-ground and pelletized samples were scanned from 2° to 60° 26 at a rate of 2° per minute on a Norelco X-ray diffractometer using nickel-filtered CuK alpha radiation generated at 48 kilovolts and 20 milliamperes and a geiger-counter. Reflection inten- sities, or peak heights, were measured above back- ground, using the principal peak for each mineral 14 GEOCHEMISTRY 0F LOWER EOCENE SANDSTONES IN THE ROCKY MOUNTAIN REGION as follows: Quartz, 4.26 A; potassium feldspar, 3.24 A; plagioclase, 3.19 A; calcite, 3.04 A; dolomite, 2.89 A; laumontite, 9.15 A; clinoptilolite, 9.00 A ; and analcime, 5.61 A. The clay minerals having principal peaks between 11 and 15 A were grouped under the name “mixed-layer clays,” abbreviated MLC, and probably include montmorillonite, chlorite, and mix- tures like illite-montmorillonite. Minerals having a principal peak at 10 A were grouped under the term “mica” but may include some illite and mixed-layer clays. The 7—A clays were grouped under the term “kaolin.” Inasmuch as detailed analyses would be needed to identify positively any specific clay min- eral, quotation marks are used in referring to clay species. Increased iron content can affect the peak heights by raising the background of the diffractogram. How- ever, in the lower Eocene sandstone samples the iron content generally was not large enough to seriously affect the comparisons. Chert, which was measured with quartz, makes the quartz peak slightly broader and lower than it would be if quartz alone were pres- ent. These are semiquantitative measurements, which are sufficient for comparative purposes but which have neither the accuracy nor the precision for quan- titative data. Because difi'erent minerals reflect the X-rays at difierent intensities, the data as they are presented here should be used only to compare amounts of one mineral in different samples, not amounts of different minerals in one sample. The limits of detection range from about 2 to 5 percent depending on the composition of the sample and on the crystallography of the specific mineral. Though semiquantitative, the data from the X-ray analyses are useful for a classification of the sandstones in- cluded in this study (fig. 3). PETROLOGY AND MINERALOGY All but the most friable samples of sandstone were examined in thin section for textural relations and for accessory minerals not detected by X-ray diffrac- tion analysis. Heavy-mineral and clay-mineral sepa- rates were also made on selected samples, and min- eral identifications, where needed to supplement the thin—section studies, were made by X-ray or optical methods. The texture and mineralogy of the samples are described in table 1. MINERAL COMPOSITION AND CLASSIFICATION OF SAMPLES Quartz and varying proportions of chert, potas— sium feldspar, plagioclase, mica, calcite, and dolomite are the principal framework minerals in most sam- ples. In most samples they occur as separate mineral grains; in some they are in composite grains such as granite, schist, shale, limestone, dolomite, and sand- stone fragments. To classify the lower Eocene sand- stone samples, a triangular diagram was constructed, Q Quartz ‘ Pattern shows ‘ enlarged Carbonate '9 Feldspar FIGURE 3. —— Average composition of lower Eocene sandstone in 36 areas in the Rocky Mountain region. Mineral data normalized from X-ray diffraction analyses for quartz, feldspars, and carbonate minerals. Minor amounts of zeo- lites were included with the feldspars for some areas. Clay‘ minerals, mica, and other accessory minerals were not in- cluded. Some areas contain minor amounts of feldspars and carbonates that are not represented in this figure because they fall below the limit of detection by the X-ray method. Numbers within each circle are keyed to the areas listed in table 1 and are shown in figure 1. FIGURE 4. — Photomicrograph of quartz-Cheri; sandstone from the Willwood Formation, Hillberry Rim area, Bighorn Basin, Wyo., showing chert grains (ch), which are readily distinguished by their mottled appearance under crossed polars, and clear quartz grains (Q). Crossed polars; sam- ple BH03. PETROLOGY AND MINERALOGY 15 having quartz, feldspar, and carbonate as end mem- end members in each set of six samples was roughly bers (fig. 3). The mean content, in percent, of the estimated from the X-ray diffraction analyses, and 110° 195° l MONTANA 45a l l r_.________—————-————————_i SOUTH DAKOTA WYOMING i i \___#______ NEBRASKA 41°— COLORADO SAN DSTONE TYPES ' i:i , Quartz rich Feldspar rich Low carbonate Carbonate rich Law feldspar 37°— '1 t i i i i __L __.——————-—‘—-——' iOKLAHOMA ———._— NEW MEXICO Intermediate Carbonate plus feldspar i l I SANDVSTONE COLORS ' Each circle represents six samples I I DO Yellow Orange Red 1 I l | TEXAS I l 100 MILES L___ié_4 1 FIGURE 5. — Distribution of lower Eocene sandstone types, and colors of the sandstone in powdered samples in 36 areas in the Rocky Mountain region. 16 GEOCHEMISTRY 0F LOWER EOCENE SANDSTONES IN THE ROCKY MOUNTAIN REGION sample sets were plotted on the diagram. The dia- gram was then divided into four parts: (1) quartz rich (greater than 90 percent quartz and chert), (2) feldspar rich (10 percent or more plagioclase and potassium feldspar but less than 10 percent car- bonates), (3) carbonate rich (10 percent or more calcite and dolomite but less than 10 percent feld- spar), and (4) intermediate (greater than 10 per- cent feldspar and greater than 10 percent carbonate). Zeolites were included with the feldspars in a few samples, but mica, clay minerals, and accessory min- erals were not included in the calculations. The sandstone samples from the Hillberry Rim area of the Bighorn Basin (figs. 3—5) are the only samples that can be classified as quartz rich accord- ing to this scheme. These samples have a clay matrix, mostly “kaolin,” and they contain about 5 percent potassium feldspar and a few grains of plagioclase. Most of the feldspar—rich sandstone is in the eastern part of the Rocky Mountain region (fig. 5). This fact reflects the dominantly crystalline source terrane (fig. 1) of the sediments in basins to the east. The feldspar-rich samples are generally poorly sorted and contain angular grains (fig. 6). Intermediate samples are in the western and west-central parts of the region. A typical intermediate sample is shown in figure 7. The carbonate-rich samples occur in a band that wraps around the northern, western, and south- central parts of the region and in the Cerrillos area in the Gallisteo basin. In most thin sections ex- amined, the carbonate minerals occur in lithic fragments (fig. 8) as well as in the matrix as a ce- menting agent. The carbonate-rich samples are generally near source terranes composed of older sedimentary rocks, including Paleozoic limestones and dolomites that probably contributed carbonate t0 the lower Eocene sandstones. Etching and partial dissolution of framework grains were observed in many thin sections. Quartz grains are usually etched in samples having a cal- cite matrix (fig. 9). Feldspar grains are commonly altered along fracture and cleavage planes. The plagioclase grains range in appearance from unal- tered to almost completely altered, regardless of matrix material (fig. 10). Sericite and clay are the most common minerals replacing plagioclase, but zeolites and carbonates also replace feldspars. Some of the more common nonopaque heavy min- erals that were noted in both thin sections and heavy- mineral separates are garnet, zircon, tourmaline, and epidote. Samples from the Firehole Basin and the Erickson—Kent Ranch areas contain garnets large enough to appear out of hydraulic equilibrium with the quartz and other framework grains. Their large size suggests unusual conditions of deposition, FIGURE 6. — Photomicrograph of arkosic sandstone from the Wind River Formation, Wallace Creek area, Wind River Basin, Wyo., showing poorly sorted, closely packed, angular grains of quartz (Q) and feldspar (F) in a clay matrix. Crossed polars; sample WR15. FIGURE 7.—Photomicrograph of an intermediate sandstone, Wasatch Formation, Oregon Buttes area, Great Divide Basin, Wyo., showing abundant feldspar (F) and calcite (ca). Crossed polars; sample GDO5. PETROLOGY AND MINERALOGY 17 FIGURE 8,—Photomicrograph of lithic sandstone from the Wind River Formation, Lost Cabin area, Wind River Basin, Wyo., showing very poorly sorted fragments of dolomite (D), chert (ch) , and quartz (Q) in a matrix of iron-stained carbonate. Crossed polars; sample WR09. FIGURE 9.—Photomicrograph of etched quartz grain from the Wasatch Formation, Woodruff area, Wasatch Range, Utah. Outline of original quartz grain (Q) is no longer discernible; the present ragged outline is the result of re- placement by calcite (ca). Crossed polars; sample WMO6. FIGURE 10. — Photomicrograph of altered feldspar grain from the Hanna Formation, Difiiculty area, northern Hanna basin, Wyoming. A large plagioclase grain (P) has been partly altered to sericitic mica, which forms the mosaic of minute light-colored crystals. The original plagioclase twinning is faintly visible in the upper part of the grain and in the dark unaltered region surrounding an embay- ment by calcite (ca). Crossed polars; sample HN06. although most of these samples are derived from obvious channel sandstones. In the thin sections, the nonopaque minerals ranged in amounts from none (Hagen area) to as much as 5 percent (Hoback basin). Hornblende and augite were rarely observed, even in the heavy-mineral separates. They seem un- usually sparse in these areas relative to their abun- dance in modern stream and terrace deposits derived from some of the same source areas (Denson, 1969, p. C29). CEMENT AND MATRIX Cementing and matrix minerals compose 5-20 per- cent of the total rock in most lower Eocene sandstone samples and may comprise as much as 50 percent of the rock in a few samples. With one or two excep- tions, none of the sandstone examined in thin sec- tions could be described as clean or well sorted (table 1). In many samples the wide range in the size of framework grains obscured any clear distinc- tion between framework grains and matrix or ce- menting minerals. A wide variety of matrix minerals was found in the lower Eocene sandstones. Clay minerals include abundant “kaolin” in the Enos Creek, Carbon basin, Cerrillos, Hoback basin, Battlement Mesa, Horse Creek—Crazy Woman Creek, Soldier Summit, and 18 GEOCHEMISTRY OF LOWER EOCENE SANDSTONES IN THE ROCKY MOUNTAIN REGION Morgan areas (1, 12, 5, 14, 16, 18, 27, and 32 in fig. 1). “Montmorillonite,” “chlorite,” or “mixed- layer clays” are abundant in the Hagen, Battlement Mesa, Huerfano Park, Little Snake River, and S01- dier Summit areas (6, 16, 19, 21, and 27 in fig. 1). Detrital clay minerals were difficult to distinguish from authigenic clay minerals, although some clay could be recognized as authigenic where a single clay mineral species forms a uniformly textured matrix, filling cavities throughout the rock (fig. 11), or where the clay mineral forms oriented coatings on framework grains (fig. 12) or partly replaces frame- work grains. FIGURE 11.—< GD 534 23.4 913 9.9 35.1 15.4 20.1 17.0 15.2 359 46 22 203 16 3.86 8.65 Geometric mean + GD... . 39 7.2 189 1.9 4.8 4.3 6.9 4.3 3.8 53 10.4 4.2 63 4.9 1.54 2.12 Geometric means of analyses of six samples from each of 36 areas Locality (area in fig. 1) Bighorn Basin. Wyo.: Enos Creek'(1) ..... 175 ...... 250 10 37 12 ...... 26 11 108 22 16 93 9.8 ‘1.71 ’6.25 Hillberry Run (2) 36 41 279 3.2 6.8 9.0 ...... 7.8 6.4 31 26 11 241 ...... ‘2.12 *4.28 Fossil basin, W30: Evanston( ) ................................................................ 201 13 289 7.8 29 10 ...... 13 129 29 19 121 .................. Fossil (4) 229 ...... 189 10 9.2 7.8 ...... 8.1 8.3 98 20 14 96 ...... *3.37 '3.64 Galisteo basin. N. Mex.: Cerrillos (S) 98 12 376 2.6 3.9 6.0 ...... 4.5 6.4 73 14 ...... 105 ...... ‘.92 ‘2.28 Haeen (6).... 133 ...... 757 6.1 5.7 8.4 ...... 5.3 .... 186 31 ...... 74 11 ‘1.06 '3.94 Great Divide Basm, Wyo Oregon Buttes (7) ................. 1022 16 529 12 44 14 ...... 24 6.5 362 60 9.0 93 15 *2.43 *4.42 Crooks Gap—Wamsutter (8) _ 65 ...... 418 2.8 6.5 6.1 13 5.8 22 79 18 6.6 91 « 18 ‘4.41 *13.80 22 GEOCHEMISTRY OF LOWER EOCENE SANDSTONES IN THE ROCKY MOUNTAIN REGION TABLE 7. —Distribution of constituents in 216 samples of lower Eocene sandstone and in 36' areas and four color subsets of the samples — Continued Minor ' in parts per million Mn B Ba Co Cr Cu Ti Ni Pb Sr V Y Zr Ga Ul Th1 Geometric means of analyses of six samples from each of 36 areas—Continued Green River Basin, Wyo.: La Barge (9)_ 349 13 550 7.6 53 10 18 21 8.5 435 45 15 201 15 ‘4.31 ‘ 14.41 Fireholge Basin (10) 442 16 393 7 9 37 10 13 19 8.4 207 36 12 146 9.6 *2.04 ‘7.22 Hanna basin, Wyo.: Difficulty (11)... 24 ...... 405 ...... 4.0 4.1 ............ 88 10 ...... 66 5.6 ‘2.26 *2.14 Carbon basin (1 223 13 331 4.8 21 6.3 19 8.3 6.6 127 27 12 150 11 *.91 ‘4.16 Hoback—western Wind asins, Wyo Dubois (13)... ..... 202 ...... 398 5.0 39 7.8 15 17 8.2 263 31 8.4 93 12 ............ Hoback basin 371 ...... 382 7.9 24 8.4 13 10 6.4 128 29 16 108 9.2 ‘2.45 *3.97 Piceance Creek basin C Rangely (15) ..... 134 20 281 2.6 9.5 6.0 13 6.6 8.2 113 22 8.0 139 6.0 ‘1.82 *4.24 Battlement Mesa 92 12 486 3.6 6.5 7.4 15 5.9 6.3 70 23 6.9 189 8.2 ‘ 1.24 ‘5.36 Powder River Basin, Wyo Pine Ridge—Pumpkin Buttes (17).. . 37 23 505 3.2 7.9 5.8 ...... 4.2 8.2 57 21 10 178 7.3 2.07 5.61 Horse Creek-Crazy Woman Creek . 483 ...... 302 6.4 24 11 ...... 12 ...... 190 30 19 103 7.9 2.11 6.18 Raton basin, Colo.: Huerfano Park (19) . 254 1254 4.0 11 11 ...... 7.0 7.9 251 32 ...... 80 16 .93 3.50 West Spanish Peak (2 217 378 ______ 10 13 8.8 6.3 165 28 8.4 93 13 1.97 5.46 Sand Wash basin, Colo.: Little Snake River (21)...113 ...... 600 2.5 9.4 4.3 20 5.5 18 164 17 8.8 91 10 3.12 13.24 Elkhead Mountains (22). 48 17 568 3.6 8.0 6.8 ...... 6.3 6.3 98 32 ...... 104 8 1 1.86 5.13 San Juan Basin, N Mex - Cuba (23) ...... 42 427 ...... 3.7 4.1 13 3.9 56 11 7.2 74 6.7 '1.88 ‘2.65 Gobernador ( . 301 897 4.4 9.6 5.8 ...... 8.3 8.1 217 33 7.9 127 11 *3.19 ‘4.45 Shirley Basin, Wyo.: Bates Hole (25)... 50 ...... 712 2.5 5.5 5.4 14 5.5 6.3 218 24 ...... 75 14 6.78 29.22 Chalk Hills (26) 25 ...... 448 ............ 4.5 18 ...... 19 77 10 6 4 102 11 3.28 17.51 Uinta Basin. Utah: Soldier Summit (27) 437 ...... 1500 9.7 24 16 ...... 12 8 5 178 48 18 135 13 *3.22 ‘10.32 Patmos Head 8) 19 740 9.4 37 24 17 19 18 246 84 8 8 145 14 *2.20 ‘8.86 Wasatch Plateau, Utah. 53111121 (29) ............................ 419 3.2 6.3 8.3 ...... 4.1 731 12 6.4 118 ...... ‘2.84 *2.72 Gunnison Plateau (30). 31 388 3.2 14 11 13 11 203 27 8.7 207 9 0 *2.13 '7.66 Wasatch Range, Utah: Woodrutf (31).. 18 64 4.3 17 8.1 ...... 8.1 ...... 56 11 19 129 “‘ .61 ‘0.72 Morgan (32) ........... 150 4.4 16 7.0 ...... 7.7 ...... 86 14 27 82 5.7 ‘1.27 *2.62 Washakie Basin, Wy .. Baggs (33) 16 408 3.6 23 13 ...... 1 8.1 146 38 15 127 9.6 '2.84 ‘7.59 Erickson—Kent Ranch (34). 21 403 2.5 19 4 ...... 8.4 ...... 192 24 11 141 7.6 '2.86 *8.06 Wind River Basin, Wyo.: Lost Cabin (35) ...... 20 227 4.8 27 11 13 12 10 117 20 14 132 8.2 2.28 4.22 Wallace Creek (36)... 61 ...... 630 2.7 16 8.2 ...... 12 15 143 20 6.9 57 14 6.52 11.46 Geometric means of analyses of samples from four subsets grouped on the basis of color 40 red sandstones ......... 14 451 4.1 12 9.6 10 7.5 6.7 117 23 10 116 8.7 ...... 67 orange sandstones... 13 341 3.8 11 7.1 11 8.1 7.0 133 21 8 9 110 7.7 98 yellow sandstones .. 13 437 4 2 13 7.7 13 8.1 8.1 133 25 9 5 108 9.3 11 green sandstones ............................................................. 396 20 632 10 54 17 16 27 11 394 58 13 174 17 ............ Minerals, in inches of deflection MLC Mica Kaolin Quartz P8532532? ngagslg' Calcite Dolomite Zeolite Geometric means and geometric deviations of analyses of 216 lower Eocene sandstone samples Geometric mean... 0.008 0.06 0.06 1. 74 0.49 0.16 0.04 0.013 0.002 Geometric dev1at1on 18 13 151.5 14 6 40 5 Geometric mean >< GD. .143 .775 .880 2. 54 7.09 4.22 1.62 .375 .019 Geometric mean + GD..... .0004 .0046 .0041 1. 19 .034 .006 .0009 .0004 .0004 Geometric means of analyses of six samples from each of 36 areas Locality (area in fig. 1) Bighorn Basin, Wyo.: Enos Creek. ( 1) ........ 0.033 0.83 1.93 0.22 0.006 0.014 0.059 Hillberry Rim (2)... ........ .013 .095 2.59 .052 ........................ Fossil basin, Wyo.: Evanston (3). ........ .21 .19 1.71 .079 ........ 1.92 .87 ~Fossfl (4) ............... .014 .27 2.29 .010 .13 1.94 1.12 Galisteo basm, N. Cerrillos (5).. .003 .42 2.46 .35 .038 .14 Hagen (6) .22 .014 .007 1.43 2.03 1.19 .007 Great Divide Basin, .. Oregon Buttes (7)y ................... .002 .50 .016 1.35 2.48 2.79 .18 . Crooks Gap—Wamsutter (8). .13 .46 .14 1.41 3.59 2.02 .009 .008 Green River Basin, Wyo.: La Barge (9) ............ .76 .011 1.31 .34 .20 .11 .13 Firehole Basin (10) .002 .54 .11 1.64 .90 1.59 .26 .16 Hanna basin, Wyo.: Difficulty (11) .............. .083 .066 1.65 3.77 1.84 .004 ................ Carbon basin (12) ........ .26 .68 1.62 .45 .26 .033 .033 ........ Hoback—western Wind Dubms (13) ............. .071 .25 .095 1.23 .39 .43 .48 .14 ........ Hoback basm (14) -- .051 .077 .34 1.78 .24 .25 .55 .069 ........ Piceance Creek basin, Co 0.. Rangely (15) ............... .018 .015 .052 2.55 .81 .095 .28 .12 ........ Battlement Mesa (16) .14 .013 .52 2.36 1. 82 .33 .007 ................ Powder River Basin, Wyo.. Pine Ridge—Pumpkin Buttes (17) ............. .003 .21 . 2.53 1 17 .016 ................ .004 Horse Creek-Crazy Woman Creek (18) .. ........ .42 .37 .93 2.2 .018 1.74 .13 ........ Raton basin, Colo.: Huerfano Park (19) ......... .17 .070 .19 1.72 4.93 .88 .003 ................ West Spanish Peak (20).. .070 .020 .006 1.94 .32 2.95 .............. .23 Sand Wash basin. Colo.: Little Snake River (21)... .18 .49 .066 1.78 3.56 1.08 .008 ................ Elkhead Mountains (22) .010 .030 .092 2.03 1.72 .077 ........................ San Juan Basin, N. Mex.: Cuba (23) ............. .003 .014 .17 2.22 1.50 .040 ........................ Gobernador (24). ........ .002 .002 1.60 2.03 4.00 .024 ................ CHEMICAL COMPOSITION OF SANDSTONES 23 TABLE 7. — Distribution of constituents in 216' samples of lower Eocene sandstone and in 36' areas and four color subsets of the samples—Continued Minerals, in inches of defection MLC Mica Kaolin Quartz $33313“ P3551: Calcite Dolomite Zeolite Geometric means of analyses of six samples from each of 36 areas—Continued Shirle Basin, W 0.: Bites Hole (y25) ....... . 0.28 0.006 1.41 4.70 3.41 ........................ Chalk Hills (26) ....... .009 .083 .092 1.91 4.71 .10 ................ Uinta Basin, Utah: Soldier Summit (27) .29 .19 .76 1.63 .035 1.07 2.00 1.59 ........ Patmos Head (28) ....... .037 .28 .017 1.25 53 .43 083 .54 0.20 Wasatch Plateau, Utah: Salina (29) ................................................................... .013 .013 .005 2.05 .10 .014 2.17 1.08 ________ Gunnison Plateau (30) ............................................... .14 .13 .033 1.87 .65 1.64 .037 .003 .006 Wasatch Range. Utah: Woodrufi (31) . .002 ........ .046 1.70 ................ .27 .008 ........ Morgan (32), ........ .015 .35 1.45 .021 ........ 3.94 .14 ........ Washakie Basin, Baggs (33) ........ .008 .033 .032 1.77 .77 .26 .042 ................ Erickson-Kent Ranch (34) ............................................... .46 .28 1.59 1.18 1.26 .14 .025 ........ Wind River Basin, Wyo.: Lost Cabin (35) ....... .008 .064 .007 2.00 .12 .010 .17 .093 ........ Wallace Creek (36) ..... .003 .23 .015 1.62 4.18 2.90 .002 .003 ........ Geometric means of analyses of samples from four subsets grouped on the basis of color 40 red sandstones ..... . 0.028 0.075 1.84 0.17 0.063 0.064 0.012 0.003 67 orange sandstones .004 .023 .043 1.87 .35 .11 .051 .015 .001 98 yellow sandstones .008 .14 .089 1.69 .93 .24 .027 .011 .001 11 green sandstones .013 .89 .009 1.22 .58 1.99 .062 .024 .008 ‘Uranium and thorium data based on stats] of only 86 samples analyzed. ‘Single sample. in the distribution of the original data. The distribu- tion patterns of the different constituents vary greatly among the sample areas. Much of this vari- ation may arise because six samples cannot give a very complete picture of the distribution patterns. In the color subsets, the distribution patterns are similar to those shown in the complete set of 216 samples. For the 216 lower Eocene sandstone samples, the concentration range of the major elements or oxides is shown in figure 15, and that of the minor elements is shown in figure 16. For comparison, on both fig- ures we plotted the mean values for an average sand- stone as estimated by Pettijohn (1963, p. 811, 815). Pettijohn computed his average from selected pub- lished analyses after estimating that the average sandstone would include 26 parts graywacke, 25 parts lithic sandstone, 15 parts arkose, and 34 parts orthoquartzite. Pettijohn’s averages closely corre- spond to the means computed for the lower Eocene sandstones for SiOQ, A1203, CaO, K20, H20—, H20+, 002, Ba, Cr, Cu, Ga, Pb, and V. Total iron as Fe203, MgO, Na20, T102, P205, MnO, B, and Zr are less in the lower Eocene sandstones than in the average sandstone. Co, La, Ni, Sr, Y, Th, and U are greater in the lower Eocene sandstone than in the average sandstone. Many replacement values in the data sets for B, Co, La, Pb, and Y (table 6) and the small number of analyses for U and Th reduce the accu- racy for those elements. The two sets of values probably reflect similarities and differences in prove- nance, lithologic types, environments of deposition, and diagenetic history. To check whether distinctive groups of minor ele- ments might characterize certain end-member types of sandstone, subsets of samples were chosen in which the mean of one mineral constituent exceeded by one standard deviation the arithmetic mean of that constituent in the total set of 216 samples. The mean value was calculated for each constituent in these subsets and was compared with the mean value for the full set of 216 samples. No unusual concen- trations of minor elements were found in any of these end-member types of sandstone. Additional subsets were chosen in which a combination of two minerals was highly concentrated or in which one mineral was very low, and still no unusual concen— trations of minor elements were found. This seems to indicate that the distribution of minor elements is independent of the major-mineral composition of the samples. Perhaps the minor-element distribution is affected more by diagenesis, provenance, and weath- ering than by the proportion of major-mineral con- stituents. This indication is further supported by the differ- ences between the geometric means of four color sub- sets. The 216 samples were distributed as follows: 40 red, 67 orange, 98 yellow, and 11 green.2 The dis— tribution of the colors is shown in figure 5, and the number of samples from each section in each color subset is listed in table 1. Little is known about the geochemical significance of color, although iron oxides are commonly re- garded as the principal coloring agent in sandstones. 2Colors are based on ground sample color and were arbitrarily selected from the “Rock-Color Chart” of Goddard and others (1948) as follows: Red, 1013 6/2, 10R 6/6, 10R 7/4, 71? 7/4, 10R 5/4, 5YR 8/1, 5YR 7/2, 5YR 8/4, 5R 8/1; orange, 10YR 8/2, IOYR 7/4, IOYR 8/6, 10YR 7/6. lOYR 6/4, 5YR 5/6, 5Y1? 4/1, 5YR 6/4: yellow. 5Y 7/2, 51" 8/1, 5Y 8/4, 5YR 6/1, N8. N9; green, 5Y 7/1, 51' 6/1, 51' 5/2. 561’ 7/2, 561’ 8/1, 5G 8/1, 101’ 8/2. 24 GEOCHEMISTRY 0F LOWER EOCENE SANDSTONES IN THE ROCKY MOUNTAIN REGION PERCENT 0.01 0.1 1 10 100 O Si02 A O A1203 0 A l 8203 O _' A MgO A 030 O C O A 0 N30 A K20 O —.._.— O EXPLANATION ‘ + H20— 0 ——.'—— O Geometric mean and approximately ‘ two-thirds range of data A TiO Highest and lowest geometric mean of 0+ 0 stratigraphic sections sampled A A P205 Average sandstone of Pettijohn (1963) Q — Q A (—0 M 0 3 .n _ 0 Mean is below limit of detection or below A scale CO; <— O ‘ t Q Total C < O O FIGURE 15.— Concentration range of major constituents in 216 samples of lower Eocene sandstone in the Rocky Mountain region. The color is related more to the oxidation state of the iron and to the mineralogy of the iron oxide and hydrate minerals than to the amount of iron (Walker, 1967) . Among the lower Eocene sandstones, the red sandstone subset contains less total iron than any of the other three color subsets. The green sandstone subset contains nearly 21/2 times as much total iron as the red sandstone subset and more of every con- stituent except Si02, quartz, “mixed layer clay,” “kaolin,” potassium feldspar, and calcite than any of the other three subsets (table 7). In the green sandstone samples, the geometric means of cobalt, chromium, copper, nickel, strontium, vanadium, gal- lium, and “mica” each exceed the geometric mean of the 216 samples by at least one geometric deviation. The mean peak height of “mica” in the green sand- stone subset is about 15 times higher than the mean of the 216 samples and is more than 6 times higher than the mean of the yellow subset. This high “mica” content may be related to the abundance of iron and minor elements in the green sandstones: by lowering the permeability of the green sandstones the “mica” could prevent the leaching of iron and minor ele- ments by oxidizing solutions. Also, micas can con- tain higher concentrations of some minor elements, including barium, manganese, chromium, titanium, and vanadium (Deer and others, 1962, p. 31, 61), than can most other common sandstone minerals such as quartz, feldspars, calcite, or kaolin, although micas in general do not scavenge minor elements as readily as do other clay minerals or organic minerals. Greenish mica may be partly responsible for the GEOCHEMICAL RELATIONS color of some of the sandstone samples, although green chlorite and epidote are also present in some of the samples. As can be seen in figure 5, the green color is not closely related to provenance or to sandstone type. The green sandstone samples were mostly intermedi- ate, but two were carbonate rich, and one was feld- spar rich. Green sandstone was collected in the Green River Basin, Great Divide Basin, Bighorn Basin, Wind River Basin, and the Wasatch Plateau, and green-tinted sandstone was noticed in some of the other basins. Both crystalline and sedimentary terranes were contributing debris to these areas (fig. 1). Possibly the development and preservation of the green color and the abundance of minor ele- ments are related to the depositional environments, but more likely the color was produced by diagenetic 25 alteration. This would explain abrupt color changes commonly observed along a single bed and among beds in stratigraphic sequence. However, more in- vestigation is needed before all the factors influenc— ing color can be evaluated. GEOCHEMICAL RELATIONS ANALYSIS OF VARIANCE The four-level hierarchial sampling plan previ- ously described was designed to allow use of analysis of variance in evaluating the relative importance of differences in element and mineral concentrations as functions of geographic or stratigraphic position. The computer program follows the computational technique of Anderson and Bancroft (1952). We used a four-level nested model in the analysis of variance, similar to the model used by Krumbein and PARTS PER MILLION 1 10 100 1000 10,000 Ba O _O—A O O Sr A Zr 0 ©—...—‘ Q__8_ AL. _ _ _ _A Cr A.-___A e B o L—n‘ La ‘71 «O """H EXPLANATION Y + ‘— G"'—.-— O Geometric mean and approximately two—thirds range of data Ga A. - _ - J . O 0 Ni Highest and lowest geometric mean of ‘ ‘_ O _—._— O stratigraphic sections sampled Cu 0 Range is below limit of detection H— ‘ Pb ‘L—_' A Average sandstone of Pettijohn (1963) C Range of average sandstone of Pettijohn {9-6— 0 (1963) «O ‘1 Th 0 Mean or average is below limit of detection ___ ' . or below scale _3_ FIGURE 16. — Concentration range of minor elements in 216 samples of lower Eocene sandstone in the Rocky Mountain region. 26 GEOCHEMISTRY OF LOWER EOCENE SANDSTONES IN THE ROCKY MOUNTAIN REGION Slack (1956) except that the distance between levels could not be held constant. The computations were made on the log-transformed data. Constituents having many indeterminate values (table 6) are not included in this discussion. Table 8 lists the percent- age of variance components at each of the four levels. The following constituents show significant vari- ance between basins: A1203, total iron as Fe203, NagO, K20, H20—, Ba, Cr, Cu, Ni, “mica,” and po- tas'sium feldspar. All the constituents tested except total iron, Cu, and “mica” show significant variance between sections within basins, and all except total iron, Ti02, MnO, Ni, and Zr show significant vari- ance between paired beds within sections. All the constituents except plagioclase, for which no test of the significance was made, show a large percentage of the total variance between samples. A1203, “mica,” and potassium feldspar show their greatest percent- age of variation at the basin level. MgO, CaO, Na20, K20, and 002, “kaolin,” and plagioclase show the greatest variation at the section level. Considered together, these two levels indicate significant re- gional variations in the feldspars and carbonate minerals that are mappable, as showu in figure 5. Sr shows the greatest variation between paired beds within sample areas, which probably reflects the ex- tremely high Sr contents (due to celestite cement) of several samples from the Salina area (table 1). The remaining constituents all show their greatest variation between samples. The variance between samples has several causes. Analytical error is included at this level, but, as can be seen from table 2, the analytical error is a small part of the total variance between samples for most constituents. Variance at this level also reflects the effects of microgeochemical environments; for ex- ample, the high variance shown by TiO2 and Zr may be partly due to differences in concentrations of heavy minerals, such as zircon and ilmenite, that commonly occur in a single bed. The variance of total iron and MnO could reflect differences in the degree of weathering of the samples and differences in diagenetic alteration due to unequal porosities of two beds. The variance between pairs of beds can be explained by local environmental differ- ences at the sites of deposition or by changes resulting from progressive erosion of the source ter- rane. The variance shown by the two lower levels of the sampling plan may be regarded as “back- ‘ GI‘QUD A .. - ,. Kaolin. O Calcite . ‘ ' : bolomite .Quartz $0.51: FIGURE 17.—Vector diagram for 216 samples of lower Eocene sandstone. “K-spar” is po- tassium feldspar, “Plag” is plagioclase, and “MLC” is mixed layer clays. GEOCHEMICAL RELATIONS 27 3 ground noise” that masks some or all of the region- ally significant geochemical trends. TABLE 8.—Variance components (in percent) [Numbers in italic indicate significant variance at the 95-percent level. Note that no test for significance was made at the lowest level, between samples within paired beds] « Variance components ........... Between Between Between Between basins sections paired beds samples within within within within region basins sections paired beds 0 35.6 25.8 38.6 . 33.7 20.8 20.5 24.7 15.0 9.9 11.7 63.4 . 16.5 38.1 8.1 37.3 . 6.2 41.1 18.1 34.6 25.3 28.0 21.0 25.8 25.4 29.2 19.8 25.6 8.5 12.0 10.6 68.9 30.9 16.9 13.9 38.3 6.3 15.1 0.8 77.9 17.9 27.2 14.4 40.6 1.4 16.3 3.1 79.2 15.5 33.6 17.6 33.3 27.5 15.7 14.3 42.5 28.6 30.0 11.4 30.4 24..) 4.7 12.9 58.2 24.0 30.1 3.7 42.3 12.8 21.8 1.0.1 25.3 19,7 27.9 14.3 38.2 1.2 15.8 4.0 79.0 33.9 9.1 25.5 31.5 0 40.3 26.3 33.4 10.7 22 8 17.4 49 1 32.6 23.9 14.6 28.8 Plagioclase ............................... 20.2 34.2 27.6 18.0 1Total iron as Fe203. The variance components in table 8 show signifi- cant differences between basins for all of the con- stituents that could be tested in the lower Eocene sandstones. This indicates that, for most constitu- ents, the sampling plan was reasonably adequate and that not much more information would have been gained from a few more samples. The feldspar min- erals and their component constituents, A1203, K20, and NaZO, show about the same variance between sample areas as they do between basins. Constituents that show significant variation between sample areas but not between basins include P205; those constitu- ents that occur in resistate minerals (including Si02, quartz, Zr, and Ti02), those that form car- bonate minerals (CaO, MgO, and 00.). Si02 and quartz are probably so ubiquitous that an unreason- ably large number of samples would be required to define variation at the basin level. Variance percent- ages for P205, Zr, TiO-_,, and the carbonate-mineral components probably reflect a greater difference be- tween source areas for different sample areas within the same basin than between basins within the re- gion. Another group, including Fe203, Cu, and “mica,” shows significant variance between basins, but not between sections. This indicates that the principal regional variation is on a very large scale and that additional samples would be required to define the variance between sample areas. COMPONENT ANALYSIS Component analysis of correlation data supple- ments the interpretation of geochemical associations of the various constituents detected. The technique used is similar to that previously described in a geo- chemical study of black shales (Vine and Tourtelot, 1969, p. All—A12; 1970b). Component analysis, as described by Harman (1960) , Imbrie and VanAndel (1964), Griffiths (1966), and Miesch, Chao, and Cuttitta (1966) , involves constructing radial vectors for each constituent so that the cosine of the angle between any two vectors is equal to the coeflicient of correlation. The vectors are then rotated and pro- .jected onto a two-dimensional diagram on which varimax axes are selected that will best represent the data. In these vector diagrams, the radial lines are omitted, and the position of each constituent is shown within a circle of unit radius. Segments of the diagram are patterned to suggest possible geo- chemical groupings of the constituents. The illustra- tions can be readily compared for geochemical interpretations. Constituents that plot close together near the outer rim of the circle usually have signifi- cant positive correlations with each other. Constitu- ents that plot close together near the center of the circle have short vectors, indicating a large angle with the plane of projection, and may or may not have significant correlations with each other. Con- stituents that plot on opposite sides of the circle have strong negative correlations; those that plot at right angles from each other show no correlations. Short vectors that are poorly represented on the two-di- mensional plot may be close together or far apart on a three-dimensional plot. Correlation data that are part of a closed array (the sum of the parts equals 100 percent) have cer- tain inherent biases, as described by Chayes (1960) , that cannot be avoided. For example, if a rock has three principal constituents, there must necessarily be a negative correlation between the predominant constituent and the other two. A positive correlation may be imposed on certain pairs of minor constitu- ents, even though there is no geochemical association between the pairs, because they both vary inversely with the predominant constituent. However, in our experience, the technique is both useful and valid because it leads to geochemical interpretations that are for the most part reasonable and consistent with geochemical theory. Figure 17 shows a vector diagram constructed from principal-component analysis of the correlation matrix for 29 constituents in the 216 samples of lower Eocene sandstones from the Rocky Mountain region. Eigenvalues show that about 50 percent of the variation in the data is represented. This low representation is probably partly due to the variety of sandstone types included in the sample set. How- ever, the representation is sufficient to give a gross 28 GEOCHEMISTRY OF LOWER EOCENE SANDSTONES IN THE ROCKY MOUNTAIN REGION picture of the geochemical relations. The array of points on the diagram may be divided into three major and two minor groups, as indicated by the patterned areas. Near the top of the diagram is a group (A) composed of quartz and Si02. The sepa- ration of the points representing these two constitu- ents is mostly a function of the amount of SiO;» in minerals other than quartz, although the separation may also reflect some bias in the X-ray data. An inverse correlation between this group and the other four groups is an indication of the predominance of quartz over other minerals and illustrates the prob- lem of closure. 0n the lower right side of the dia- gram is a segment characterized by potassium feldspar, plagioclase, K20, Na20, A1203, and barium, referred to as the feldspar group (B). On the left side is a segment referred to as the carbonate group (C) that includes calcite, dolomite, 002, and CaO. Most of the remaining constituents are scattered be- tween the carbonate and feldspar groups. The seg- ment adjacent to the carbonate group is designated group D and includes P205, MnO, MgO, Co, Cr, Ni, and Cu. The segment adjacent to the feldspar group contains total iron, “mica,” Ga, H20+, Ti02, Sr, and V and is designated group E. Group D is prob- ably more closely related to the carbonate group, and group E to the feldspar group, although there is not unequivocal separation of all constituents. Groups D and E tend to correlate with each other as well. Vector diagrams (fig. 18) were constructed for each of the four color subsets previously described (p. 23). Although these diagrams and the vector diagram for the 216 samples (fig. 17) are generally similar, certain differences are evident that may have geochemical significance. The vector diagram for the red sandstones (fig. 18A) shows a closer correlation of “mica,” “MLC,” and Ti02 with the feldspar group and Cu and Sr exchange places in groups D and E as compared with their positions in the 216 sand- stone samples. The differences between the two dia- grams are partly a result of greater homogeneity among the red sandstone samples than among the samples in the complete set. For example, in the vec- tor diagram for the red sandstones, group E may represent mostly the constituents found in the hema- tite that gives the color, whereas in the vector dia- gram for the 216 variously colored sandstones, group E includes constituents of a variety of hydrous iron oxide minerals. The close association of V and Cu with iron (fig. 18A) suggests that these two ele- ments occur almost exclusively in the iron oxide min- erals in the red sandstones, whereas in the 216 sandstones Cu must occur in other constituents as well, as suggested by its position in group D (fig. 17 ). For the ofange subset the vector diagram (fig. 183) is most like that of the whole set and repre- sents a similar mixture of sandstone types, as indi- cated in figure 5. For the yellow subset (fig. 180), which contains the most samples of the four subsets, the vector dia- gram shows some major differences when compared with the diagram for the entire group of samples. Group D includes the calcite group, and group E has moved away from the feldspar group and is almost indistinguishable from the calcite group. Ga is in group B rather than group E and is more defi- nitely associated with the feldspar group. Several constituents, including Zr, “kaolin,” Ti02, H20+, and “mica”, show no close association with any of the groups. Inspection of the correlation coefficients reveals that Zr tends to correlate positively with both quartz and Ti02 at the 99-percent confidence level. “Kaolin” shows positive correlations with calcite and dolomite; TiOQ, with Zr and with total iron; and H20+, with group E constituents. “Mica” shows no significant positive correlations. The yellow subset is composed principally of very faintly colored sam- ples. Although the content of total iron (table 7) is close to that of the whole set of 216 samples and to that of the orange subset, both color and correlation statistics suggest that, in the yellow subset, the con— tent of iron oxide minerals may be significantly lower and that more of the iron occurs in carbonate minerals than in iron oxide minerals. Sr seems to be the only element that occurs about equally in the car- bonate group and in the feldspar group. The position of Ga in the feldspar group also suggests a decrease in iron oxide minerals. Ga may occur in iron oxide minerals and, more likely, in aluminum minerals such as the feldspars (Goldschmidt, 1954, p. 327). Inspection of the correlation coefl‘icients and of the vector diagram suggests that, in the yellow sand- stones, the minor elements, except Ba, Ga, and part of the Sr, occur more with carbonate and phosphate minerals (probably mostly the dolomite) than with the iron oxide minerals, whereas in the orange and red subsets, they occur more with the iron oxide minerals. For the green subset the vector diagram (fig. 18D) is distinctly different from that of the other subsets. The three major groups—quartz, feldspar, and car- bonate——retain their same general orientation, with respect to each other, but group E has disappeared, and group D has lost most of its constituents and is on the other side of the calcite group. A new group, F, composed of total iron, Cu, “kaolinite,” “MLC,” and H20+, can be differentiated next to group D, but the strong correlation shOWn by total iron with “kaolin,” “MLC,” and H20+ in group F suggests GEOCHEMICAL RELATIONS that most of the iron occurs in micaceous and clay minerals. Groups D and F are inversely related to the feldspar group. Potassium feldspar is positioned near the quartz group, perhaps reflecting a predomi- nance of plagioclase over potassium feldspar (table 7). Many constituents fall close to the center of the circle and show no strong association with any group. The general scattering of those constitu- ents means partly that 11 samples arc’insufficient for correlation statistics and partly that, in the green subset, the elements are partitioned differ- ently among the major constituents than in the other subsets. For instance, the correlation coefficients show that Co has a positive correlation with potas- sium feldspar at the 99-percent confidence level. TABLE 9: —— Areas where any constituent mean value for six samples deviation of all samples (+) or is less than one geometric mean 29 Some elements—for example Ga, Sr, and MnO—are about equally associated with two inversely re- lated groups. The correlation coefficients are below the 90-percent confidence level for any positive cor- relations for K20 and Ba. Other elements—for ex- ample Ni and V—show strong positive correlations only between each other. The inverse relation of group F to the feldspar group suggests that the con- stituents of group F may replace plagioclase diage— netically in the green sandstone. GEOCHEMISTRY OF SAMPLE AREAS Table 9 lists areas that are rich in various con- stituents; the areas and constituents are grouped in a sequence suggested by geochemical similarities. is greater than one geometric mean times the geometric divided by the geometric deviation of all samples (0) CONSTITUENT AREA (figure 1) SiO2 TiO2 + Ti Zr B A1201 NaZO K20 Ga 002 +total C Pb “MLC” Ba Total Fe MnO Sr Cr N V Co Cu MgO Dolomite Calcite CaO P20 Zeolite “Kaolin” K-spar Plagioclase + Quartz 2, Hillberry Rim 23. Cuba +++ 17. Pine Ridge-Pumpkin Buttes 30. Gunnison Plateau + + + + 25. Bates Hole + + 24. Gobernador + 26. Chalk Hills 8. Crooks Gap-Wamsutter l9, Huerfano Park ++++ + 36. Wallace Creek 0 + 21. Little Snake River ++ 6. Hagen 28. Patmos Head + + 1. Enos Creek 7. Oregon Buttes ++ + + 27. Soldier Summit 9. La Barge 101 Firehole Basin 13. Dubois 29. Salina . . ++ o+++ 4. Fossil 32. Morgan 0 O O O + O O 18. Horse Creek-Crazy Woman Creek 0 0 3. Evanston O O O O... +++++ 20. West Spanish Peak 22. Elkhead Mountains 5. Cerrillos ooo+++++ + + 1 . Difficulty O ._. 3 . Woodruff . O O O .— 30 GEOCHEMISTRY OF LOWER EOCENE SANDSTONES IN THE ROCKY MOUNTAIN REGION .Zr u - ,, . Kaolln .“MLC” .Tota] Fe Group E FIGURE 18.—Vectot diagrams for color subsets of the sandstone samples: feldspar, “Flag” is plagioclase, GEOCHEMICAL RELATIONS 9%. A Group B ' ‘ potassium D, 11 green. “K-Spar” 15 A, 40 red; 3, 67 orange; C, 98 yellow; and “MLC” is mixed layer clays. 31 32 GEOCHEMISTRY OF LOWER EOCENE SANDSTONES IN THE ROCKY MOUNTAIN REGION The constituents were arranged first, beginning with quartz, in a sequence suggested by the correlation statistics. The areas were then arranged in a related sequence, beginning with the Hillberry Rim area because it is rich in quartz. The Hillberry Rim area is also rich in Si02 and Zr, so these are the next constituents in the sequence. The Cuba area, listed next because it is rich in Si02, is followed by the Pine Ridge—Pumpkin Buttes area because it is rich in Si02 and B. The Gunnison Plateau area follows in fourth place because it is rich in Zr, B, and Ti. These four areas can be classified as rich in resistate minerals such as quartz, chert, zircon, tourmaline, and rutile or in ilmenite. The Hillberry Rim and Cuba areas are notably lean in many other constituents. The arrangement of the remaining areas and con- stituents was continued in much the same manner to produce a diagonal scatter diagram. The areas rich in A1203, Na20, K20, and Ga are all characterized by large amounts of feldspar (fig. 3), as would be expected, although neither po- tassium feldspar nor plagioclase occurs in amounts large enough to be classified as rich according to the formula described in table 9. The areas rich in these constituents include Bates Hole, Gobernador, Chalk ‘ Hills, Crooks Gap—Wamsutter, Huerfano Park, and Wallace Creek, a list that is slightly less inclusive than the areas shown as feldspar rich in figures 3 and 5. The constituents Pb, Ba, and “MLC” are abundant in areas with high amounts of feldspar as well as in areas with high amounts of several other minor ele- ments. The Chalk Hills area is notably lean in the minor elements Cr, Co, Ni, and V. Although the data are not complete enough to permit their inclusion in table 9, U and Th contents are higher in the feld- spar-rich sandstones. Both U and Th are abundant in the Wallace Creek, Bates Hole, Crooks Gap—Wam- sutter, and La Barge areas. In addition, Th is abun- dant in the Chalk Hills, Little Snake River, Patmos Head, and Soldier Summit areas. Some of these areas are classed as intermediate in figures 3 and 5. Another major group of constituents consists of total iron, Sr, Cr, Ni, Co, Cu, MgO, and dolomite. One or more of these constituents are rich in the following areas: Enos Creek, Dubois, La Barge, Ore- gon Buttes, Fossil, Firehole Basin, Soldier Summit, Patmos Head, and Salina. Rocks in these areas are mostly intermediate sandstones (figs. 3 and 5), ex- cept at Enos Creek, Fossil, and Salina, where they are carbonate rich, and form a band across western Wyoming to south-central Utah. Total iron and dolo- mite appear to be the only major constituents related to the increased contents of these minor elements. The Patmos Head area has the greatest number of constituents classified as rich, including Na20, zeo- lite, total iron, dolomite, MgO, Cr, Ni, V, and Cu. The high content of Na20 is probably related to the high content of zeolite, or analcime (table 1) more than to feldspar content (fig. 3 and table 7). The broad spectrum of the constituents in high amounts and the presence of analcime suggest that the com- position of the Patmos Head samples may be the result of diagenetic concentration of some constitu-i ents. In general, the position of the band of areas rich in minor elements suggests that both source terrane and diagenesis affected the composition of the areas. These areas, although they probably re- ceived detritus from several different source terranes (fig. 1) are all near Paleozoic sediments that include enriched black shale units such as the Meade Peak Phosphatic Shale Member of the Phosphoria Forma- tion (Vine and Tourtelot, 1970b). The older sedi- ments could be a source for the minor elements which either were deposited directly with the sand- stones or, more likely, were later concentrated in the sandstones of these sections by ground water. C02, calcite, Ca, and P205 are abundant in the Evanston, Horse Creek—Crazy Woman Creek, Mor- gan, Fossil, and Salina areas. Except for the Fossil area, which is rich in Co, these areas show no high values for other constituents. DISCUSSION The source, degree of mobility, and ultimate fate of the minor elements in the lower Eocene sand- stones remain subjects for speculation, for the sedi- ments, even though they are geologically young, have had an extremely complex geochemical history. Al- though the data for B are admittedly censored by a large number of indeterminate values, there is a positive correlation between high B content and high quartz content. A weak positive correlation also exists between quartz, Ti, and Zr. These minor ele- ments probably occur in resistate minerals such as tourmaline, zircon, and ilmenite and, because of their resistance to chemical decomposition, tend to be con- centrated in second-cycle sediments along with quartz. In contrast to the minor elements locked up in resistate minerals, Co, Cr, Cu, Ni, V, and Sr appear to be more readily mobilized, perhaps by alkaline carbonate solutions. Of these minor ele- ments, all but Cu are richer in some areas than in any average granitic or sedimentary source terrane, as listed by Turekian and Wedepohl (1961, table 2). This greater local abundance may indicate some con- centration of these elements during deposition and diagenesis. The correlation statistics (figs. 17, 18 and table 9) show that these elements tend to correlate with secondary constituentssuch as dolomite, iron REFERENCES CITED 33 oxides, and clays, and that they tend to change asso- ciations in the different color groups (figs. 17, 18), again suggesting mobility. The apparent depletion of minor elements in red sandstone suggests leaching by oxidizing solutions, perhaps by oxidizing meteoric ground waters intro- duced into permeable aquifers during late Cenozoic uplift and erosion as suggested by Adler (1970, p. 334). SUMMARY Lower Eocene sandstones in the basins of Wyo- ming, Utah, Colorado, and New Mexico show large variations in chemical and mineral composition. Feldspar-rich and carbonate-rich sandstones are common, and intermediate sandstones, containing significant quantities of both feldspar and carbonate minerals, are characteristic of southwestern Wyo- ming and eastern Utah (figs. 3, 5). Sandstones from only one area can be classified as quartz rich. Most of the sandstone samples are poorly sorted and con- tain varying amounts of clay minerals. Evidence of diagenetic alteration and the formation of authigenic clay, carbonate, zeolite, and sulfate minerals is pres- ent in all areas (table 1). In many samples quartz has been etched, and the mafic minerals have been partly to completely dissolved. Although the quantity of mafic minerals such as hornblende in a sediment is primarily controlled by provenance and by the amount of weathering and alteration during erosion and sediment transport, subsequent diagenetic alter- ation is a major factor in the preservation of the minerals. Evidence of low-grade metamorphism was noted in the Raton basin. The laumontite-rich sand- stones of the West Spanish Peak area are most likely the result of reactions of the feldspar-rich sand- stones with heated ground water, and thus may be classified as hydrothermally altered. All the constituents tested showed significant vari- ance components either between basins or between sample areas or both. A significant component of variance was found at both levels for A1203, Na20, K20, H20—, Ba, Cr, Ni, “mica,” and potassium feld- spar. The following constituents, however, showed significant variation only between sample areas and not between basins: Si02, MgO, CaO, H20+, TiOQ, P205, MnO, C02, Sr, V, Zr, “kaolin,” quartz, and plagioclase. Total iron as Fe203, Cu, and “mica” showed significant variation between basins but not between sample areas. These relations suggest that significant differences exist in the content of feld- spars and some clay minerals whereas the content of carbonate minerals and resistate minerals varies on a smaller scale and would require additional sam- pling to show useful geographic differences. The variance between basins and the variance between sample areas are probably both reflections of the variety of source areas; sedimentary source terranes are dominant in western Wyoming and Utah, and granitic source terranes are dominant in central Wyoming, Colorado, and New Mexico. The composition of the average lower Eocene sandstone is similar in many respects to that of the average sandstone of Pettijohn (1963), which rep- resents a composite of various sandstone types. The differences between the two sets of values can prob- ably be explained chiefly on the basis of differing proportions of the end-member types of sandstone. The data are inadequate to distinguish between dif- ferences resulting from various environments of deposition, diagenesis, or provenance, although we suspect that these factors may explain the higher concentrations of some minor elements in the aver- age lower Eocene sandstone. The highest concentrations of minor elements occur in a group of sample areas in western Wyo- ming and northeastern Utah extending from the western Bighorn Basin to the southern Wasatch Plateau. Within this area one to five of the minor elements Co, Cr, Cu, V, Ni and Sr, exceed the geo- metric mean of all samples by at least one geometric deviation. Dolomite and (or) MgO are also abundant in four of the sample areas in this band, and calcite is abundant in two. Although the area rich in minor elements includes a number of intermediate sand- stone samples, none of them is rich in feldspar alone or in quartz. On the basis of incomplete data for U and Th, some correlation seems to exist between high U and feldspar-rich sandstone and between high Th and feldspar-rich or intermediate sandstone. A significant difference was found between color subsets of the samples, the greatest difference being between the red and green subsets. The green sand- stones are rich in iron, “mica,” and many of the minor elements. Some early diagenetic redistribution of minor ele- ments is suggested by the greater concentration of the elements in intermediate rocks rather than in quartz-, feldspar-, or carbonate—rich rocks. Later mobility is suggested by the apparent depletion of minor elements in red sandstones that were prob- ably exposed to oxidizing meteoric solutions. REFERENCES CITED Adler, H. H., 1970, Interpretation of colour relations in sand- stone as a guide to uranium exploration and ore genesis, in Uranium exploration geology, Vienna, 1970, Proc. Vienna, Austria, Internat. Atomic Energy Agency, p. 331—344. 34 GEOCHEMISTRY OF LOWER EOCENE SANDSTONES IN THE ROCKY MOUNTAIN REGION Ahrens, L. H., 1954, The lognormal distribution of the ele- ments (a fundamental law of geochemistry and its sub- sidiary): Geochim. and Cosmochim. Acta, v. 5, no. 2, p. 49—73. Anderson, R. L., and Bancroft, T. A., 1952, Statistical theory in research: New York, McGraw-Hill Book Co., 399 p. Antweiler, J. C., and Love, J. D., 1967, Gold—bearing sedimen- tary rocks in northwest Wyoming—A preliminary re- port: U.S. Geol. Survey Circ. 541, 12 p. Baltz, E. H., 1967, Stratigraphy and regional tectonic impli- cations of part of Upper Cretaceous and Tertiary rocks, east-central San Juan Basin, New Mexico: U.S. Geol. Survey Prof. Paper 552, 101 p. Barghoorn, E. S., 1953, Evidence of climatic change in the geologic record of plant life [Chap.] 20, in Shapley, Har- low, ed., Climatic changes—evidence, causes, and effects: Cambridge, Mass., Harvard Univ. Press, p. 236—248. Bell, K. G., and Hunt, J. M., 1963, Native bitumens associ- ated with oil shales, Chap. 8, in Berger, 1. A., ed., Organic geochemistry: New York, Macmillan Co., Internat. Ser. Mon. on Earth Sci., v. 16, p. 333—366. Bradley, W. H., 1963, Paleolimnology, in Frey, D. G., ed., Limnology in North America: Madison, Wis., Univ. Wis- consin Press, p. 621—652. Bradley, W. H., 1964, Geology of Green River Formation and associated Eocene rocks in southwestern Wyoming and adjacent parts of Colorado and Utah: U.S. Geol. Survey Prof. Paper 496—A, 86 p. Bradley, W. H., and Eugster, H. P., 1969, Geochemistry and paleolimnology of the trona deposits and associated authigenic minerals of the Green River Formation of Wyoming: U.S. Geol. Survey Prof. Paper 496—B, 71 p. Cashion, W. B., Jr., 1957, Stratigraphic relations and oil shale of the Green River Formation in the eastern Uinta Basin [Utah], in Intermtn. Assoc. Petroleum Geologists Guide- book 8th Ann. Field Conf., Uinta Basin: p. 131—135. 1967, Geology and fuel resources of the Green River Formation, southeastern Uinta Basin, Utah and Colo- rado: U.S. Geol. Survey Prof. Paper 548, 48 p. Chayes, Felix, 1960, On correlation between variables of con- stant sum: Jour. Geophys. "Research, v. 65, no. 12, p. 4185-4193. Coombs, D. 8., Ellis, A. J., Fyfe, W. S., and Taylor, A. M., 1959, The zeolite facies, with comments on the interpre- tation of hydrothermal syntheses: Geochim. et Cosmochim. Acta, v. 17, p. 53—107. Culbertson, W. C., 1964, Oil shale resources and stratigraphy of the Green River Formation in Wyoming: Mtn. Geolo- gist, v. 1, no. 3, p. 181. 1965, Tongues of the Green River and Wasatch For- mations in southeastern part of the Green River Basin, Wyoming, in Wyoming Geol. Assoc. Guidebook, 19th Ann. Conf., Sedimentation of Late Cretaceous and Tertiary outcrops, Rock Springs uplift, 1965: p. 151—155. 1969, Facies changes in the Eocene rocks in the southeastern part of the Green River basin, Wyoming, in Intermtn. Assoc. Geologists (and Utah Geol. Soc.) Guidebook 16th Ann. Field Conf., Uinta Mountains, Utah’s maverick range: Salt Lake City, Utah, Utah Geol. Survey, p. 205—211. Deer, W. A., Howie, R. A., and Zussman, J., 1962, Sheet silicates, v. 3 of Rock-forming minerals: New York, John Wiley & Sons, Inc., 270 p. Denson, N. M., 1969, Distribution of nonopaque heavy min- erals in Miocene and Pliocene rocks of central Wyoming and parts of adjacent States, in Geological Survey re- search 1969: U.S. Geol. Survey Prof. Paper 650—C, p. 025—032. Dobbin, E. C., Bowen, C. F., and Hoots, H. W., 1929, Geology and coal and oil resources of the Hanna and Carbon Basins, Carbon County, Wyoming: U.S. Geol. Survey Bull. 804, 84 p. Donnell, J. R., 1961a, Tertiary geology and oil-shale resources of the Piceance Creek basin between Colorado and White Rivers, northwestern Colorado: U.S. Geol. Survey Bull. 1082—L, p. 835—891. 1961b, Tripartition of the Wasatch formation near De Beque in northwestern Colorado, in Short papers in the geologic and hydrologic sciences: U.S. Geol. Survey Prof. Paper 424—B, p. B147—B148. Dorr, J. A., Jr., 1952, Early Cenozoic stratigraphy and verte- brate paleontology of the Hoback Basin, Wyoming: Geol. Soc. America Bull., v. 63, no. 1, p. 59—93. Evernden, J. F., Savage, D. E., Curtis, G. H., and James, G. T., 1964, Potassium—argon dates and the Cenozoic mammalian chronology of North America: Am. Jour. Sci., v. 262, no. 2, p. 145—198. Fisher, D. J., Erdmann, C. E., and Reeside, J. B., Jr., 1960, Cretaceous and Tertiary formations of the Book Cliffs, Carbon, Emery, and Grand Counties, Utah, and Garfield and Mesa Counties, Colorado: U.S. Geol. Survey Prof. Paper 332, 80 p. Fleischer, Michael, 1969, Additional data on rocks G—1 and W—l, 1965—1967, pt. 1 of U.S. Geological Survey stan— dards: Geochim. et Cosmochim. Acta, v. 33, no. 1, p. 65—79. Gazin, C. L., 1962, A further study of the lower Eocene mammalian faunas of southwestern Wyoming: Smith- sonian Misc. Colln., v. 144, no. 1 (Smithsonian Inst. Pub. 4474), 98 p. Gill, J. R., Merewether, E. A., and Cobban, W. A., 1970, Stratigraphy and nomenclature of some Upper Creta- ceous and lower Tertiary rocks in south—central Wyo- ming: U.S. Geol. Survey Prof. Paper 667, 53 p. Goddard, E. N., chm., and others, 1948, Rock-color chart: Washington, D.C., Natl. Research Council (repub. by Geol. Soc. America, 1951, 1971), 6 p. Goldschmidt, V. M., 1954, Geochemistry [Alex Muir, ed.]: Oxford, Clarendon Press, 730 p. Griffiths, J. C., 1966, A genetic model for the interpretive petrology of detrital sediments: Jour. Geology, v. 74, no. 5, pt. 2, p. 655—672. Hardy, C. T., and Zeller, H. D., 1953, Geology of the west— central part of the Gunnison Plateau, Utah: Geol. Soc. America Bull., v. 64, no. 11, p. 1261-1278. Harman, H. H., 1960, Modern factor analysis: Chicago, Univ. Chicago Press, 469 p. Harshman, E. N., 1968, Geologic map of the Shirley Basin area, Albany, Carbon, Converse, and Natrona Counties, Wyoming: U.S. Geol. Survey Misc. Geol. Inv. Map 1—539. 1969, Wyoming uranium issue [spec. ed.] of Soc. Economic Geologists Field Conf., 1969: Wyoming Univ. Contr. Geology, v. 8, no. 2, pt. 1, p. 65—149. Hawkins, J. W., Jr., and Whetten, J. T., 1969, Graywacke matrix minerals: Hydrothermal reactions with Columbia River sediments: Science, v. 166, p. 868—870. Hay, R. L., 1966, Zeolites and zeolitic reactions in sedimen- tary rocks: Geol. Soc. America Spec. Paper 85, 130 p. REFERENCES CITED 35 Henderson, G. V., 1958, Geology of the northeast quarter of the Soldier Summit quadrangle, Utah: Brigham Young Univ. Research Studies Geology Ser., v. 5, no. 5, 40 p. Hite, R. J., and Dyni, J. R., 1967, Potential resources of daw- sonite and nahcolite in the Piceance Creek basin, north- west Colorado: Colorado School Mines Quart., v. 62, no. 3, p. 25—38. Hose, R. K., 1955, Geology of the Crazy Woman Creek area, Johnson County, Wyoming: U.S. Geol. Survey Bull. 1027—B, p. 33—118. Imbrie, John, and VanAndel, T. H., 1964, Vector analysis of heavy—mineral data: Geol. Soc. America Bull., v. 75, no. 10, p. 1131—1156. Johnson, R. B., 1959, Geology of the Huerfano Park area, Huerfano and Custer Counties, Colorado: U.S. Geol. Sur- vey Bull. 1071—D, p. 87—119. Johnson, R. E., Wood, G. H., Jr., and Harbour, R. L., 1958, Preliminary geologic map of the northern part of the Raton Mesa region and Huerfano Park in parts of Las Animas, Huerfano, and Custer Counties, Colorado: U.S. Geol. Survey Oil and Gas Inv. Map OM—183. Keefer, W. R., 1957, Geology of the DuNoir area, Fremont County, Wyoming: U.S. Geol. Survey Prof. Paper 294—E, p. 155—221. . 1965, Stratigraphy and geologic history of the upper- most Cretaceous, Paleocene, and lower Eocene rocks in the Wind River Basin, Wyoming: U.S. Geol. Survey Prof. Paper 495~—A, 77 p. L 1970, Structural geology of the Wind River Basin, Wyoming: U.S. Geol. Survey Prof. Paper 495—D, 35 p. King, P. B., 1959, The evolution of North America: Prince- ton, N.J. Princeton Univ. Press, 189 p. Knight, S. H., 1951, The Late Cretaceous-Tertiary history of the northern portion of the Hanna Basin, Carbon County, Wyoming, in Wyoming Geol. Assoc. Guidebook 6th Ann. Field Conf., south—central Wyoming: p. 45—53. Krumbein, W. G., and Slack, H. A., 1956, Statistical analysis of low-level radioactivity of Pennsylvanian black fissile shale in Illinois: Geol. Soc. America Bull., v. 67, no. 6, p. 739—762. Love, J. D., 1964, Uraniferous phosphatic lake beds of Eocene age in intermontane basins of Wyoming and Utah: U.S. Geol. Survey Prof. Paper 474-E, 66 p. 1970, Cenozoic geology of the Granite Mountains area, central Wyoming: U.S. Geol. Survey Prof. Paper 495—0, 154 p. Love, J. D., McGrew, P. 0., and Thomas, H. D., 1963, Rela- tionship of latest Cretaceous and Tertiary deposition and deformation to oil and gas in Wyoming, in Backbone of the Americas: Am. Assoc. Petroleum Geologists Mem. 2, p. 196—208. Masursky, Harold, 1962, Uranium-bearing coal in the eastern part of the Red Desert area, Wyoming: U.S. Geol. Sur- vey Bull. 1099—B, 152 p. McGookey, D. P., 1960, Early Tertiary stratigraphy of part of central Utah: Am. Assoc. Petroleum Geologists Bull., v. 44, no. 5, p. 589—615. McKenna, M. C., 1960, Fossil Mammalia from the early Wasatchian Four Mile fauna, Eocene of northwest Colo- rado: California Univ. Pubs. Geol. Sci., v. 37, no. 1, p. 1—130. Miesch, A. T., Chao, E. C. T., and Cuttitta, Frank, 1966, Multivariate analysis of geochemical data on tektites: Jour. Geology, v. 74, no. 5, pt. 2, p. 673—691. Muffler, L. J. P., and White, D. E., 1969, Active metamor- phism of upper Cenozoic sediments in the Salton Sea geothermal field and the Salton trough, southeastern California: Geol. Soc. America Bull., v. 80, no. 2, p. 157—182. Mullens, T. E., 1971, Reconnaissance study of the Wasatch, Evanston, and Echo Canyon Formations in part of north- ern Utah: U.S. Geol. Survey Bull. 1311—D, 131 p. Myers, A. T., Havens, R. G., and Dunton, P. J., 1961, A spectrochemical method for the semiquantitative analysis of rocks, minerals, and ores: U.S. Geol. Survey Bull. 1084—1, p. 207—229. Oil and Gas Journal, 1970, West Uinta seen as new “Red Wash freeway”: Oil and Gas Jour., v. 68, no. 48, p. 91. Olive, W. W., 1957, The Spotted Horse coalfield, Sheridan and Campbell Counties, Wyoming: U.S. Geol. Survey Bull. 1050, 83 p. Olson, A. B., 1959, Photogeologic map of the Flat Top Moun- tain NE. quadrangle, Carbon County, Wyoming: U.S. Geol. Survey Misc. Geol. Inv. Map I—301. Olson, J. S., and Potter, P. E., 1954, Statistical methods, pt. 1 of Variance components of cross-bedding direction in some basal Pennsylvanian sandstones of the Eastern Interior Basin: Jour. Geology, v. 62, no. 1, p. 26—49. Oriel, S. S., 1961, Tongues of the Wasatch and Green River formations, Fort Hill area, Wyoming, in Short papers in the geologic and hydrologic sciences: U.S. Geol. Sur- vey Prof. Paper 424-B, p. B151—B152. 1962, Main body of Wasatch Formation near La Barge, Wyoming: Am. Assoc. Petroleum Geologists Bull., v. 46, no. 12, p. 2161—2173. Oriel, S. S., and Tracey, J. 1., Jr., 1970, Uppermost Creta- ceous and Tertiary stratigraphy of Fossil basin, south- western Wyoming: U.S. Geol. Survey Prof. Paper 635, 53 p. Pettijohn, F. J., 1963, Chemical composition of sandstones— excluding carbonate and volcanic sands, chap. S. of Fleischer, Michael, ed., Data of geochemistry [6th ed.]: U.S. Geol. Survey Prof. Paper 440—S, p. S1—S21. Privrasky, N. C., 1963, Geology of the Big Piney area, Sub- lette County, Wyoming: U.S. Geol. Survey Oil and Gas Inv. Map OM—205. Rich, E. I., 1962, Reconnaissance geology of Hiland—Clarkson Hill area, Natrona County, Wyoming: U.S. Geol. Survey Bull. 1107—G, p. 447—540. Robinson, Peter, 1957, Age of the Galisteo formation, Santa Fe County, New Mexico: Am. Assoc. Petroleum Geolo- gists Bull., v. 41, no. 4, p. 757. Roehler, H. W., 1965, Early Tertiary depositional environ- ments in the Rock Springs uplift area, in Wyoming Geol. Assoc. Guidebook 19th Ann. Field Conf., Sedimen- tation of Late Cretaceous and Tertiary outcrops, Rock Springs uplift, 1965: p. 141—150. 1970, Nonopaque heavy minerals from sandstone of Eocene age in the Washakie Basin, Wyoming, in Geo- logical Survey research 1970: U.S. Geol. Survey Prof. Paper 700—D, p. D181—D187. Rohrer, W. L., 1966, Geology of the Adam Weiss Peak quad- rangle, Hot Springs and Park Counties, Wyoming: U.S. Geol. Survey Bull. 1241—A, 39 p. Sears, J. D., and Bradley, W. H., 1924, Relations of the Wasatch and Green River formations in northwestern Colorado and southern Wyoming, with notes on oil shale in the Green River formation: U.S. Geol. Survey Prof. Paper 132, p. 93—107. 36 GEOCHEMISTRY OF LOWER EOCENE SANDSTONES IN THE ROCKY MOUNTAIN REGION Shapiro, Leonard, 1967, Rapid analysis of rocks and miner- als by a single-solution method, in Geological Survey re- search 1967: U.S. Geol. Survey Prof. Paper 575—8, p. B187—B191. Sharp, W. N., 1970, Extensive zeolitization associated with hot springs in central Colorado, in Geological Survey research 1970: U.S. Geol. Survey Prof. Paper 700—B, p. Bl4—B20. Sharp, W. N., McKay, E. J., McKeown, F. A., and White, A. M., 1964, Geology and uranium deposits of the Pump— kin Buttes area of the Powder River Basin, Wyoming: U.S. Geol. Survey Bull. 1107—H, p. 541—638. Sheppard, R. A., and Gude, A. J., 3d, 1969, Diagenesis of tuffs in the Barstow Formation, Mud Hills, San Bernar- dino County, California: U.S. Geol. Survey Prof. Paper 634, 35 p. Simpson, G. G., 1948, The Eocene of the San Juan Basin, New Mexico: Am. Jour. Sci., [pt. 1], v. 246, no. 5, p. 257—282; [pt. 2], v. 6, no. 6, p. 363—385. Smith, J. W., and Milton, Charles, 1966, Dawsonite in the Green River Formation of Colorado: Econ. Geology, v.61, n0. 6, p. 1029—1042. Snedecor, G. W., and Cochran, W. G., 1967, Statistical meth- ods [6th ed.] : Ames, Iowa, Iowa State Univ. Press, 593 p. Soister, P. E., 1968, Stratigraphy of the Wind River Forma— tion in south—central Wind River Basin, Wyoming: U.S. Geol. Survey Prof. Paper 594—A, 50 p. Stearns, C. E., 1943, The Galisteo formation of north-central New Mexico: Jour. Geology, v. 51, no. 5, p. 301-319. Stephens, J. G., 1964, Geology and uranium deposits at Crooks Gap, Fremont County, Wyoming, with a section on Gravity and seismic studies in the Crooks Gap area, by D. L. Healey: U.S. Geol. Survey Bull. 1147—F, 80 p. Stokes, W. L., and Madsen, J. H., Jr., compilers, 1961, Geo- logic map of Utah—northeast quarter: Utah Geol. and Mineralog. Survey. Taylor, J. H., 1964, Some aspects of diagenesis: Adv. Sci., v. 21, p. 417-436. Theobald, P. K., 1970, Preliminary geologic map of the north half of the Craig quadrangle, Moffat County, Colorado: U.S. Geol. Survey open-file rept. Tourtelot, H. A., 1946, Tertiary stratigraphy in the north- eastern part of the Wind River Basin, Wyoming: U.S. Geol. Survey Oil and Gas Inv. Prelim. Chart 22. Tracey, J. L, Jr., Oriel, S. S., and Rubey, W. W., 1961, Diamictite facies of the Wasatch formation in the Fossil basin, southwestern Wyoming, in Short papers in the geologic and hydrologic sciences: U.S. Geol. Survey Prof. Paper 424—8, p. B149—Bl50. Turekian, K. K., and Wedepohl, K. H., 1961, Distribution of the elements in some major units of the Earth’s crust: Geol. Soc. America Bull., v. 72, no. 2, p. 175—192. Van Houten, F. B., 1944, Stratigraphy 0f the Willwood and Tatman formations in northwestern Wyoming: Geol. Soc. America Bull., v. 55, no. 2, p. 165—210. __1948, Origin of red-banded early Cenozoic deposits in Rocky Mountain region: Am. Assoc. Petroleum Geolo- gists Bull., v. 32, no. 11, p. 2083—2126. Vine, J. D., 1969, Authigenic laumontite in arkosic rocks of Eocene age in the Spanish Peaks area, Las Animas County, Colorado, in Geological Survey research 1969: U.S. Geol. Survey Prof. Paper 650—D, p. D80—D83. Vine, J. D., and Tourtelot, E. B., 1969, Geochemical investi- gations of some black shales and associated rocks: U.S. Geol. Survey Bull. 1314—A, p. A1—A43 [1970]. ___1970a, Preliminary geochemical and petrographic analysis of lower Eocene fluvial sandstones in the Rocky Mountain region, in Wyoming Geol. Assoc. Guidebook 22d Ann. Field Conf., Wyoming Sandstones Symposium: p. 251—263. 1970b, Geochemistry of black shale deposits—A sum- mary report. Econ Geology, v. 65, no. 3, p. 253-272. Walker, T. R., 1967, Formation of red beds in modern and ancient deserts: Geol. Soc. America Bull., v. 78, no. 3, p. 353-368. Zeller, H. D., and Stephens, E. V., 1969, Geology of the Ore- gon Buttes area, Sweetwater, Suhlette, and Fremont Counties, southwestern Wyoming: U.S. Geol. Survey Bull. 1256, 60 p. U. S. GOVERNMENT PRINTING OFFICE : 1973 0 - 490-247 Stratigraphy, Morphology, and Paleoocology of a Fossil Peccary Herd From Wostern Kentucky By WARREN I. FINCH, FRANK C. WHITMORE, JR., and JOHN D. SIMS GEOLOGICAL SURVEY PROFESSIONAL PAPER 790 Work done in part in cooperation with the Kentucky Geological Survey Nature of the death and burial of the herd and comparison with other fossil peccary herds of North America UNITED STATES GOVERNMENT PRINTING OFFICE, WASHINGTON : 1972 UNITED STATES DEPARTMENT OF THE INTERIOR ROGERS C. B. MORTON, Secretary GEOLOGICAL SURVEY V. E. McKelvey, Director Library of Congress catalog-card No. 72-600269 For sale by the Superintendent of Documents, U.S. Government Printing Office Washington, D.C. 20402 - Price 70 cents (paper cover) Stock Number 240k00248 CONTENTS Page Page Abstract ..................................................................................... 1 Fossils—Continued Introduction .............................................................................. 1 Platygonus compressus Le Conte 1848—Continued Acknowledgments ..................................................................... 1 Composition and inferred habits of the herd ...... 17 Geologic setting _______ 2 Other herds of Platygonus compressus ................ 18 Silt stratigraphy ................................................... 2 Denver, Colo., herd ........................................... 18 Silt petrology ...................................................... 4 Goodland, Kans., herd ............. 19 Mineralogy and size analyses. ................... 7 Sandusky County, Ohio, herd ......................... 19 Interpretation ......................................................... 8 Reading, Ohio, herd ......................................... 19 Silt fabric and paleowind directions .. 8 Columbus, Ohio, herd. ..... 19 Fossils ........................................................................................ 11 Belding, Mich., herd .................. 20 Platygonus compressus Le Conte 1848 ......................... 12 Gainesville, N.Y., herd ......... 2O Positions of the skeletons .................. 12 Pittsford?, N.Y., herd ...................................... 20 Skeleton 1 .......................................................... 12 Size distribution in Platygonus compressus ________ 20 Skeleton 2 .......................................................... 14 Mollusca ..................................................................... 23 Skeleton 3... 15 Age of fossils .................................................................... 23 Skeleton 4 .......................................................... 16 Paleoecology and the death of the herd ................................ 23 Skeleton 5 .......................................................... 17 References cited ........................................................................ 24 ILLUSTRATIONS Page FIGURE 1. Map of Hickman 71/2-minute quadrangle, showing the fossil peccary locality in western Kentucky ................... 2 2. Section showing stratigraphic setting of fossil peccaries in Roxana Silt at White’s gravel pit ........................... 3 3. Chart showing correlation of silt units at the peccary locality with the classification of the Wisconsinan in Illinois ................................................................................................................................................................................ 5 4 Graphs showing X—ray mineralogy, microscopically visible carbonate content, and ¢>-mean of samples 1—18 ..... 6 5. Graph showing mean and standard deviation of clay-mineral content of silt from peccary locality ................... 7 6 Silt-fabric rose diagrams for Roxana Silt and basal 7 feet of Peoria Loess ............................................................. 9 7 Full-circle wind rose diagram based on readings of directions from which wind blew at Barkley Field, Pa- ducah, Ky., and semicircle wind rose of strike directions of same wind data ................................................... 9 8. Silt—fabric rose diagram for samples from peccary zone ............................................................................................... 10 9. Plan View and section showing relative position of the five peccary skeletons ............................................................ 10 10—13. Photographs of peccary Platygonus compressus: 10. Skeleton 1 ................................................................................................................................................................. 13 11. Skeleton 2 and posterior part of skeleton 3 ..... 14 12. Anterior part of skeleton 3 and posterior part of skeleton 4.... 15 13. Parts of skeleton 4 and skeleton 5 ....................................................................................................................... 17 TABLES Page TABLE 1. Measurements of skulls, mandibles, and teeth of Platygonus compressus from near Hickman, Ky., and Den- ver, Colo ............................................................................................................................................................................. 16 2. Measurements of postcranial skeletons of Platygonus compressus from near Hickman, Ky., and Denver, Colo. 21 3. Measurements indicative of stature of Platygonus compressus ..................................................................................... 22 III u .4 mama...» ’ 2M5! STRATIGRAPHY, MORPHOLOGY, AND PALEOECOLOGY OF A FOSSIL PECCARY HERD FROM WESTERN KENTUCKY By WARREN I. FINCH, FRANK C. WHITMORE, JR., and JOHN D. SIMS ABSTRACT Fossils of a herd of five adult peccaries, Platygonus com- pressus Le Conte, were discovered in May 1967 beneath about 65 feet (20 m) of silt. Their position was 4—5.5 feet (1.2—1.6 m) above the base of the Roxana Silt of Wisconsinan age, which rests unconformably on continental gravel deposits of Pliocene(?) and Pleistocene age. The peccaries were confined to an area 3 feet wide and 11 feet long (1 by 3.3 m) and lay with their heads pointed southeastward on an indistinct up- ward—sloping surface. This surface marks a gradational change from mixed alluvial and eolian silt below to entirely eolian silt above. X—ray mineralogic and size analyses and thin-section examinations of samples from the full 11 feet (3.3 m) of Roxana and the basal 7 feet (2.1 m) of the over- lying Peoria Loess confirm the field stratigraphic relations of the two units and indicate several probable paleosols. Silt fabric, based on orientation of quartz grains, shows maxi- mums that fit the dominant southwest and northwest present- day wind directions. Morphology and position of each skeleton are described in detail. The poses of the skeletons indicate that the herd died quietly and was buried quickly, probably overwhelmed in a duststorm as they walked southeastward up a path from the river with their backs to the world. Prob- ably the animals either died of cold or were smothered while comatose because of cold, and they were buried almost imme- diately. Size distribution of other Platygonus compressus herds is compared with that of the Hickman herd. Mollusca associated with the skeletons have a radiocarbon age of >34,000 years B.P. INTRODUCTION In May 1967, an alert heavy-equipment operator, John Henson, unearthed bones of a fossil peccary while removing thick silt overburden from gravel in a. pit about 1.5 miles (2.5 km) southwest of Hick- man, Ky., along the Chickasaw Bluffs of the Missis- sippi River (fig. 1; Morgan, 1967). Mr. Henson im- mediately informed Richard W. White, M.D., owner of the pit. The following day Warren I. Finch, who was mapping the geology of the Hickman 71/2—min— ute quadrangle, visited Dr. White to obtain permis— sion for a field party to visit his properties, and Dr. White told Finch of the discovery. Dr. White unhesi- tatingly gave the U.S. Geological Survey permission to remove the bones. Excavation took about 5 days and disclosed a total of five peccaries. It was done under the direction of John D. Sims and Finch with the assistance of Roger Swanson, Don Russell, and Fred Wreck of the U.S. Geological Survey and of Robert Langford, Dr. White’s nephew. Owing to the fragility of the bones, they were trucked by Finch and Sims to the U.S. National Museum, Washington, DC, where under the direction of Frank C. Whit- more, Jr., they were cleaned for study and display. In the resulting study, presented here, the geologic portion is by Finch, the report on the peccaries is by Whitmore, and the portion on silt petrology is by Sims. All three authors collaborated in the interpre- tation of the data. Original measurements were made in both the English and the metric systems; where possible, equivalent metric values and scales are shown. ACKNOWLEDGMENTS We thank and commend Mr. John Henson, exca- vator, of Columbus, Ky., for his alertness in noticing the bones and in stopping his excavation to save them from destruction. We are grateful to Dr. Rich- ard W. White for allowing us to excavate on his property, which made possible the preservation-of the skeletons. Our appreciation is extended to all those who helped excavate the bones, particularlyto Robert Langford, a student of archeology at Memphis State University, for his capable and spirited assistance in the excavation. The peccary skeletons were pre- pared for study and exhibition by Sigmund J. Sweda, Division of Vertebrate Paleontology, U.S. National Museum; Ralph E. Eshelman, U.S. Geological Sur- vey; and Susan H. Whitmore. Photographs of the skeletons shown in figures 9 through 13 are by Rob- ert H. McKinney and Haruo E. Mochizuki, U.S. Geo- logical Survey. John E. Guilday, Carnegie Museum, made helpful suggestions, as did Clayton E. Ray, U.S. National Museum, who reviewed the paleonto- logical portion of the manuscript. We thank Craig C. Black of the Carnegie Museum and Mrs. Elizabeth Dalvé of the University of Cincinnati for permission 1 2 STRATIGRAPHY, MORPHOLOGY, AND PALEOECOLOGY, PECCARY HERD, KENTUCKY to study specimens at their institutions,1 and we thank G. Edward Lewis, US. Geological Survey, for furnishing measurements and data on the group of peccaries he collected in Denver, as well as for help- ful discussion. GEOLOGIC SETTING The fossil peccary locality is near the axis of the northern part of the Mississippi embayment. The oldest rocks that crop out in the vicinity of the fossil locality are silt and clay of the Jackson Formation of late Eocene age. The Jackson is unconformably overlain at most places by continental gravel de- posits of Pliocene(?) and Pleistocene age. Along the bluffs in the Hickman quadrangle, these gravel de- posits range in thickness from 5 feet to more than 40 feet (1.5—12 m), and at the peccary site they av- erage about 20 feet (6 m). A few miles east of the bluffs in the quadrangle, the gravel deposits are 10- cally absent (Finch, 1971). The gravel deposits, and locally the Jackson Formation, are overlain by one or more blankets of predominantly eolian silt, whose combined thickness in the quadrangle ranges from 45 feet to 80 feet (13.5—24 m). The relations of the various geologic units at the fossil locality are shown in the section in figure 2. SILT STRATIGRAPHY In the area of the fossil locality, the silt can be divided into a lower unit, the Roxana Silt, and an upper unit, the Peoria Loess. The Roxana Silt is characteristically chocolate brown, noncalcareous, and unfossiliferous and in many places forms a fairly sharp contact with the overlying Peoria, which is commonly moderate-yellowish-brOWn cal- careous fossiliferous silt. At the peccary site and at gravel pits 1,000 feet (305 m) to the north and about 1 mile (1.5 km) to the southwest (fig. 1), however, the Roxana is generally various shades of brown, but locally is gray, and is variably calcareous and fos- siliferous. At these localities, the contact between the Roxana and Peoria is difficult to determine. In the Hickman quadrangle, the Roxana is exposed only along and near the Chickasaw Bluffs (fig. 1), where it ranges in thickness from 0 to 15 feet (0—4.6 m). It either is indistinguishable from the Peoria or, as is more likely, is absent from the east half of the quadrangle. The Peoria is exposed throughout the upland area and ranges in thickness from 0 to 65 feet (0—20 m). Its thickness is greatest along the bluffs and thins to 25 feet (7.6 m) or less 1After Whitmore made his measurements of the Carnegie Museum speci- men, the specimen was transferred to the National Museum of Canada, Ottawa. We are grateful to C. R. Harington of the National Museum for providing three additional measurements. at the eastern boundary. The maximum combined thickness of the Roxana and Peoria in western Ken- tucky is 82 feet (25 m) in the Bondurant quadran- gle, about 2 miles (3.2 km) south of the peccary site; at the peccary site the combined thickness is about 70 feet (21 m). xx" 0 3,: locality (9 Fossil peccary locality %V\§Gravel FUL[O_N cougij _ _ KENTUCKY 36 OBION COUNTY _ — TENNESSEE _ ‘ 30’ 89615, O 1 2 MILES 89°07’30 0 1 2 3 KI LO M ET E R8 LL-J—J—J FIGURE 1. — Map of the Hickman 71/2-minute quadrangle, showing the fossil peccary locality in western Kentucky. A measured section of 66 feet (20.1 m) of silt at the peccary site follows. The calcareous character of the unit, determined with dilute hydrochloric acid, begins each description for emphasis. Color names and their numerical designations were determined by use of National Research Council’s rock-color chart (Goddard and others, 1948). Feet Meters Peoria Loess (upper 4: ft excavated) : 8. Noncalcareous silt, light—yellowish—brown (10YR 6/4), clayey, friable, unstratified, nonfossiliferous; A and B soil horizons not present owing to excavation .............. 17 5.2 7. Calcareous silt, very pale brown (IOYR 7/3), slightly clayey, micaceous; abun- dant dark minerals; fossiliferous, most fossils occur singly (fossil mollusk local- ity 1,000 ft (305 m) north contains Deroceras laeve (Miiller) , Discus crank— hitei (Newcomb), Punctum minutissi— 2. Mixed noncalcareous Peoria Loess—Continued mum (Lea), Succinea grosvenori gelida (Baker), Vertigo gouldii (Binney),Hen- dersom‘a occulta Say, and other ter- restrial gastropods according to Browne and Bruder (1963)); nonstratified ........ 6. Calcareous silt, pale—yellowish-gray (5Y 7/2), clayey; abundant dark minerals, micaceous; contains large Gastropoda Allogona cf. A. profunda and Angui- spira cf. A. koehi in lower part; fairly compact; nonstratified ................................ Total measured thickness of Peoria Loess ...................................................... Roxana Silt: 5. Mostly noncalcareous silt but some spots . weakly calcareous, light-yellowish-brown (10YR 6/4) to medium-yellowish-brown (10YR 5/2), clayey; abundant dark minerals; abundant root casts where cal- careous; micaceous; nonfossiliferous; compact; nonstratified ................................ 4. Noncalcareous and subordinately calcar- eous silt, grayish-orange (10YR 7/4) to moderate-yellowish—brown (10YR 5/4), slightly clayey; abundant dark minerals; contains carbonized wood frag- ments as much as 3 cm long and, where calcareous, abundant carbonate root casts; compact; nonstratified ____________________ 3. Mixed noncalcareous and calcareous silt, moderate-yellowish-brown (10YR 5/4); Liesegang—like bands of grayish-orange (10YR 7/4) and dark-yellowish-orange (10YR 6/6), sparse clay, abundant dark minerals; pebbly; contains smooth-sur- faced spheroidal nodules of calcium car— bonate called loess kindchen 5—20 mm in diameter; contains fossil peccaries Platygonus compressus Le Conte and gastropods Anguispz'ra kochi that occur singly but more commonly in clusters 5—8 m in diameter; pebbles in peccary zone consist of brown well-rounded to angular chert from granule size to 6 cm long, yellowish—white quartz as much as 15 mm across, yellow and red quartzite as much as 2 cm across, and one silici- fied crinoid stem section; compact; non- stratified ......................................................... and subordinately calcareous silt, light—brown (5YR 6/4) to moderate-yellowish-brown (10YR 5/4), slightly clayey, micaceous; abun- dant dark minerals; sparse very fine quartz grains in lower 1 ft (0.3 111); some carbonate root casts; very sparse brown chert pebbles; locally contains sparse to abundant carbonized wood fragments 1—4 mm across in upper 1 ft (0.3 m) of dark-yellowish-brown (10YR 4/2) silt; a few clusters of gastropod Anguispira kochi; compact; nonstrati- fied .................................................................. SILT STRATIGRAPHY 3 Feet 32 55 2.5 1.5 3 Meters 9.8 .76 .46 Feet M eters Roxana Silt—Continued 1. Noncalcareous clayey silt, very pale brown (10YR 7/2) to moderate-yellowish—brown (10YR 5/4) ; sparse coarse to very coarse quartz grains, very sparse brown chert pebbles; compact, nonstratified; intergrades with underlying gravel ________ 1 .3 Total thickness of Roxana Silt ............ fi Total measured thickness of Roxana ”— Silt and Peoria Loess .......................... 66 20.1 Continental deposits: Gravel, moderate-brown; pebbles, chiefly brown chert, as much as 5 cm across; clayey sand matrix ............................................ 1+ .3+ WEST 470' - 450'— Peoria Loess 400'_ Roxana Sill 350’— - Alluvium 300'— " 270' O 100 200 FEET O 30 60 METERS FIGURE 2. —— Section showing the stratigraphic setting of fos— sil peccaries in Roxana Silt at White’s gravel pit near Hickman, Ky. Vertical exaggeration x 4. Altitude is in feet above mean sea level. The silt at the peccary site is of rather uniform texture and is unstratified. Division of the silt into stratigraphic units is based mainly on color and on the presence or absence of calcareous zones. Three distinct but uneven and gradational color changes occur at the tops of units 3, 5, and 7, and one less 4 STRATIGRAPHY, MORPHOLOGY, AND PALEOECOLOGY, PECCARY HERD, KENTUCKY distinct occurs at the top of unit 6. The silt is cal- careous, except for unit 8 and for parts of units 1 through 5. At the outcrop the most noticeable break comes at the top of unit 5, about 11 feet (3.4 m) above the base. This break marks the highest non- calcareous silt and also separates moderate-brownish shades below from pale-grayish and pale-brownish shades above. The lower five units are designated here as the Roxana Silt; those above, as the Peoria Loess. Units 7 and 8 are typical Peoria, but unit 6 seems to be a gradational zone between the Peoria and Roxana. The uppermost part of the Peoria, as represented by unit 8, is everywhere leached of its carbonate content. On the basis of a study of fossil Mollusca from a locality about 1,000 feet (305 m) to the north (fig. 1), Ruth G. Browne (written commun., 1967) placed the contact between “prob- able Farmdale” (Roxana as used here) and “typical Peoria” at about 15 feet (4.6 m) above the base of the silt section, which is just above the highest occur— rence of the large Gastropoda Allogona profunda. and Anguispira kochi. This contact corresponds with the highest occurrence of large gastropods, probably the same species, in unit 6 at the peccary site. The Roxana Silt is mostly of eolian origin, but the lower 1—4 feet (0.3—1.2 m) is at least partly alluvial. Scattered quartz and chert grains and pebbles in this lower part become more abundant toward the con- tact with the underlying gravel. A few crudely im- bricate layers of whole and broken snail shells in this lower part indicate stream action. The lower 1—4 feet (0.3—1.2 m) of the Roxana represents a transition between alluvial and eolian deposition, which was accompanied by soil formation. Such a transitional zone, though common between gravel and loess in this region, is generally better developed and is char— acterized by red color and very abundant sand and pebbles. At the peccary site the upper limit of the transition zone seems to be an indistinct, uneven sur- face beneath which the silt is more compact than the silt above. Liesegang-like bands of limonitic coloration occur in the upper part of the transition zone; in some places these bands extend upward several feet into the overlying silt, and in others they extend down- ward nearly to the base of the Roxana. Most of the bands concentrically enclose silt of normal color; these centers may contain fossils. The fossil pec- caries appear to lie on the upper surface of the tran- sition zone and are enclosed by Liesegang-like bands. The banding is thought to have been produced by ground water carrying organic by-products of reduc- tion away from the peccaries and other organisms. Occurrence of similar color bands was noted also at a gravel pit about 1 mile (1.5 m) southwest of Dr. White’s pit (fig. 1). The association of the bones with color bands suggests that the areas in and around, and possibly between, these pits may be fa- vorable sites for finding more vertebrate fossils. The name Roxana Silt is used here for the lower formation in preference to Farmdale Loess as pre- viously used by Leighton and Willman (1949, 1950) in the Hickman area. A radiocarbon age of >34,000 years B.P. for the unit suggests that the unit is the Roxana Silt (Frye and Willman, 1960) and that it is correlative to part of the Altonian Substage of the Wisconsinan Stage, as used nearby in Illinois by Frye and Willman (1960). They restrict the term Farmdale to the organic-rich radiocarbon-dated silt and peat at and near the type locality in Illinois, and they define the Roxana Silt as the loessal and alluvial silt lying beneath the Farmdale (fig. 3). Leighton (1960) redefined the original Farmdale and pro- posed “a new name, Farm Creek, for the deposits of the original Farmdale type section in order to retain the name Farmdale for the glacial substage repre- sented by loess deposition” (Frye and others, 1968, p. 15—16). Farmdale Silt as used by Frye and Will- man (1960) is not present at the peccary locality. Loveland Silt of Illinoian age was not found in the Hickman area, but Leighton and Willman (1949, p. 52, 55; 1950, p. 614) reported Loveland Loess be— neath Farmdale Loess (Roxana Silt as used here) in the Mississippi River bluffs at localities that are 15 miles ( 24 km) to the north and 15 miles to the south- west of Hickman. The Sangamon soil zone in the upland areas of the northern Mississippi embayment commonly occurs beneath either Peoria Loess or Roxana Silt, but it occurs only beneath Roxana where both the Peoria and the Roxana are present (fig. 3) . It is developed only on Loveland Silt. Most of the zone is 3—6 feet (0.9—1.8 m) thick. It has not been found in the-Hick— man area. SILT PETROLOGY Mineralogic and size anayses were made of the seemingly uniform silt of the Roxana and the lower part of the Peoria to investigate the validity of the field-determined contact between the two formations and to search for fossil soils developed at this contact as well as within the Roxana, especially at the pec- cary horizon. Fabric studies were made of oriented silt samples to test the generally accepted eolian ori- gin of the silt. Paleowind directions indicated by the fabric study fit well the eastward thinning of the silt in the Hickman area and provide useful data on the paleoecology of the peccaries. SILT PETROLOGY ILLINOIS CLASSIFICATION RAgl'gé/g‘g‘ém FRYE, GLASS, AND WILLMAN (l962) pECCARY LOCALITY, YEARS LEIGHTON AND FULTON COUNTY, 3_ p, WILLMAN (1950) KENTUCKY TIME ROCK UNIT 0 - RECENT - VALDERAN - MANKATO TWOCREEKAN ‘ CARY — — —' — — — — ‘ Z — TAZEWELL g WOODFORDIAN Peoria Loess Peoria Loess 0 DJ “ IOWAN ‘1 Farmdale Peat and Silt 25,000 — FARM DALIAN [IE _ 7?? Roxana Loess, .7 ? _ zone IV Roxana Loess, , _‘ E zone 111 Roxana SM 2 E e 7; Q Roxana Loess, 5 8 zone II _ O (I) ‘ _7 7 7 7 9 E 7“ ‘r ' - 7 3 — FARMDALE LOESS ALTONIAN 50,000— ‘ ? 7 7 /' . ‘ / ’ Roxana Silt Roxana ‘\ \ zone I—b _7_ _ \ *— , Peocary “Grimm ‘7‘ 2 Silt '1 ‘ '4 I —? 1 1 Roxana Silt — zone 1—2. 70,000 . SANGAMONIAN (Sangamon Soul) FIGURE 3.— Correlation of silt units at the peccary locality with the classification of the Wisconsinan in Illinois (modified from Frye and Willman, 1963). *Approximate position of peccary horizon plotted ac- cording to probable stratigraphic position rather than radiocarbon age of >34,000 years B.P. 479-128 0 — 73 — 2 STRATIGRAPHY, MORPHOLOGY, AND PALEOECOLOGY, PECCARY HERD, KENTUCKY VISIBLE CARBON- ATE SAMPLE BULK MINERALOGY CLAY MINERALOGY NUMBER 25 5:0 715 25 5L0 715 1 I \\\\ \- ;\\ ¢—MEAN 18 ///////// ///////// /// , ////////////// /, - ' ' -//// //Mlxed-Iayer// ///// , ’I// H // ///// Clay///// ////////,///// ////////////// ///////// 17 Quartz PIagIow 16 ems PEORIA LOESS §§\, N Dolomite \ arbonat INTERVAL D 1“ :feld'sbari 13 /////// ////// 12 11 ////////////// ////////////// ////////////// 10 INTERVAL C MontmoriIIonite 7 \ \ \ Noncarbonate //////// //////// RQXANA SILT m Peccary zone // INTERVAL B ' . . / K 6.0. The magnitude of o' of the clay-mineral con- tents increases in the order kaolinite < chlorite < illite < montmorillonite w mixed-layer clay miner— als. Interval D (Peoria Loess) averages more ka- olinite, illite, and mixed-layer clay and less chlorite and montmorillonite than intervals B and C (Rox- ana Silt, above the transition zone). 7O 50— _ # o l + I '— z u U m U n. 30 — — 20 — _ 10 — ++ + + + _ Kaolinite Chlorite Illite Montmore Mixed-layer illonite clay V Interval D (lower part of Peoria Loess) O Intervals B and C (Roxana Silt _ above transition zone) ‘— X A Interval A (transition zone) a I Average of 18 samples FIGURE 5. — Mean (2) and standard deviation (0) of clay-mineral content of silt from peccary locality. Bulk mineralogic determination by X-ray diffrac- tion shows that interval D (Peoria Loess) has a mean carbonate content of 24 percent (fig. 4). Inter- val C (upper part, Roxana Silt) contains no visible 8 STRATIGRAPHY, MORPHOLOGY, AND PALEOECOLOGY, PECCARY HERD, KENTUCKY carbonate minerals as seen in thin sections, but it contains as much as 9 percent as determined by X-ray diffraction. Sample 11 shows an increase in total clay and a decrease in feldspar and dolomite over sample 12. Sample 7 shows a near absence of feldspars and carbonates and a high clay content. This clay contains more than 75 percent montmoril- lonite, the highest proportion of the mineral in any of the 18 samples. The average total carbonate con- tent of interval D is 24 percent, whereas the average of intervals B and C is 7 percent. Thin-section exam- ination shows that the carbonate grains in interval B have two habits: some grains are subangular to subrounded and appear to be detrital; the others have intricate grain boundaries that interlock with grains of quartz, feldspar, and clay clasts, and they appear to be recrystallized or authigenic. Dolomite is the dominant carbonate mineral in interval D (Pe- oria) and a minor constituent in intervals A, B, and C (Roxana). Calcite is rare or absent in interval D (Peoria) but is present in amounts greater than those of dolomite in intervals A, B, and C (Roxana), except in sample 5. Frye, Glass, and Willman (1962, p. 13) stated that calcite is more common than dolo- mite in the Roxana south of the Missouri River, whereas dolomite is the dominant carbonate mineral in the Peoria in Illinois. These differences are ex- plained by the fact that the older western tills, which were sources of detritus for the Roxana, are high in calcite, whereas the younger tills to the east in the Lake Michigan lobe, which were sources of detritus for the Peoria, are high in dolomite. Size analyses of the silt were made by means of a Coulter Counter (electronic zone-sensing particle- size analyzer). The 25M fraction was analyzed at in- tervals of 05¢. Phi-means range from 4.8 to 5.3 units (36p—25lu.) and are plotted in figure 4. These data show little variability though the ¢-mean (coarsen- ing) in samples stratigraphically above sample 11 (fig. 4) in general slightly decreases. Sample 11 probably represents the contact zone between the Roxana Silt and the Peoria Loess. Phi-skewness values are also relatively constant throughout the sampled interval. Visual inspection of the plot of ¢-means (M4,) in figure 4 shows M4, to be <5.0 for six samples from the Peoria and shows M1, to be >5.0 for eight samples and <5.0 for one sample from the Roxana (excluding samples where Md,:5.0). The value M¢:5.0 serves as a discrimi- nant between the coarser grained silt of the Peoria above and the finer grained silt of the Roxana below at this locality. Analysis of the ¢-means by the Fisher exact probability test (Siegel, 1956) at the >99.8-percent confidence level results in the rejec- tion of the hypothesis that ¢-means from Peoria and Roxana belong to a single population. INTERPRETATION The mineralogic and size data may be interpreted two ways with respect to stratigraphy and possible fossil soils. The preferred interpretation is the desig— nation of samples 1 and 2 (interval A) as pre—Wis- consinan paleosol and mixed loess and bedrock mate- rials formed at the onset of loess deposition on the pre-Wisconsinan land surface; samples 3—11 (inter- vals B and C) , as Roxana Silt (for purposes of map- ping and general stratigraphic discussion samples 1 and 2 are included in the Roxana) ; and samples 12— 18 (interval D), as the lower part of the Peoria Loess. If samples 3—11 represent Roxana, then samples 7 and 8 are interpreted as soil formed on the Roxana I—b of Frye and Willman (1963) and samples 9—11 are interpreted as part or all of Roxana zones II, III, and IV of Frye and Willman (1963) (fig. 3). Radio- carbon dating of Mollusca associated with peccary bones indicate an Altonian age for the fossil horizon, which is placed in Roxana zone I—b (fig. 3). The Farmdalian Substage (Frye and Willman, 1960) may be represented by a period of leaching that modified the Roxana to an apparent depth of 5 feet (1.5 m) (visible-carbonate column, fig. 4). Such a depth of modification is supported by the leaching that has accompanied Holocene scil formation on the Peoria to depths as great as 12 feet (3.7 m). Simi- lar leaching has been proposed by Frye, Glass, and Willman (1962) to partially explain differences in carbonate mineral assemblages in the Roxana and Peoria. The second interpretation is the definition of three stratigraphic units that have intervening fossil soils. Samples 1 and 2 are again designated as pre—Wiscon- sinan paleosol and mixed loess and bedrock materi- als. The Roxana (Frye and Willman, 1960) may be represented by samples 3-8, whose paleosol is repre- sented by samples 7 and 8. An equivalent of the Farmdale Silt (Frye and Willman, 1960) may be represented by samples 9—11, whose paleosol is rep- resented by sample 11. The lower part of the Peoria Loess is then represented by samples 12—18, interval D. This scheme allots a rock-stratigraphic unit to each time-stratigraphic subdivision of the Wiscon- sinan Stage (Frye and others, 1962). Frye and Will- man’s (1960) restriction of their Farmdale to the area near the type locality, however, probably rules out this interpretation. SILT FABRIC AND PALEOWIND DIRECTIONS The silt fabric was studied to determine the paleo- wind directions for the Roxana Silt, the peccary SILT PETROLOGY 9 PERCENT Ul PERCENT m 7,=121.4° A 180° 180° PERCENT 7221185“ 180° FIGURE 6. — Silt—fabric rose diagrams. Data for the Roxana Silt and Peoria Loess were shown to be significantly bimodal at >0.99 significance level by the X2 text. (See footnote on page 10.) 7, vector mean direction; 0, dispersion factor; a, contribution of dominant mode. A, Roxana Silt, samples 3—11, 870 observations. Bimodal model values: a:61.3 percent, 0121.26, 0221.56. B, Roxana Silt and Peoria Loess, sam— ples 3—18, 1,547 observations. Bimodal model values: 11:61.7 percent, 0121.5, 0221.7. C, Peoria Loess (basal 7 ft), samples 12—18, 677 observations. Bimodal model values: a:62.2 percent, 0121.57, 0221.91. zone, and the lower part of the Peoria Loess. The azimuths of the apparent longest physical (not opti- cal) axes of about 100 elongated quartz grains were measured in each of 16 horizontally oriented thin sections from samples 3—18. Quartz grains were chosen for two reasons. First, silt-sized quartz is the dominant mineral in the sections, and second, quartz lacks cleavage that might give rise to orientations based chiefly on cleavage as would calcite, dolomite, and feldspar. The effect of imbrication on the appar— ent grain shape was tested in one vertically oriented thin section from sample 13. The mean imbrication in this section is 8° E. This value is within the range of imbrication data reported by Matalucci, Shelton, and Abdel-Hady (1969) for loess at Vicksburg, Miss, and indicates that imbrication has only a slight effect on the apparent grain shape. The quartz grains are assumed to be oriented parallel to paleo- wind-current directions (Potter and Pettijohn, 1963, p. 54). The resultant paleowind directions are com- pared with the dominant presenbday wind direc- tions recorded by the U.S. Weather Bureau for Barkley-Field, Paducah, Ky., about 40 miles (65 km) northeast of Hickman. The azimuths of quartz grains measured for sam- ples from the Roxana and Peoria, separately and combined, are plotted in 100 classes and are ex- pressed as percentages of total observations in the eastern semicircles of silt-fabric rose diagrams in figure 6. A similar diagram for the peccary zone is presented in figure 8. Diagrams constructed for each of the 16 individual samples are very similar to those ShOWn in figures 6 and 8. In figure 7, the present-day wind data are plotted in sixteen 221/2O classes as the stippled area on the full-circle wind rose and are re- plotted as the black area on the eastern semicircle to conform to the convention used in the plots in figures 6 and 8. * 180“ FIGURE 7.—Full-circle wind rose (stippled area) based on about 40,000 azimuthal readings of directions from which wind blew (calm readings not included) at Barkley Field, Paducah, Ky., January 1950—December 1954, taken by US. Weather Bureau; semicircle wind rose (black area) of strike directions of same wind data. Vector mean direction (9) and confidence interval (dashed arc span) at >0.95 significance level. I 10 South Forelimb Probable location of skull broken before block 8 was removed Hindlimbs articulate with pelvis in block A 400’ FIGURE 8.—— Silt-fabric rose diagram for samples 5 and 6 combined to give 200 measurements from pec— cary zone. (See footnote 2.) y, vector mean; 0, dispersion factor; or, contribution of dominant mode. Bimodal model values: a=66.4 percent, 01=—0.302, 622—0066. Diagram is oriented with respect to plan view. 399’ FIGURE 9. ~——P1an view (upper) of five peccary skele- tons, labeled 1—5 (grid is in 1-ft squares), and sec- tion (lower) showing relative position of blocks B, A, J, and K. Altitude is in feet above mean sea level. 398/ The silt-fabric rose diagrams in figures 6 and 8 and the semicircle wind rose in figure 7 are obviously bimodal and cannot be treated by statistics assum- ing a unimodal model (T. A. Jones, written commun., 1971). Various bimodal statistical parameters are shown for the silt-fabric rose diagrams in figures 6 and 8.2 Tests by use of the bimodal model are not available for comparison of the statistical fit of the vector means (7), or implied paleowind directions, of the silt-fabric rose diagrams with the present-day wind directions. The silt-fabric rose diagrams shown in figures 6 and 8 are strikingly similar to the semicircle wind rose for present-day winds shown in figure 7. Thus, if the long axes of quartz grains record wind-current STRATIGRAPHY, MORPHOLOGY, AND PALEOECOLOGY, PECCARY HERD, KENTUCKY Mandible, skull, and cervical vertebrae Lumbar vertebrae, sacrum. and pelvis l Ulna-radius and pelvis fragment possibly from peccary 3 partial pelvis possibly from peccary3 6IO 90 1 Horizontal line BL OCK B BLOCK A directions, it is apparent that the wind directions during the time represented by the peccary zone were similar to the dominant southwestern and the northwestern winter-storm winds of the present time. 2The bimodal model is based on a. circular normal distribution (Stephens, 1962, 1964; Jones, 1967, 1968). In order to treat the semicircularly distrib- uted silt-fabric data by use of a bimodal model, the azimuthal angles must be doubled. according to James and Jones (1969. p. 129). The distribution of those data. with doubled angles is also bimodal. which indicates that the bimodal model is needed to calculate the statistical parameters for these populations. The parameters—vector means (71 and 72), dispersion factors (91 and 92), and percentage contribution (a) of the dominant mode—are calculated for this new distribution on the basis of the bimodal model of James and Jones (1969) and T. A. Jones (written commun., 1971). The data shown in figures 6 and 8 are significantly bimodal by the x2 test at >0.99 significance and have means, dispersion factors, and percentage contribu- tions consistent with the data and assumption of the statistics. FOSSILS 11 I Thoracic vertebrae, ribs, and \ fragments of humerus and scapula 2 Distal end of femur 240 270 300 L BLOCK J Smith (1942), in his study of loess (silt) in Illi- nois, found that silt is thickest along major rivers and thins rapidly away from them, as it does away from the Mississippi River in the Hickman area. He found that. the rate of thinning is a linear function of the logarithm of the distance from the source of silt and that the depositional pattern of silt is also a function of variations in the major wind directions with respect to flood—plain configuration (Smith, 1942, p. 162). Smith’s study and a more comprehen— sive one of the Whole western Kentucky region by J. D. Sims (written commun., 1970) suggest that the two major wind directions, southwesterly and north- westerly, caused the silt to pile up thickest along the bluffs near the Mississippi River. —1 399/ BLOCK K 398’ FOSSILS Peccaries are the only vertebrate fossils found at the Hickman site. The skeletons are the remains of a herd or family group of Platygonus compressus Le Conte 1848. Other such occurrences of this species have been reported by Klippart (1875), Leidy (1889), Williston (1894), Wagner (1903), Clarke (1916), Hoare, Coash, Innis, and Hole (1964) , Lewis (1970), and Elizabeth Dalvé (written commun., 1969). Of these occurrences, only one appears to have been in loess: that reported from Goodland, Kans., by Williston. Although Williston stated that his peccary group was found in a loose sandy marl, Hay (1924, p. 158) was certain that the bones were buried in loess. Hay’s opinion is corroborated by the 12 STRATIGRAPHY, MORPHOLOGY, AND PALEOECOLOGY, PECCARY HERD, KENTUCKY map of Pleistocene eolian deposits of the United States (Natl. Research Council, 1952), which shows Goodland, Kans., in an area where 67—100 percent of the land is covered by loess 8—16 feet (2.4—4.9 m) thick. The skeletons reported by Hoare, Coash, Innis, and Hole (1964) may have been buried in dune sand. Lewis (1970) reported the skeletons of at least five individuals found in eolian sand and silt in Denver. Besides these few discoveries of skeletons, there have, of course, been scattered surface finds of dis- articulated bones and a remarkably large number of occurrences of peccary bones in caves. Hay in 1923 and 1924 summarized discoveries up to those dates, and subsequent reports have been made by many other authors. Associated with the peccaries are fossil gastropods. At and within a foot or so of the stratigraphic level of the peccaries are widely scattered bits of carbon- ized plant matter, probably wood. The plant matter is fairly abundant in pockets and layers. In the cal- careous silt, carbonate root casts are abundant. Near the peccaries and in a zone 3-6 feet (0.9—1.8 m) above the base of the Roxana are numerous white powdery, but well-preserved, calcareous fossil hack- berry seeds. PLATYGONUS COMPRESSUS LE CONTE 1848 The five fossil peccaries where found appeared to lie on a surface that marks the end of mixed alluvial and eolian silt deposition; they were covered by wholly eolian silt. The surface forms an irregular plane dipping from 0° to 10° NW. The five skeletons, all headed toward'the east roughly parallel to one of the two vector mean directions shown in figure 8, lay in a line trending approximately S. 80° E. within an area about 3 feet (1 m) wide and 11 feet (3 m) long (fig. 9). The snout of skeleton I touched the rear of skeleton 2; the snout of 2 was about 1 foot (0.3 m) southeast of the sacrum of 3; the snout of 3 touched the rear of 4; the heads of 4 and 5 (de- stroyed before recovery) were probably within 1 foot (0.3 m) of each other and pointed in the same direction. The westernmost pair of skeletons (1 and 2) are best preserved (figs. 10, 11) ; they lay at a lower altitude and were apparently covered more deeply and quickly than the three skeletons clustered to the east of them. The disposition of the skeletons indicates that the peccaries were suddenly overcome and were very quickly buried by eolian dust. Skele- tons 1 and 2 were undisturbed after death. This is undoubtedly true also of skeletons 3 and 4, but their higher altitude may have resulted in damage and movement of some bones during bulldozer passes associated with the discovery. Also two bones which we associate with skeleton 3 were found 1 and 2 feet, respectively, north of the skeleton. One forelimb and one hindlimb of skeleton 5 were lying in a proper position relative to each other so that we conclude that they were not moved after the animal died. The other hindleg associated with this skeleton was found, still articulated, 1 foot (0.3 m) west of its mate. This is slightly out of position for a natural pose, and it might be assumed that a predator had dragged the leg to this position. This is deemed un- likely, however, because the bones have not been broken or even disarticulated. G. E. Lewis (oral commun., 1970) suggested that the displaced bones were shifted by movement of windblown silt. None of the skeletons showed evidence of disturbance by predators, or even of gnawing by rodents. Burial must have been very rapid to have avoided this. POSITIONS OF THE SKELETONS Although several herds or family groups of Platy- gonus compressus have been discovered, the postures and relative positions of the skeletons have not been described in detail, and none of the groups has been preserved with the skeletons in their original poses. A group of Platygonuspearcei Gazin from the Hag- erman Lake Beds of Idaho (Gazin, 1938), consisting of an adult and two young individuals, has been pre- served in the position in which found (USNM 13800). Gazin figured these specimens but did not discuss their poses. Accordingly, it seems worthwhile to describe, as well as figure, the positions of the Hickman peccary group. The skeletons are preserved in the US. National Museum (USNM 26098—26102) in the positions in which they were found. SKELETON 1 Skeleton 1 (figs. 9, 10) is lying on its right side. The right side (underside) of the skull seems to have been crushed, inasmuch as the right temporal con- dyle has been rotated downward and outward and turned 180°. The neck is flexed to the right (down- ward), probably by the weight of the overlying ma- terial. The atlas has been disarticulated from the occipital condyles and moved to the right (downward) about 70 mm. With only slight gaps, the vertebrae are articulated from the atlas to the third caudal; two other caudal vertebrae, separate but articulated with each other, lie beside the third caudal. The an- terior part of the back is lying on its side, in the same plane as the head. Between the 10th thoracic vertebra (most posterior one with a long neural spine) and the 11th, rotation has occurred; posterior to the 11th thoracic vertebra the neural spines are vertical. The pelvis, which is in normal articulation with the sacrum, is positioned with its dorsal side up. The posterior half of the vertebral column is FOSSILS 13 15 O 15cm FIGURE 10.——Peccary Platygonus compressus, skeleton 1, in natural position. probably in the position the animal assumed just before death. The hyoid bone is approximately in place at the level of the posterior border of the man- dible. Because of the decision to leave the skeleton in death position, the hyoid could not be exhumed, but the posterior 35 mm of the right thyroid cornu is exposed. It is 3.5 mm in diameter. The hyoid has been rotated so that its right side is about 15 mm lower than its left side. The left femur is in almost natural articulation with the pelvis. The tibia is slightly twisted out of articulation at the knee. The rest of the left hindlimb is in normal articulation. The right femur has moved outward and downward from a normal crouching position so that the femur lies with its medial side up, rotated in the acetabulum beyond the position of normal articulation. The tibia is in normal articula- tion with the femur. The patella is present, only slightly out of position. The rest of the right hind- limb is buried beneath the lumbar region. The left front limb is in normal flexed position. The carpals are slightly disarticulated from the ulno- radius. The right scapula is rotated upward and for- ward so that its neck is at the level of the seventh cervical vertebra and 120 mm above its centrum. The head of the right humerus is in contact with the glenoid cavity of the scapula but is disarticulated and rotated backward so that it is parallel to the vertebral column. 0n the humerus, the line of the proximal epiphysis is visible; the distal epiphysis is closed. The radius and ulna are fused. Both epiphysis of the radius are closed. The olecranon epiphysis of the ulna is barely visible. The epiphysis of the metacarpus and of the first and second phalanges of the forelimb are fused. The main pelvic bones are fused, but the ilium is not fused to the tuber coxae, and the tuber ischii is not fused to the ischium. On the femur, the proximal epiphysis is not clearly visible; the distal epiphysis is still visible but is probably nearly closed. On the tibia, the proximal epiphysis is fused; the bone is cracked at the distal end so that it is impossible to be sure of the condition of the epiphysis. The metatar- sal epiphyses are fused. 14 STRATIGRAPHY, MORPHOLOGY, AND PALEOECEOLOGY, PECCARY HERD, KENTUCKY M3 and probably M3 are fully erupted. M2 and M3 are slightly worn. M1 is worn to concavity. P3 and P4 are worn so that the dentine shows. P2 is only slightly rounded by wear. There is heavy wear on the canines. According to the age criteria developed by Kirk- patrick and Sowls (1962) on the basis of tooth eruption in the collared peccary (Tayassu tajacu), skeleton 1 represents an individual more than 74 weeks old. SKELETON 2 Skeleton 2 (fig. 11), like skeleton 1, represents an individual that seems to have died lying in a natural position on its ventral side, with forelimbs flexed under it and probably with chin resting on the ground. In contrast to skeleton 1, the skull of skele- ton 2 rests on its left side. The cervical vertebrae have been displaced ventrally (horizontally) about 80 mm. All vertebrae are in articulation from the atlas to about the fourth thoracic; there the verte- brae are disarticulated, with the anterior surface of the centrum of the fifth thoracic vertebra at an Thoracic vertebrae, ribs, and fragments of humerus and scapula 3?» Distal end of femur 15 O | | angle of about 100° to the posterior surface of the fourth thoracic (fig. 11). The attitude of the poste- rior thoracic vertebrae cannot be seen, but the ribs protruding from the matrix are in sequence, imply- ing that the corresponding vertebrae are in or near articulation. Six lumbar vertebrae are in articula- tion. Most of the sacrum has been lost, but parts of three sacral vertebrae, which had been linked only by cartilage, were collected With the skeleton. The scapulae are approximately in place. The fore- limbs are not visible. The hindlimbs are in place. The pelvis is separated into right and left halves, and the two parts are out of orientation with each other (fig. 11). From the orientation of the hindlimbs, it ap- pears that they were spraddled much like those of skeleton 1. The right femur is rotated backward and lies on its anterior side in a position that suggests that it was in or near articulation with the acetabu- lum. The right tibia has been dislocated from the femur and moved forward 45 mm from the distal end of the femur (fig. 11). The tibia, tarsus, and metatarsus are articulated; the phalanges, if pres- . A}! ’11" .. . W: 150m 1 FIGURE 11.—Peccary Platygonus compressus, skeleton 2 and posterior part of skeleton 3 (lower left part of block), in natural position. FOSSILS 15 ent, are embedded beneath the anterior end of the lumbar series. The left hindlimb is articulated. Both hindlimbs are flexed. The epiphyseal sutures are closed on the hind- limbs, except for the distal ends of the tibiae, where they are open. Epiphyses are open at the ends of the centra of the thoracic vertebrae; they probably are closed on the lumbars. The upper dentition is not visible. M3 is erupted and slightly worn. M2 has about the same wear as M3. M1 is worn, and concavity was just beginning. P4 is slightly worn; P2 and P3, less so. The anterior side of the upper canine is considerably worn. Skeleton 2 represents an individual younger than that of skeleton 1 but is judged to be still older than 74 weeks on the basis of tooth eruption. SKELETON 3 Skeleton 3, though articulated at burial, is less nearly complete than skeletons 1 and 2. It was col- lected in several pieces: posterior part from the left end of block J (fig. 11) , anterior part from the right end of block A (fig. 12), and isolated bones from Mandible, skull, and cervical vertebrae \ Lumbar vertebrae, sacrum, and peIVIs 15 north of blocks A and B (fig. 9). A sacrum com- prised of four vertebrae, lies 150 mm north of the skull of skeleton 2 (fig. 11). The right femur is in proper position for articulation (fig. 11) though the pelvis is not in place—iliac blades collected sepa- rately probably represent the pelvis of this individ- ual. The distal epiphysis of this femur is open. A femur, tibia, ulno-radius, and two pelvic fragments (fig. 9) that are isolated from the other bones of skeleton 3 may belong to that skeleton. Two thoracic vertebrae, with open epiphyses on their centra, the distal end of another femur (fig. 11), with epiphysis closed, and several ribs are also present. At the west end of block A, 38 cm left of the east end of block J, are an articulated scapula, humerus, and fused radius and ulna (fig. 12), all with epiph- yses closed and in extremely flexed position. East of these are articulated cervical vertebrae 2 through 7, with 7 in contact with the head of the humerus (fig. 12). The anterior spine of the axis vertebra lies only about 10 mm from the foramen magnum of the skull. The atlas was displaced but lay near the other ‘ Scapula, humer _, ~ radius, and ulna .. H . 15 cm ——0 FIGURE 12. —Peccary Platygonus compressus, anterior part of skeleton 3 (right two—thirds of block) and posterior part of skeleton 4 (left part of block), in natural position. 16 STRATIGRAPHY, MORPHOLOGY, AND PALEOECOLOGY, PECCARY HERD, KENTUCKY cervical vertebrae; it was removed during excava- tion. Most of the cranium of this individual was de— stroyed, probably by the first pass of the bulldozer blade at the time of discovery. However, a partial maxilla was recovered, bearing P3, M1 and M2. P3 is almost unworn; P4 is missing—no alveolus is pres- ent and the position of P“ is represented only by a gap of 7.3 mm between P3 and M1. M1 is worn to concavity; in M2, the anterior loph is appreciably worn, and the posteriorLloph is fresh. The mandible is present, nearly in arti ular position with the skull; its anterior end is broken off 30 mm anterior to the posterior end of the symphysis. P2 is missing, and the part of the mandible bearing it is chipped away. The right P3 is missing; its alveolus indicates a small tooth (table 1). The right P4 is present, is fully erupted, and is only very slightly worn. The right M1 is slightly worn. M2 and M3 of both sides are present. M2 is in beginning wear; M3 is unworn, not quite erupted. Skeleton 3, on the basis of tooth eruption, is be- lieved to represent an individual of about the same age as that of skeleton 2. SKELETON 4 Skeleton 4, like skeleton 3, is divided between two blocks (figs. 9, 12, 13). The pelvis of skeleton 4 (fig. 12) was in contact with the skull of skeleton 3. The basal part of the sacrum is present and in place, but the tops of the sacral vertebrae have been destroyed. Four associated and articulated lumbar vertebrae are in place (fig. 12). The central epiphyses of these vertebrae are open, as are those on five thoracic ver- tebrae, probably from this individual, that were col- lected loose at the east end of block A. At the south end of block B (almost in contact with the east end TABLE 1. —Measurements, in millimeters, of skulls, mandibles, and teeth of Platygonus compressus from near Hickman, Ky., and Denver, Colo. [Specimen numbers: USNM, U.S. Vational Museum; D, U.S. Geological Survey, Paleontology and Stratigraphy Branch, Denver fossil vertebrate locality. Leaders ( ...... ). measurement not possible] Kentucky Colorado Specimen No ........... [M 26098 USNM 26099 USNM 26100 USNM 26101 D714A D7140 D714H D714E D7146 Skeleton .................... 1 2 3 4 Young adult Old adult Young adult Juvenile Newborn male female male or fetal Skull Greatest length .............. 330 1310 ............ 330 298 325 ............ Condylobasal length ..... 299 1285 ............ 302 275 303 ............ Length from anterior margin of foramen magnum to anterior extremity of premaxillaria ........................... 290 ‘274 ...... 291 265 290 ............ Anterior border of orbit to extreme anterior end of premaxillae ................................. 199 1155 ...... 215 191 212 ............ Depth of zygoma. from end of postorbital process to end of preglenoid process ............... 64 172 ...... 76 68 73 ............ Depth of zygoma at middle below orbit ................................. 48 44 ............ 40 35 38 ............ Dorsoventral diameter of orbit. 227.8 42 36 35 33 ............ Length of superior precanine diastema ...................................... 26.7 .................. 23 21 21 ............ Length of superior postcanine diastema ...................................... 53 __________________ 48 40 55 ............ Length of temporal fosa from inion to postorbital process... 87.3 81 ............ 78 68 75 ............ Height of occiput from ventral border of occipital condyles... 90 1108 ............ 100 93 97 ............ Length of inferior postcanine diastems ...................................... 65.9 153 122 354.2 57 ............ 53 9.2 Mandible Length condyle to incisors ......... 231 1200 ____________ 231 .................. Delitlllvl below anterior margin 40 0 i ................................. 44.3 42.5 40 342 43 .................. Depth, at coronoid proces 1105 197\ 96 4 1.3104 108 ............ 90 27.3 Teeth Length ILM3 ............... 1179 \ .. . ____________ 171 161 177 , ............ Eengtg 1521:1113 ............ ‘ ...... 1 350 48.5 52 49 ............ engt .. 178 .................. 77 ........................ Upper canine l ...... 17.4 X 10.0 13.5 X 10.3 17.2 X 9.6 ............ P“. ...... 10.0 x 11.5 10.3 x 11.7 10.0 x 10.7 ...... 311 X 13.2 10.0 X 13.1 11.9 X 13.1 9.5 X 11.4 313.2 X 14.9 314.3 X 12.2 13.5 X 13.2 14.3 X 14.5 13.2 X 12.3 316.4 X 14.6 318.1 X 15.5 16.4 X 15.2 16.1 X 16.2 15.9 X 13.6 ...... \ 31419.1 X 15.2 19.5 X 15.5 21.6 X 17.2 20.5 X 13.8 .................. 14.0 X 10.0 19.9 X 6.4 ...... 310.2 X 7.5 9.7 X 7.3 ........... 110.5 X 8 11 7 X 10 310.7 X 8.9 10.4 X 8.6 ............ 111.3 X 9.6 ...... 311.4 X 9.7 11.0 X 9.9 ............ 113.5 X 9.5 “14.1 X 10.5 13.1 X 11.1 14.2 X 11. ............ 713.7 X 10.3 116.6 X 12.6 “16.6 X 13.0, 315.9 X 12.8 15.9 X 12.5 ‘ ............ 716.4 X 13.2 . 123.0 X 12.6 “22.7 x 13.9, 323.3 x 14.1 22.0 x 15.5 ........................ 1521.8 X (5) Length P2—Mri ...... 82 185 383.7 83.4 ........................ Length Mr—Mm 53 55 352.4 52.7 .................. 1Estimate. 5Buried in matrix: unable to measure. ”Dorsoventrally crushed. 6Right. 3Bone not in place, but assumed to belong to this skeleton. 7Left. ‘Anterior loph. FOSSILS 17 of block A) is a badly broken ulna articulated to a complete lower forelimb (fig. 13) , in flexed position. A skull was located immediately north of this (fig. 13) but was broken before the block was removed. Two fragments found at the site of this skull contained, respectively, P4 and M1, and M2 and M3. P4 is only very slightly worn. The anterior loph of M1 is worn to the point of dentine exposure; the posterior loph is only Slightly worn. The anterior loph of M2 is slightly worn; the posterior loph is unworn. M3 is unworn and was probably unerupted. Surrounding the site of the skull are two articulated lower hind- limbs (tibia, tarsus, metatarsus, and phalanges; fig. 13). Of these, the tibiae appear to have open epiph— yses at both ends; all other epiphyses are closed. The individual represented by skeleton 4, on the basis of tooth eruption, is interpreted as younger 15 0 15cm I FIGURE 13. — Peccary Platygonus compressus, parts of skele- ton 4 (top) and skeleton 5 (bottom), in natural position. than the individuals represented by skeletons 1, 2, and 3 and as not much older than 74 weeks by the criteria used with Tayassu tajacu, inasmuch as M3 is fully formed but unerupted. SKELETON 5 Skeleton 5 (figs. 9, 13) is represented by a scapula, part of an articulated hindlimb (tibia, tarsus, meta- tarsus, and phalanges), and part of an articulated forelimb (distal end of humerus articulated with fused radius and ulna) (fig. 13). The hindlimb is extended and the forelimb flexed—this animal died in the same position as the animal represented by skeleton 1 (figs. 9, 10). The distal epiphysis of the humerus appears to be open; all other epiphyses are closed. All that can be said on the basis of the limbs of skeleton 5 is that the individual was probably an adult. A right mandible found near block A (fig. 9) must belong with either skeleton 4 or skeleton 5. It has a complete cheek-tooth series: P2 and P3 with enamel slightly worn, P4 with dentine showing on the anterior lophid, M1, worn to concavity, M2 with dentine showing on both lophids, and M3 with den- tine just beginning to show on the first and second lophids. This would indicate an individual of about the same age as that of skeleton 1, and the teeth would not match the upper teeth assigned to skeleton 4. The mandible is, therefore, tentatively assigned to skeleton 5. COMPOSITION AND INFERRED HABITS OF THE HERD All five peccaries were adult. Absolute ages cannot be assigned with confidence, but the relative ages of the individuals are significant because of their asso- ciation. It is estimated that the herd consisted of two fully adult animals (skeletons 1 and 5), two slightly younger (skeletons 2 and 3), and a still younger ani- mal (skeleton 4), that was nearly full grown. On the basis of size of canines and heaviness of zygo- matic arch, skeleton 1 is believed almost certainly to have been a male. On the same basis, skeleton 2 is believed probably to have been a female. The sex of the other skeletons cannot be judged. As found, the skeletons were probably lined up in the order in which they were walking when overwhelmed: the two oldest animals leading and bringing up the rear, respectively; the youngest animal following the leader and the two slightly younger behind it. Despite morphological differences that certainly indicate contrasts in habit between Platygonus com- pressus and living peccaries, both forms share the characteristic of gregariousness as indicated by dis- coveries of herds of Platygonus, cited before. Both the collared peccary (Tayassu tajacu) and the white- 18 STRATIGRAPHY, MORPHOLOGY, AND PALEOECOLOGY, PECCARY HERD, KENTUCKY lipped peccary (Tayassu pecari) run in herds. Har- ems do not exist; a herd consists of both sexes and all ages, more or less in random combination. The collared peccary has a year-round breeding season (Sowls, 1961). There is no evidence of pairing of males and females for either a short or a long period. Sows appear to have several mates during a single estrous period, and there is no aggressive behavior between males in the herd (Sowls, 1966). In Vera- cruz, Mexico, the collared peccary usually travels in herds of 5—10 individuals; the white-lipped peccary, in bands of 12—25, with herds of 50 or more reported (Hall and Dalquest, 1963). Leopold (1959, p. 494) pointed out that peccary herds are larger when the population is denser. The collared peccary, in areas of sparse population, runs in herds of two or three up to a half-dozen. In areas of plenty they are seen in groups of 10—20, rarely more. Collared peccaries follow their mothers until at least 1 year old, so smaller herds may well be family groups. The composition of the Hickman herd of Platy- gonus compressus is consistent with the loose social organization observed for living peccaries. Skeleton 4 may represent a yearling still following its mother. Among several structural differences distinguish- ing Platygonus from living peccaries are longer legs, presence of sharp, high transverse cusps on the cheek teeth, and much larger nasal cavities which have more extensive turbinal bones. As Leidy (1853, p. 325—326) pointed out, the tooth pattern of Platy- gonus indicates a lower jaw movement that was “less rotary and more simply ginglymoid than exists in Dicotyles; for at the same stage of trituration of the teeth in the latter, the principal lobes would be worn down to a level with the bottom of the inter- lobular fissures, approaching more in this respect the Hog than the Tapir.” The strongly lophodont teeth of Platygonus indicate a diet different from that of Tayassu; G. E. Lewis (written commun., May 1970) has suggested that Platygonus is better equipped to chew harsher vegetation from a different environ- ment than Tayassu, which has a bunodont dentition. The geographic range of Platygonus embraced a more rigorous environment than does that of Ta- g/assu, but the range of both genera is latitudinally large enough to show that they share the ability to subsist on a wide variety of food. The collared peccary is distributed from the Rio de la Plata of Argentina, at about lat 350 S., to south- western United States at lat 35° N. This range in- cludes both desert (Sowls, 1961) and tropical forest (Chapman, 1936; although Chapman’s paper is titled “White-lipped Peccary,” it also records obser- vations of collared peccary herds on Barro Colorado Island, Canal Zone). Platygonus compressus is known to have ranged from California to New York and from Michigan south at least as far as Florida and Texas (Hibbard and Taylor, 1960, p. 184) ; Ray, Denny and Rubin (1970) reported a specimen that must have died “within a few miles of the wasting late Wisconsinan continental glacier [in Pennsyl- vania], and [must have] lived there in a periglacial environment.” Leidy also remarked (1853, p. 334—335) on the “enormous cellular dilatation” between the maxilla and the palatine bones, whose interior communicates anteriorly with the cavity of the nose. The malar bone also has a cellular interior. A set of skulls sagit- tally sectioned by John E. Guilday showed that Platygonus has a much larger nasal cavity and more extensive turbinal bones than Tayassu or Mylohyus; it also has false internal nares, leading into cul-de- sacs at the posterior end of the palate (Guilday and others, 1971, p. 304). The passage of air through the nasal passage was thus very tortuous, which may have been an advantage in a dust-laden atmosphere. Platygonus, unlike Tayassu, adapted to a cold en- vironment, although, as indicated by the southern limits of its range, it also could live in warmer cli- mates. Most occurrences of Platygonus are in areas that were probably characterized by harsh vegeta- tion, such as silica-rich grass, in late Pleistocene time. Tayassu, by contrast, lives in warm climates but can subsist on both tropical and desert vegeta- tion. The cold-climate range of Platygonus implies great differences from Tayassu in pelage and other adaptations to the environment; a comparable con- trast is that between the Pleistocene periglacial woolly mammoth and the living elephants of Africa and India. OTHER HERDS OF PLA TYGONUS COMPRESSUS We have found eight reports of herds of this spe- cies that died together. Two of these consisted of only two individuals each, but they are included be- cause the individuals were lying close together in the pose that is characteristic of larger herds. DENVER, COLORADO, HERD G. Edward Lewis (1970) reported a group of Platygonus compressus in eolian sand and silt of late Wisconsinan age (USGS fossil vertebrate loc. D714). Parts of five skulls were found: those of two adult males, an adult female, a juvenile, and either a newborn or a fetal animal. Numerous postcranial bones were found, but it was impossible to determine with which skulls these bones belonged. The proxim- ity of the bones to each other indicates that the in- dividuals constituted a herd. No complete articulated FOSSILS 19 skeleton was preserved. Lewis (written commun., 1970) believed that dissociation was accomplished by dune movement, perhaps over a period of years. For this reason, only measurements of individual bones are available, and most bones cannot be re- ferred with certainty to distinct skeletons. Previ- ously unpublished measurements of these bones, furnished by G. E. Lewis, are presented in tables 1 and 2. GOODLAND, KANSAS, HERD One of the best knOWn reports of a herd burial is that of Williston (1894) involving nine skeletons buried in what was almost certainly loess. The ani- mals were facing southwest, with the heads of the hinder ones resting on the posterior parts of those in front. The bones were “all or nearly all in the po- sition in which they had been at the animals’ death” (Williston, 1894, p. 23). As in all other excavations of Platygonus compressus herds, the skeletons were collected bone by bone rather than being left in death position. The skeleton of a female was mounted under Williston’s direction at the University of Kan- sas. A second skeleton was assembled and sent to the American Museum of Natural History, and a third was sent to the Carnegie Museum (Peterson, 1914, p. 114) and subsequently to the National Mu- seum of Canada. Presumably the other specimens remained at the University of Kansas. Williston (1894, p. 27) applied the new name Platygonus lep- torhz‘nus to the members of this herd. This species has been synonymized with P. compressus (Simpson, 1949, p. 22). Williston’s type of P. leptorhinrus bears the University of Kansas Museum number KU 458; the other skeletons of the herd are lumped together under number KU 3108. Williston stated that the animals were of different ages and sexes, but he does not give the number of animals in each category. From the descriptions and measurements in his paper, it is apparent that the herd included two old individuals, one adult female, two adult males, and two young individuals. Of the young, one had only the first molars erupted of the permanent tooth series, and the other still had its deciduous teeth. SANDUSKY COUNTY, OHIO, HERD Parts of at least four individuals were recovered in sand, possibly a dune deposit (Hoare and others, 1964). The bones were partially articulated. The skeleton of a mature female (BGSU 2353) was mounted for exhibition. At least one individual (BGSU 2354) was immature, with open epiphyses on the leg bones and vertebrae. The other peccary bones are numbered BGSU 2355. No mention was made of the positions in which the skeletons were found. The specimens are in the collections of the Department of Geology, Bowling Green (Ohio) State University. READING, OHIO, HERD We thank Mrs. Elizabeth Dalvé of the University of Cincinnati for information concerning one peccary skeleton (UC 40311) and part of a second (UC 40312) excavated by her in 1948 and for permission to measure the complete skeleton, which is mounted in the University of Cincinnati Museum. The skele- tons were found on the Albert Stegmann farm, Cooper Road, just northeast of Reading. They were about 6.5 feet (2 m) below the surface in what ap- pears to be an area of lateral outwash from kame deposits. The altitude of the kame terrace is approxi- mately 635 feet (195 m) above sea level, in contrast to the common 800-foot (245 m) altitude of the bed- rock hills of the region and the 580-foot (175 m) altitude of the terrace in Reading of latest Wiscon- sinan age. Mrs. Dalvé pointed out that more than 11/2 skeletons may have been present, but that excavation had to be discontinued owing to danger of collapse of the trench wall. COLUMBUS, OHIO, HERD Twelve skeletons were found in 1873 “embedded in calcareous clay, intermingled with a very gener- ous quantity of calcareous sand” (Klippart, 1875). Six of the smallest animals were found together at a depth of 8 feet (2.4 m) ; the six larger animals were 6 feet (1.8 m) from the first group and 4 feet (1.2 m) deeper. Klippart reported that “the strata immediately above the remains were confused, pre- senting evidence of disturbance so far as the deposi- tion of sand and gravel were concerned, whilst the overlying clay was, so far as the eye could discern, entirely intact or undisturbed.” This stratigraphic situation could result from the burial of the pecca- ries in the cave-in of a bluff under which they had taken shelter; the overlying undisturbed clay could be an alluvial deposit laid down on top of the land- slide material at a later time. The animals were lying side by side with their heads toward the southeast, the same orientation as the Hickman peccaries. The bones were entire except where broken from “mani- fest pressure,” and Klippart could find no evidence of the bones “having been gnawed, or ‘crunched.’ ” He reported that nearly all the “bases” and “heads” of the femora, humeri, ulnae and other bones became separated from the shafts while the bones were be- ing boiled in glue. This indicates the presence of a number of juveniles considerably less than a year old, which is to be expected in so large a herd. Half the specimens were apparently sent to O. C. Marsh at 20 STRATIGRAPHY, MORPHOLOGY, AND PALEOECOLOGY, PECCARY HERD, KENTUCKY Yale (Hay, 1923, p. 214). One skull and one man- dible (USNM 16557) are in the US. National Mu— seum. BELDING, MICHIGAN, HERD Wagner (1903) reported on five skeletons in the museum of the University of Michigan. He reported that they had been found in a peat bog, but C. W. Hibbard and R. E. Eshelman, in restudying the specimens, have determined that they came from up- land loess (Eshelman, written commun., Mar. 8, 1972). No record had been made of the positions of the skeletons when found. The group contains one adult whose skull sutures are obliterated. The skele- ton of this individual (V7325) was mounted at the University of Michigan and was figured by Case (1921, p. 295). Wagner identified this specimen as a female, but Hibbard and Eshelman identify it as a male. The other skeletons are incomplete, and Wag- ner did not mention the age or sex of the individuals. Hay (1923, p. 216) pointed out that the Belding herd was found close to a part of the Charlotte moraine system of Wisconsinan age. GAINESVILLE, NEW YORK, HERD Two skeletons were found in a sand and gravel bank (Clarke, 1916). One belonged to an older in- dividual; the other, to a younger, though grown, specimen which, according to Clarke, still had its temporary molar teeth. No mention was made of the position of the skeletons, which were disturbed be— fore being collected, inasmuch as many of the bones had been carried away with sand for mixing con- crete. The recovered bones were placed in the New York State Museum (N.Y.S.M. V41a and V41b). Hartnagel and Bishop (1922, p. 85) stated that the Gainesville skeletons were found in a drumlin less than 1 mile south of the Silver Springs—Rock Glen kame area, in a pocket of sand with clasts “up to one foot in diameter, those of three to four inches in diameter being most abundant.” This depositional environment is reminiscent of that reported by Ray, Denny, and Rubin (1970) for a skeleton of Platy- gonus compressus from Mosherville, Penn. PI'I‘TSFORD (P) , NEW YORK, HERD Leidy (1889) reported the occurrence of two adult individuals, one represented by the greater part of a skeleton and the other by a skull. They were found in a gravel bank. Leidy did not mention the orienta- tion or position of the skeleton. The specimens were deposited in the Academy of Natural Sciences in Philadelphia (skull, ANSP 11544; postcranial bones, ANSP 11545). SIZE DISTRIBUTION IN PLATYGONUS COMPRESSUS Herds such as the Hickman group, found buried together, are helpful as indicators of the size range that existed in groups of known contemporaneity and probable blood relationship. Although several such groups are known, as described above, available measurements are few. Measurements of the Hick- man herd are presented in tables 1 and 2 together with measurements of the Denver herd. Publication of measurements of other existing groups would be useful in determining whether population differences can be discerned, and possibly correlated with ecologic conditions, over the wide geographic range of this species. Table 3 reproduces measurements, all made from articulated skeletons of P. compressus, for comparison with measurements of the members of the Hickman herd. Other measurements are avail- able in the sources cited; those reproduced here were chosen because they may be indicators of body size. Only for the Goodland, Kans, herd (Williston, 1894) and the Denver, Colo., herd (tables 1 and 2) were measurements for more than one skeleton avail— able. The Goodland and Denver herds differed from the Hickman herd in that they contained juveniles. For the Goodland herd, Williston presented post- cranial measurements of only two members, an adult male and an adult female (1894, p. 27—28). The adult female'was mounted for exhibit at the University of Kansas. A composite skeleton, including an imma- ture skull, was traded to the Carnegie Museum, where it was mounted (CM 2806; Peterson, 1914) ; later it was transferred to the National Museum of Canada (NMC 17912). . The upper dentition of NMC 17912 consists of dP1'2-3, M1 just coming into wear, M2 erupted but un- worn, and M3 unerupted. In the lower dentition, P2, P3, and P4 are beginning to erupt (the milk teeth have been shed), M1 is slightly worn, M2 is erupted but unworn, and M3 is unerupted. The following measurements were made by Whitmore: dPl—MZ, 56.2 mm; Ml—M3, 44.1 mm; C-M3, 129 mm; upper diastema, 43.5 mm. Despite the composite nature of the National Museum of Canada skeleton, measure- ments are included in table 3 as indicative of the size range represented by the members of the Goodland herd. Another skeleton from the Goodland herd was for- warded in exchange to the American Museum of Natural History, New York (AMNH 10388). Simp- son (1949, p. 38-44) published measurements of the cheek teeth of this specimen which indicate that it was fully adult. The American Museum skeleton is probably a composite, although this is nowhere 21 :33me £5 3 95—3 3 vac—=53 «an 6an :m 00: w:om« dauEsmm—H ,,,,,, 5mm mdm 43:3 <3 «3:. .5512: :meoum Hm: «.mw Yum ENE: .>H Aafiaaaacfi m5: ‘‘‘‘‘ ms 6.me— .fluflh ...... «Na «N 3.....‘::.....:...:.::oma£:umu.~a 732: «o 5.352: 093.5:2? aw «5N— wm~ sew—~01.— ...... wdv 9% "Eu 3518.3 «9 523 . "15:. ........... mNN .flO—n—OOha HO hwavgflmfi 0mh0>m=dhyfl .45“ mm: 873:8 «a .5353“. 3.55:9; mtnmn mu: ....... wuw: «o Sam—:3“. waggauh run 5H .33: no .5353: homuwamonohoan< ...... 3:; 3: £3.54 nude—Eh FOSSILS bdm .................................... COEGhOU hon—flhggo HO LSWEdmmu UBMCO‘H 1mm .................................................. 53:233.: no 33:55 «NH . Est—9mm ya :33: .8133: 8 525833: «o 3:50 am .............. EB: «a .5va .5135: 3 53:858.» «0 §=00 mm« .295— umwuawnmy ”0:3 03550:: 9mm ............................................... 53:2 .>~ :36 .x:£u;: 3:555 flan 53:3 .>H #3::qu mm ......... 33:5 ungumfia 13mg «o 553.5 $35.5 $363553 MVH Gan—Hr Myrna mica» 353:8 me .553: 13mm. «o 5:55 235qu ca :55: 65: mica: “550.38 «a 533 umaoa mdm 51cm EoEwmm «0 523 92 . 55:: no flung: NNN 5;: «0 59:3 rm: ....:o5£:u$.§ :33: we flung: 558132 a? $335030 amok: .v:o 13mm“. :8: 5:3.5 Ezfimxdfi am: .2333 33va 323 umsm 35% .«c .532:qu notwumoaouwu:< wwu 6:358 .55: «0 $895 .5133: 8 v3: EPG 5m:w.H ..................................... eON Ian—neg 530:5” .............................. v.3 «an 29:33 Eo:w_uuna:m 95:39:“ 5:32: #8: mo 5325:: swan—3923:“ $3:th ........................ Hm mdu .33: «o 5356:. how—38923:“ «984 fig «N 33.25 Eofixm .«o nave—:3“. 3:95:35 «Ema m.wN “EV—5m EOE—u «0 3.3525 nomuwumoaouwu:< om: w: 553 ..... gm . 59:3 32::qu and 3553:?» 2;: 55:33 me 53:0..— Sa ...aanwfi~> 53:5— 3.: no "ENE: N: .Eafluo> 3995.: 6:003 «o ”3&me SH ......................... 933:?» 33.5.5 firs... no asmmwm ................. méou .EEa—Sv «o his—Saga 32:23:: 3 059m 193...: no a: .dnn§w> 33:3 5:93; «a 230mm .................... ”.mm v.3 ......wasd «0 300mm 5:73:00 «0 53>? .................. m4: «a 2:: «o 523 3355 "wannafimxr 2:2 .53 .5 «164 52:32 253w 3:02;. 30 9:55:55 MESM n v m N a 5:355 NoSN SSN cogs aacmn moowu mhvpn— «REED ##1an NQEbG Havpn mzxu ¢v§fl EZmD SZmD SZmD 22w: EZmD ..... oz :wEEonw awake—00 3—35.an .330 .33ng ~33» {Cw £3§~3m s3: Six mnwmwafioo magmfifim ”3 3.838%. Nswgsouwea xo .msu3§§.s§ § .2§§§§$§~ I .N mamfiw STRATIGRAPHY, MORPHOLOGY, AND PALEOECOLOGY, PECCARY HERD, KENTUCKY 22 gauge: 3238mm .EES—Ou 19535; «O Jaw—pl wa—Q =:u_m mo 55951? .5523? $5 3 wagon 3 cwESmwn “:5 50.29 E ”3: 95m? 4.93:: ms @3m «:9»? .dhawtw.» Empoc “EEC j}. .EzEumm we 25 3 hhufizxwfiwun .«o can"? .3585”? Sufi: $385. .on .................. mm: ......... m2 ......... i. .22:an .o a .3560 .3 uwhdmwwfi €36: ES vuofivwhm £32 u 83: 223° 9; gum is Sfi ...... .2: mg :N .2: NE «on ““““““““““““ 52558 30 3th 2783 6:358: numfla>§mzamh .3550 09:53 .on 2: m2 mmm «m: ................................ 23% £52 me< cgmwfiom :3: 4me“. .g 6%: 2:3 ....................... 2N ................................. 2:2 «$2 $72 €283an Capo? >62 (3:500 zxmzvcdw .QEmCBOE .GHNINHN .n €me 328 ER Bacm “““ 1:: 2: SN new mam 8N mew mew ................. 24.53 2:22 mmmm Dmom :oflflflwwg :8:Eam 3;. fluent—MS?» .O .nw %n wuflwgwudwaoz .................. oaH aNNm MNN mm: HON omm ........ wan—hwy yawn MESOV Humo¢ OD .M—JfidwfiOF—O an 3? ,,,, «2% 22m: . 6Q a: 2:3 2:3 EzwD . .on ...... om: ,,,,,,,,,,,, 83a SZmD .3550 .3 ,,,,, 92 m2... 3: if 233 EZwD .525 .:. Seat—E .o .m 3 355253: SN; m2 EN EN 9% ....2§= :35. was.“ Ezmb auimfiflim .Awmuom .9 «mm: 5.52:3 ........................................ 3m ............................. 33> was DM .Nfiwfldo mo Efimmfls Mdflomuflz .flmer—Eflm :MH .0 GEN ihh. .QhOEfiJg .O .nH >3 fiwkdmflwg Wuhan» LE d 48: 523$ c3 3.: $2 2:. war. 2: Na? 2m .............................. was; as: 022 .3550 6Q ............ mom ...... ms NNN a: «3 .. 32.2: 23% me; D: nafissm .Amwawm d 4.33 SEE? ............ m2 emu NE 8“ m: gm 2.25“ as: E bx 6:238 "mamunm 0:93 . 3% Ho 83% "Emsm 23:3 «5:. 3.5m Sareonfi .25 wEofism £53m $55 33.3%.: oz 3:83 :ofizwxw 55an 5.5526 mwsmmmog no: uimeaflwdwE .A ...... V $5“.an .mmsfiwvflzm .30:ma 3.2552 .«0 meGNQQ .mwz< ”hemwuwfinb waduw AOEOV .5qu METFOM “wam ”madszmoflmo «o hammh; JED .OD ”Edmmfls Ezofinz mquw fiwumcp .EZWD udvdfldo Ho 5535: Enofiwz .OEZ ”mum—Haw“ we hummhwfizp 5.5%me EEK—dz «a Efimmzs .DM 5me 3 «we? :5: “Emdumsndawowu "Sam: mmEZdoo‘: 3833?. 33230338 $03K 3238 hmammmfimgoo mucowfiflm xe EEG?» 98 mafisoxofis .mxfigtmfigg 3 .m§m§ux$mam§~ I .m mqmfiw PALEOECOLOGY 23 stated. Simpson (1949) compared the American Mu— seum skeleton and Williston’s and Peterson’s mea- surements of the Goodland skeletons with his large suite of P. compressus bones from Cherokee Cave, St. Louis, Mo. He concluded (p. 44) that It is nevertheless clear that as populations, that represented by Williston’s Kansas specimens is somewhat different from that represented by the series from St. Louis. Their variations fully intergrade, and many individuals could not well be dis- tinguished taken separately, but the mean size of their teeth (and perhaps some other averages for details) are somewhat different. They represent slightly different segments of one specific population. The Goodland, Hickman, and Denver herds of P. compressus are the only ones for which measure- ments of several individuals are available. In several parameters (Pg—M3 length, skull length, scapula length, ulna length), the range of measurements of Hickman bones demonstrates a larger size than does that of bones recovered at Goodland (table 3). Ray, Denny, and Rubin (1970) suggested aclinal decrease in size northward in P. compressus. This was indi— cated by a regular decrease in cheek-tooth length from southwest to northeast, and the cheek-tooth lengths in the Hickman herd fit within this trend. The single upper dentition measured from the Den- ver herd also fits the trend, but the lower dentition from the Denver herd is larger than would be ex- pected if a cline existed, and it is also unusually large considering the size of the upper dentition (to be compared to Ray and others, 1970, table 5). Mea- surements of postcranial elements of articulated skeletons (table 3) do not give a clear clinal picture. If the Goodland and Hickman herds are combined and used as a standard of comparison, almost all measurements from the other cited skeletons fall within their range. Total length and height of skele- ton, noted in table 3, should be considered only as approximations because of differences in posture and mounting techniques for exhibit specimens and be- cause of distortion of specimens left in natural posi- tion. In most parameters, the Denver herd falls within the size range of the combined Hickman and Goodland herds (tables 1—3; Williston, 1894, p. 26— 28). In length of temporal fossa, however, the larg— est Denver measurement is smaller than the smallest Hickman measurement and only slightly overlaps the Goodland range (Denver, 68—78 mm; Goodland, 75—82 mm). The Denver herd has markedly longer metatarsals and metacarpals than the Hickman herd (table 2), but the proximal limb bones in both herds fall within the same size range. MOLLUSCA Large gastropods ranging from 1.5 to 3 cm across occur around the peccaries as single shells, in clus- . ters of two or more shells 5—10 cm in diameter, and in layers 2.5—5 cm thick. One such layer 5 cm thick at the peccary level was composed of whole and broken shells of Aquispira alternata (Say) that cov- ered an area at least 10 by 30 cm. A few gastropods occurred among the bones of the peccaries. At the mollusk locality north of the peccary site (fig. 1), Anguispira kochi (Pfeiffer) and the giant Allogona proftmda (Say) were found by Mrs. Ruth Browne (written commun., 1967) to be limited to the Farmdale (Roxana as used here), though small Allo- gona proftmda were identified in the lowermost typi- cal Peoria Loess. AGE OF FOSSILS Radiocarbon dating of the peccary bones is not feasible because they are largely replaced by calcium carbonate (Meyer Rubin, oral commun., 1968). Aragonitic gastropod shells of Anguispim alter— nata (Say) collected about 4 feet (1.2 m) above the base of the Roxana and within 50 feet (16 m) of the peccary skeletons were radiocarbon dated as >34,000 years B.P. by Meyer Rubin (sample W—2127, in Sul- livan and others, 1970). This date is within Altonian Substage interval of 28,000—70,000 years (fig. 3; Frye and others, 1968). This assignment would hold true even if as much as 3,000 years were subtracted from the radiocarbon date to compensate for pos- sible presence of inorganic carbon incorporated into the gastropod shell. Rubin and Taylor (1963) dis- cussed sources of error in carbon-14 dates deter- mined on shell material. PALEOECOLOGY AND THE DEATH OF THE HERD During Altonian time the glacial front was in the vicinity of Shelbyville, 111., about 200 miles (320 km) north-northeast of Hickman. As the glacial ice melted, the resulting streams flowed southward heavily laden with silt. The low-gradient drainage was no doubt poorly defined, giving rise to extensive mudflats and marshlands. The Mississippi is believed to have flowed southward through the Morehouse lowland west of Sikeston, Mo., toward Hickman and at the most only a few miles west of the present Chickasaw Bluffs. Aquatic gastropod fossils found in Roxana Silt just east of the bluffs indicate the presence of another perennial stream in the Vicinity of Hickman (Finch, 1971, Roxana description). The climate and vegetal cover at the time the pec- caries roamed the Hickman area can be ascertained from associated fossil material. The large number of Anguispira kochi (Pfeiffer) and large specimens of Allogona profunda (Say) “seems to confirm the pres- ence of true forests during the time of Farmdale 24 STRATIGRAPHY, MORPHOLOGY, AND PALEOECOLOGY, PECCARY HERD, KENTUCKY [Roxana as used here] deposition. Climate was prob- ably somewhat cooler than at present” (Ruth G. Browne, written commun., 1967). Hackberry trees were common and locally abundant. At the peccary horizon and within 50 feet (15 m) a vertical zone about 1 cm wide, 30 cm long, and 2.5—8 cm high trending N. 60° E. was found to be filled with hack— berry seeds. This zone probably represents a mud- crack filling, indicating at least local ponding and drying periods. The great abundance of carbonate root casts suggest a heavy grass cover. The poses of the peccary herd indicate that the pec- caries died quietly and that their bodies were buried quickly. There is no evidence of the death pose—the head thrown back—that results from desiccation of the large tendons of the back of the neck when the body lies exposed on the surface. From the poses, death could have occurred in one of three ways: smothering in a duststorm, freezing, or smothering beneath the slumping of a vertical or overhanging loess blufi’. The third possibility was suggested to Whitmore by Cornelia C. Cameron, who is familiar with sudden and heavy slumps of loess in Iowa. The Hickman peccaries may have died in a gully that was subsequently filled; perhaps they had taken refuge in the lee of an overhanging part of the gully wall which caved in on them. This would have preserved the skeletons in undisturbed position. Modern pec- caries seek shelter in burrows, sometimes sleeping in them (Hall and Dalquest, 1963). However, compari- son of the strong orientation of the silt particles sur- rounding the Hickman peccary skeletons with the linear orientation of skeletons (figs. 6A, 8) tends to favor burial by windblown silt rather than by slump- mg. Numerous observations of windblown silt (Péwé, 1951 ; Lugn, 1968) leave no doubt that dust burial can be rapid enough to cover animal bodies in such a way as to preserve them from predators. It is much less certain, however, that a herd can be smoth- ered in a duststorm and remain in the relaxed natu- ral posture in which the Hickman herd and other herds of Platygonus compressus were found. We have made an extensive, though not exhaustive, search of travel literature on central Asia and Ara- bia and have found no reports of smothering of ani- mals in the many duststorms and sandstorms that have been described in those areas. Heat, cold, and drought are frequently cited as causing localized and, sometimes, mass mortality, which emphasizes the lack of citation of dust as a cause of death. Possibly, however, dust inhalation can be a contributary cause of death when coupled with immobilization because of cold. If extreme weather conditions rendered a Platygonus compressus herd comatose, or dead, it could be quickly covered with dust. The way in which this might happen was graphically described by Hedin (1905, p. 48) : “You have only to crouch down with your back to the wind, and you are at once well- nigh suffocated by the amount of dust and sand that accumulates in the eddy formed by the shelter of your own body.” Hay (1923, p. 214, 216) and Clarke (1916, p. 34—35) suggested that peccary herds were overcome by cold. In Altonian time the prevailing southwesterly wind combined with the intermittent northwesterly (figs. 6, 8) wind swept across the seasonally dry mudflats, picking up silt and depositing it as a blanketlike mantle on the low hills east of the flood plain. It was probably during one of the northwesterly dust- storms, accompanied by a drop in temperature, that the peccaries met their sudden death and were buried. Their linear arrangement and inclined posi- tions (fig. 9) suggest that they were walking up a trail away from the river with their backs to the wind when they became cold, crouched down, and were overwhelmed. REFERENCES CITED Browne, R. G., and Bruder, P. M., 1963, Pleistocene Mollusca from the loesses of Kentucky: Sterkiana, no. 11, p. 53—57. Case, E. C., 1921, Something about the paleontological col- lections in the University: The Michigan Alumnus, v. 27, p. 292-300. Chapman, F. M., 1936, White-lipped peccary: Natural His- tory, v. 38, p. 408—413. Clarke, J. M., 1916, Report on the geological survey, Paleon- tology, in Twelfth Report of the Director of the State Museum and Science Dept.: New York State Mus. and Science Service Bull. 187, p. 33—39. Finch, W. I., 1971, Geologic map of part of the Hickman quadrangle, Fulton County, Kentucky and Mississippi County, Missouri: US. Geol. Survey Geol. Quad. Map GQ—874. Frye, J. C., Glass, H. D., and Willman, H. B., 1962, Stratig- raphy and mineralogy of the Wisconsinan loesses of Illi— nois: Illinois Geol. Survey Circ. 334, 55 p. Frye, J. C., and Willman, H. B., 1960, Classification of the Wisconsinan Stage in the Lake Michigan glacial lobe: Illinois Geol. Survey C‘irc. 285, 16 p. 1963, Development of Wisconsinan classification in Illinois related to radiocarbon chronology: Geol. Soc. America Bull., v. 74, no. 4, p. 501—505. Frye, J. C., Willman, H. B., Rubin, Meyer, and Black, R. F., 1968, Definition of Wisconsinan Stage: U.S. Geol. Survey Bull. 1274—E, 22 p. Gazin, C. L., 1938, Fossil peccary remains from the upper Pliocene of Idaho: Washington Acad. Sci. Jour., v. 28, no. 2, p. 41—49. Goddard, E. N., chm., and others, 1948, Rock-color chart: Washington, Natl. Research Council, 6 p. REFERENCES CITED 25 Guilday, J. E., Hamilton, H. W., and McCrady, A. D., 1971, The Welsh Cave peccaries (Platygonus) and associated fauna, Kentucky Pleistocene: Carnegie Mus. Annals, v. 43, art. 9, p. 249—320. Hall, E. R., and Dalquest, W. W., 1963, The mammals of Veracruz: Kansas Univ. Mus. Nat. History Pub., v. 14, no. 14, p. 165-362. Hartnagel, C. A., and Bishop, S. C., 1922, The mastodons, mammoths, and other Pleistocene mammals of New York State: New York State Mus. Bull., nos. 241—242, 110 p. Hay, 0. P., 1923, The Pleistocene of North America and its vertebrated animals from the States east of the Missis- sippi River and from the Canadian provinces east of longitude 95°: Carnegie Inst. Washington Pub. 322, 499 p. 1924, The Pleistocene of the middle region of North America and its vertebrated animals: Carnegie Inst. Washington Pub. 322A, 385 p. Hedin, S., 1905, Lop-nor, v. 2 of Scientific results of a journey in central Asia, 1899—1902: Lithographic Inst., General Staff, Swedish Army, Stockholm, 716 p. Hibbard, C. W., and Taylor, D. W., 1960, Two late Pleisto— cene faunas from southwestern Kansas: Michigan Univ. Mus. Paleontology Contr., v. 16, no. 1, p. 1-223. Hoare, R. D., Coash, J. R., Innis, Charles, and Hole, Thorn- ton, 1964, Pleistocene peccary Platygonus compressus Le Conte from Sandusky County, Ohio: Ohio Jour. Sci., v. 64, no. 3, p. 207—214. James, W. R., and Jones, T. A., 1969, Analysis of bimodal orientation data: Math. Geology, v. 1, no. 2, p. 129—135. Jones, T. A., 1967, Estimation and testing procedures for cir- cular normally distributed data: Office of Naval Re- search, Geography Branch, Tech. Rept. 3, ONR Task No. 388—078, Contract Nonr—1228(36) , 61 p. 1968, Statistical analysis of orientation data: Jour. Sed. Petrology, v. 38, no. 1, p. 61—67. Kirkpatrick, R. D., and Sowls, L. K., 1962, Age determination of the collared peccary by the tooth-replacement pattern: Jour. Wildlife Management, v. 26, no. 2, p. 214—217. Klippart, J. H., 1875, Discovery of Dicotyles (Platyyonus) compressus, Le Conte [at Columbus, Ohio]: Am. Assoc. Adv. Sci. Proc., v. 23, pt. 2, p. 1—6. Le Conte, J. L., 1848, Notice of five new species of fossil mammalia from Illinois: Am. Jour. Sci., ser. 2, v. 5, p. 102—106. Leidy, Joseph, 1853, A memoir on the extinct Dicotylinae of America: Am. Philos. Soc. Trans, n. s., v. 10, p. 323—343. 1889, On Platygonus, an extinct genus allied to the peccaries: Wagner Free Inst. Sci. Trans, v. 2, p. 41—50. Leighton, M. M., 1960, The classification of the Wisconsin glacial stage of north Central United States: 'Jour. Ge- ' ology, v. 68, no. 5, p. 529—552. Leighton, M. M., and Willman, H. B., 1949, Late Cenozoic geology of Mississippi Valley, southeastern Iowa to cen- tral Louisiana, in Itinerary of [2d Pleistocene] field con- ference, June 12—25, 1949: Urbana, 111., State Geologists, 86 p. 1950, Loess formations of the Mississippi Valley: Jour. Geology, v. 58, no. 6, p. 599-623: reprinted as Illi- nois Geol. Survey Rept. Inv. 149. Leopold, A. S., 1959, Wildlife of Mexico—The game birds and mammals: Berkeley, California Univ. Press, 568 p. Lewis, G. E., 1970, New discoveries of Pleistocene bisons and peccaries in Colorado, in. Geological Survey research 1970: U.S. Geol. Survey Prof. Paper 700—B, p. B137— BI40. ’ Lugn, A. L., 1968, The origin of loesses and their relation to the Great Plains in North America, in Schultz, C. B., and Frye, J. C., eds., Loess and related eolian deposits of the world—Internat. Assoc. Quaternary Research, 7th Cong., Boulder-Denver, Colo., 1965, Proc., v. 12: Lincoln, Ne- braska Univ. Press, p. 139—182. Matalucci, R. V., Shelton, J. W., and Abdel-Hady, M., 1969, Grain orientation in Vicksburg loess: Jour. Sed. Petrol- ogy, v. 39, no. 3, p. 969—979. Morgan, Randy, 1967, Fossil bones in gravel pit may be 40,000 years old: Paducah, Ky., Sun-Democrat, v. 90, no. 122, p. 1, 2, May 22. National Research Council, Committee for the Study of Eolian Deposits, 1952, Pleistocene eolian deposits of the United States, Alaska, and parts of Canada [Map]: Geol. Soc. America, scale 1 : 2,500,000. Peterson, 0. A., 1914, A mounted skeleton of Platigonus lep- torhinus in the Carnegie Museum: Carnegie Mus. Annals 9, p. 114—117. Péwé, T. L., 1951, An observation on wind-blown silt: Jour. Geology, v. 59, no. 4, p. 399—401. Potter, P. E., and Pettijohn, F. J., 1963, Paleocurrents and basin analysis: New York, Academic Press, 296 p. Ray, C. E., Denny, C. S., and Rubin, Meyer, 1970, A peccary, Platygonus compressus Le Conte, from drift of Wiscon- sinan age in northern Pennsylvania: Am. Jour. Sci., v. 268, no. 1, p. 78—94. Rubin, Meyer, and Taylor, D. W., 1963, Radiocarbon activity of shells from living clams and snails: Science, v. 141, no. 3581, p. 637. Schultz, L. G., 1964, Quantitative interpretation of mineralog- ical composition from X-ray and chemical data for the Pierre Shale: U.S. Geol. Survey Prof. Paper 391—0, 31 p. Siegel, Sidney, 1956, Nonparametric statistics for the behav- iorial sciences: New York, McGraw-Hill Book Co., 312 p. Simpson, G. G., 1949, A fossil deposit in a cave in St. Louis [Mo.]: Am. Mus. Novitates 1408, 46 p. Smith, G. D., 1942, Illinois loess—variations in its properties and distribution, a pedologic interpretation: Illinois Univ. Agr. Expt. Sta. Bull. 490, p. 139—184. Sowls, L. K., 1961, Hunter-checking stations for collecting data on the collared peccary (Pecari tajacu): North American Wildlife and Nat. Resources Conf., 26th, trans, p. 496—505. 1966, Reproduction in the collared peccary (Tayassu tajacu) in Comparative biology of reproduction in mam- mals: London and New York, ‘Academic Press, Symposia, Zool. Soc. London, no. 15, p. 155-172. Stephens, M. A., 1962, Exact and approximate tests for di- rections, I: Biometrika, v. 49, p. 463—477. 1964, The testing of unit vectors for randomness: Am. Statistical Assoc. Jour., v. 59, p. 160—167. Sullivan, B. M., Spiker, Elliott, and Rubin, Meyer, 1970, U.S. Geological Survey radiocarbon dates XI: Radiocarbon, v. 12, no. 1, p. 319—334. Wagner, George, 1903, Observations on Platygonus compres- sus Le Conte: Jour. Geology, v. 11, p. 777—782. Williston, S. W., 1894, Restoration of Platygonus: Kansas Univ. Quart, v. 3, no. 1, p. 23—39. U. S. GOVERNMENT PRENTING OFFICE: 1973 O » 479-128 3%in My); V ’4; 35? @979” Oligocene Molluscan Biostratigraphy and Paleontology of the Lower Part of the 7.5 Type Temblor Formation, California 77/ GEOLOGICAL SURVEY PROFESSIONAL PAPER 791 DOCUMENTS- fiEPl‘tRWlEl-ET JUN 12 l973 1 1133.92?! IIr-uvié'renv is" mgr-”.mw .u—e—r—— , ~ ~~~7~A~~~ Oligocene Molluscan Biostratigraphy and Paleontology of the Lower Part of the Type Temblor Formation, California By WARREN O. ADDICOTT GEOLOGICAL SURVEY PROFESSIONAL PAPER 791 Marine mollusks from the basal shale (Cymric Shale Member) and the overlying sandstone (Wygal Sandstone Member) are of provincial Oligocene age. Warm-water assemblages from the Wygal Sandstone Member represent a previously unrecognized biostratigraphic unit of late Oligocene age in California UNITED STATES GOVERNMENT PRINTING OFFICE, WASHINGTON : 1973 UNITED STATES DEPARTMENT OF THE INTERIOR ROGERS C. B. MORTON, Secretary GEOLOGICAL SURVEY V. E. McKelvey, Director Library of Congress catalog-card No. 72-600377 For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, DC. 20402 Price: paper cover—$1.25, domestic postpaid; $1.00, GPO Bookstore Stock No. 2401-00284 CONTENTS Page Page Abstract ________________________________________ 1 Systematic descriptions~Continued Introduction _____________________________________ 1 Class Pelecypoda _____________________________ 22 Acknowledgments ________________________________ 3 Order Nuculoida _________________________ 22 Reports dealing with mollusks from the lower part Family Nuculidae ____________________ 22 of the Temblor Formation ______________________ 3 Family Nuculanidae __________________ 22 Stratigraphy ____________________________________ 4 Order Arcoida ___________________________ 23 Wygal Sandstone Member ____________________ 4 Family Arcidae ______________________ 23 Provincial age and correlation ____________________ 6 Order Mytiloida _________________________ 24 General considerations _______________________ 6 Family Mytilidae ____________________ 24 Cymric Shale Member ________________________ 8 Order Pterioida _________________________ 24 Wygal Sandstone Member ____________________ 10 Family Pectinidae ____________________ 24 Santos Shale Member ________________________ 13 Family Ostreidae ____________________ 27 Agua Sandstone Member _____________________ 14 Order Veneroida _________________________ 27 Paleobathymetry _________________________________ 14 Family Lucinidae ____________________ 27 Zoogeography and paleoclimatology ________________ 15 Family Ungulinidae _________________ 29 Systematic descriptions ___________________________ 16 Family Mactridae ____________________ 29 Class Gastropoda ____________________________ 17 Family Solenidae ____________________ 30 Order Archaeogastropoda ________________ 17 Family Tellinidae _______ . _____________ 30 Family Trochidae ___________________ 17 Family Veneridae ____________________ 35 Order Mesogastropoda ___________________ 17 Order Myoida ___________________________ 37 Family Cerithiidae? __________________ 17 Family Myidae ______________________ 37 Family Calyptraeidae ________________ 17 Class Scaphopoda ____________________________ 38 Family Naticidae ____________________ 18 Family Dentaliidae __________________ 38 Family Ficidae ______________________ 20 Fossil localities __________________________________ 38 Order Neogastropoda _____________________ 20 References ______________________________________ 40 Family Neptuneidae __________________ 20 Index __________________________________________ 45 PLATE FIGURE TABLE 1. ILLUSTRATIONS [Plates 1—9 follow index] Avila, Anadara, Crenomytilus?, Leptopecten, Vertipecten, Yoldia. Crassostrea, Lucinoma, Pecten, Vertipecten. Crenomytilus?, Here, Lucinoma, Miltha, Vertipecten. Crenomytilus, Felaniella, Solen, Spisula, Tellina, Tellina7. Heteromacoma, Manama, Pseudocardium?, Tellim. Amiantis, Dosim'a, Heteromacoma, Pitar, Securella, Tellina. Amiantis, Clementia, Dosim'a. Calyptraea, cerithiid?, Clementia, Crepidula, Dentalium, Nation, Panopea, Tegula. Bruclarkia, Calicantharus, Ficus, Kelletia?, Never'ita, Polimlces, Sinum, Siphonalial P Index map showing location of Temblor Range _________________________________________________ “2 Stratigraphic nomenclature of the type Temblor Formation, Temblor Range, Kern County, Calif ___ 4 Geologic index map showing distribution of the Temblor Formation, central Temblor Range, Kern County, Calif __________________________________________________________________________________ 5 Molluscan and foraminiferal provincial stage classification of the type Temblor Formation _________ 8 Zoogeographic profiles for selected Tertiary faunal assemblages from the San Joaquin basin, California _ 16 TABLES P Systematic list of mollusks from the Wygal Sandstone Member of the Temblor Formation, Temblor we Range, Calif ___________________________________________________________________________________ 7 III IV TABLE 2. 9’ CONTENTS Chart showing specifically determined mollusks from the Wygal Sandstone Member of the Temblor For- mation _________________________________________________________________________________________ Chart showing ranges of some species of Bruclarkia and Vertipecten _____________________________ Molluscan species from the Wygal Sandstone Member that are not known to occur in strata of Refugian age ____________________________________________________________________________________________ Molluscan genera and subgenera from the Wygal Sandstone Member that are not known to occur in strata of Refugian age _________________________________________________________________________ Mollusks from the Agua Sandstone Member of the Temblor Formation ______________________________ Page 11 12 12 13 14 OLIGOCENE MOLLUSCAN BIOSTRATIGRAPHY AND PALEONTOLOGY OF THE LOWER PART OF THE TYPE TEMBLOR FORMATION, CALIFORNIA By WARREN O. ADDICOTI‘ ABSTRACT The type Temblor Formation consists of about 1,800 feet of alternating shale and sandstone units exposed in the central part of the Temblor Range along the southwest margin of the San Joaquin Valley in central California. The lower part constitutes a unique stratigraphic continuum of marine middle Tertiary strata in the California Coast Ranges. Molluscan stages of provincial late Oligocene and early Miocene age can be recognized on the basis of assem- blages from the lowest sandstone members of the formation. These units are directly associated with deeper water foram- iniferal assemblages that constitute the type section of the Zemorrian Stage of the provincial microfaunal sequence. Elsewhere in central and southern California, the Oligocene and the earliest part of the Miocene are usually represented by nonmarine deposition. A small assemblage from the Cymric Shale Member at the base of the Temblor Formation contains Bmclarkia columbiana (Anderson and Martin), the presence of which suggests correlation with the upper part of the Refugian Stage [:“Lincoln Stage” of western Washington]. The Cymric Shale Member was originally included in the lower part of the overlying Zemorrian Stage of the microfaunal chronology based on meager foraminiferal data. Assemblages from the overlying Wygal Sandstone Mem- ber comprise a biostratigraphic unit of provincial late Oligocene age that had previously been assigned to the early Miocene “Vaqueros Stage.” Included in this fauna are the stratigraphically restricted species Pecten sanctaecru- zensis Arnold, Bruclarkia seattlcnsis Durham, and Verti- pecten alexclarki n. sp. This fauna is correlated with the lower part of the Zemorrian Stage of the microfaunal chronology. The stratigraphically higher Agua Sandstone Member contains small assemblages referable to the early Miocene “VaquerOS Stage.” Included are the stratigraphically re- stricted species Crassostrea vaquerosensis (Loel and Corey) and Macrochla/mis magnolia (Conrad). This member is cor- related with the upper part of the Zemorrian Stage. The molluscan fauna of the Wygal Sandstone Member is indicative of a warm, shallow-water depositional environ- ment—less than 20 fathoms and possibly much shallower at the base. Many of these genera are today restricted to sub- tropical and tropical latitudes along the Pacific coast of Mexico and Central America. The hermatypic coral Fam‘tes provides further evidence of warm-water conditions. Yet, this inner sublittoral biofacies contrasts with the middle to lower bathyal environments inferred from benthonic foram- iniferal assemblages of the underlying Cymric Shale Member and the overlying Santos Shale Member. Fifty mollusks, almost all of them previously unrecorded from the Wygal Sandstone Member, are treated systemati- cally. Six of these represent new species, although only one, Vertipecten alexclarki n. sp., is represented by material adequate to permit formal description. A new tellinid sub- genus, Olcesia (type: Macoma piercei Arnold), has its low- est stratigraphic occurrence in assemblages from the Wygal Sandstone Member. INTRODUCTION The type area of the Temblor Formation is in the central part of the Temblor Range near the western margin of the San Joaquin basin about 40 miles west of Bakersfield, Calif. (fig. 1). In this area the Temblor Formation consists of alternating shale and sandstone totaling about 1,800 feet in thickness (Woodring and others, 1940). The most definitive biostratigraphic study of the type Temblor Formation (Kleinpell, 1938) was based on benthonic Foraminifera. In that report, and in an earlier one (Kleinpell, 1934), he showed that the lower part of the type Temblor Formation was of pre-middle Miocene age. Prior to Kleinpell’s studies most workers regarded the Temblor Forma- tion to be wholly of middle Miocene age and to be referable to what later became known as the “Tem- blor Zone” (Clark, 1941) or “Temblor Stage” (Weaver and others, 1944; Durham, 1954). Klein- pell’s (1938) data were substantiated in a brief biostratigraphic analysis of known molluscan as- semblages from the vicinity of the type Temblor Formation by Woodring, Stewart, and Richards (1940, p. 131), who concluded that the lower part of the type Temblor Formation is “of the age of the Vaqueros elsewhere” [early Miocene]. The molluscan faunas of the type Temlblor For- mation and its lateral correlatives in the northern part of the Temblor Range have received only cur— sory study. The largest recorded faunal list prior 1 MOLLUSCAN BIOSTRATIGRAPHY, PALEONTOLOGY, TYPE TEMBLOR FORMATION, CALIF. @ Il/l./ \S‘ 509/, ‘7 an. ”49/” Lost Hills z; .: n’l/ 1;, ) 4”, w, 1 4/ ’2, ”/7. ’4 5‘ ’7/ 43 47 (”fa - % I; .glE/I 35.10 4F; 6 I?“ 617 < w; ¢ } O “2 (’0 ’ _ ’//',S,/ Area shown 7,1, "’4” ’9' ’r, in Figure 3 O (4" 7/6 ,9 ’4 ”4. o ‘94,, . n '7 745/ I [’2 ‘9 Z;- "/0 3‘“; //I ”5/ 7/. ”é, McKittrick ’7 ’6; 3:2,} 9' 7"». /’/ o / L ”4,, 7/”! E f}, / 7", I, h/ Cs - 2/01 ’¢ Eta/’2’, Taft . I ”801.4%,” 4,// 8 E 1;;,'/ _~ -, , ’I x .11.] I‘\ \V e ,/ //// ’/, E //r ,—.=::7_””'.'\l\\\\ 44:7,, “54’4” 4/109 g Coumy ’ /’//4/ ’5'”- ’//// ’ ,7; (2/ —COUNTY l 9/, - u ,; ,,/ . // ’29 2III; 196/) 04/70 ”1’ _l (5’2 Marlcopa 7/, /////, § 4 , {5/6” //,, | ‘7'.’ , ’4 / ’ ' ’ n 4/, -: o ”1', , 0 ///a 7 ,. C9 ’5‘; .47 I’ll/(0”, ’4’ $ ® J 350 I .=—: // ’ - I; 4",, ~\.\ l 120 ° 1 19 ° FIGURE 1.—Location to this study (Anderson, 1905, p. 170), for example, is from the Wygal Sandstone Member, but it in- cludes only 11 mollusks in contrast to the 50 mol- luscan taxa treated in this report. The lack of in- tensive paleontologic and biostratigraphic study of the Temblor Formation may result from the fact that in the type area all the sandstone members ex- cept the lowest one, the so-called Phacoides Sand [Wygal Sandstone Member of this report], have sparse assemblages of larger marine invertebrates. Nevertheless, collections at hand from the type area in the central part of the Temblor Range and from more fossiliferous exposures in the northern part of the range are adequate to establish a local bio- stratigraphic sequence. Although this report is pri- marily concerned with the Wygal Sandstone Mem- ber, the sequence of Oligocene and early Miocene molluscan faunas of the Temblor Range is briefly 20 MILES of the Temblor Range. outlined and correlated with the established pro- vincial molluscan stages of Weaver and others (1944) and with the benthonic foraminiferal stages of Kleinpell (1938) and Schenck and Kleinpell (1936). The conclusions of this report differ from those of earlier workers in that at least one, and possibly two, Oligocene biostratigraphic units of stage magnitude are represented in the lower part of the type Temblor Formation. The next to the lowest of these two units, which is represented by the fauna of the Wygal Sandstone Member, is the subject of this report. Molluscan data from the underlying Cymric Shale Member and from the overlying Santos Shale and Agua Sandstone Mem- bers are considered in a following section on pro- vincial age and correlation in order to establish the limits of the late Oligocene biostratigraphic unit represented by the Wygal fauna. ANNOTATED REFERENCES TO THE MOLLUSCAN FAUNA 3 ACKNOWLEDGMENTS Several people provided helpful assistance during the course of this study. The manuscript has been critically read by E. E. Brabb, T. W. Dibblee, Jr., E. J. Moore, and R. L. Pierce. Their comments and suggestions have improved the report and are deeply appreciated. Dibblee mapped the geology in this report and discussed stratigraphic relationships and formational nomenclature. Olgerts Karklins, US. Geological Survey, and Alan H. Cheetam, US. National Museum, kindly identified an en- crusting bryozoan, and J. Wyatt Durham, Uni- versity of California, Berkeley, identified the corals. Dibblee, Saburo Kanno, R. L. Pierce, H. S. Sonneman, and H. C. Wagner assisted in making some of the mega-invertebrate collections from the Wygal Sandstone Member. J. G. Vedder collected mollusks from the Quail Canyon Sandstone Member of the Vaqueros Formation (Dibblee, 1972) to which this report refers. M. A. Murphy, Uni- versity of California, Riverside, granted access to museum collections and lent specimens for figuring in this report. A. M. Keen and N. J. Silberling, Stanford University, and J. H. Peck, J r., University of California, Berkeley, also granted access to mu— seum collections and type collections at their respec- tive institutions. R. L. Pierce and C. C. Church, Bakersfield, Calif., discussed problems of forami- niferal correlation with the writer. Fossil photog- raphy is by Kenji Sakamoto. I am particularly indebted to H. S. Sonneman, who initially suggested the biostratigraphic study of mollusks from the type Temblor Formation and who collaborated during much of the fieldwork and preparation of this report. REPORTS DEALING WITH MOLLUSKS FROM THE LOWER PART OF THE TEMBLOR FORMATION Annotated references to the molluscan fauna of the Wygal Sandstone Member are listed chronologi- cally in this section. Stratal terminology and fossil names indicated in the annotations are those of the authors cited. Place names are indicated in figure 3. Complete bibliographic citations are included in the references at the end of the report. 1905. Anderson, F. M. Designated the area between Car- nara Spring [Carneros Creek] and Temblor [Tem- blor Ranch], to the south, as the type area for his Temblor Beds. Seven species and four genera of mollusks are listed from his lower sandstones. This is the most complete published fauna] list from the Wygal Sandstone Member of the Temblor Formation. 1910. Arnold, Ralph, and Johnson, H. R. The basal 25 ft of a 445-ft measured section of the Vaqueros Sand- stone [Temblor Formation] between Salt Creek and Temblor Creek (W‘yé sec. 25, T. 29 S., R. 20 E.) contains Ostrea and many indeterminate, poorly preserved fossils. A collection from between Chico Martinez and Carneros Creeks (USGS 5149) made during fieldwork for this report was not re- corded. Species from this locality are listed in table 1. 1921. English, W. A. Four measured sections of the Va- queros Formation [Temblor Formation] between Salt and Media Agua Creeks are included in a correlation diagram. The basal sandstone beds con- tain a large [molluscan] fauna, best developed near Carneros Spring and Media Agua Creek. 1936. Hanna, G D. Abundant Phacoides acutilineatus occur in the basal part of Anderson’s (1905) Temblor Formation. The jaw of a small-toothed whale, a zeuglodont, is reported from these strata. 1936. Schenck, H. G. Reported specimens of Acila conradi (Meek) var. from beds of “Vaqueros” age on the southwest flank of McDonald anticline (sec. 31, T. 27 S., R. 19 E.). The associated molluscan as- semblage was regarded by Alex Clark as correla- tive with the lower sandstone member of the Temblor Formation on Carneros Creek [Wygal Sandstone Member], although it contains species such as Bruclarkia barkeriamz and Chione cf. C. panzana that are indicative of a Miocene rather than Oligocene provincial age. The fossil locality is probably from near the base of the Buttonbed Sandstone Member, which unconformably overlies the Eocene Point of Rocks Formation in this area. A recent collection from the basal 5 ft of this unit (USGS loc. M3988) contains Trophosycon ker- nianum, Olive califomia, Amussiopecten sp., and Lyropecten crassicardo, fossils suggestive of a middle Miocene (“Temblor”) provincial age. 1938. Kleinpell, R. M. Four molluscan taxa including Chlamys sespeemis in the Phacoides reef [Wygal Sandstone Member] and Bruclarkiu barker-tuna [B. seattlensis] are recorded from localities in the Zemorra Creek—Chico Martinez Creek area. This assemblage occurs in the lower part of his Zemor- rian Stage and was correlated with the Tur'rz'tella inezana. zone. 1940. Woodring, W. P., Stewart, Ralph, and Richards, R. W. Abundant Lucinoma acutilineata occur in their “Phacoides” reef member of local usage [Wygal Sandstone Member] near the base of a measured section of the type Temblor Formation on Zemorra Creek. Typical “Vaqueros Stage” spe- cies are not known to occur in this member, al- though the faunal facies is such that they might be expected to occur in it. “Vaqueros Stage” index species occur in stratigraphically higher units in this part of the Temblor Range. 1943. Curran, J. F. Six mollusks are listed from the “reef” [Wygal Sandstone Member] at the base of his lower Temblor Sand-stone in the vicinity of Chico Martinez Creek. 1959. Stinemeyer, E. S., Beck, R. S., Ortalda, R. A., Espen— scheid, E. K., Bainton, J. D., and O’Keefe, M. S. Exposures of the basal part of the Phacoides Sand [Wygal Sandstone Member] on Zemorra Creek contain Lucimr acutilmeata [Lucino'ma acutiline- ata], Pecten brannem’ [Vertipecten alexclarki n. sp.], Ostrea sp., and Bmclarkia barkeriana [B. seattlensis]. Kleinpell, R. M., and Weaver, D. W. The mollusk- bearing lower (“Phacoides Reef”) sandstone mem- ber of the Temblor Formation [Wygal Sandstone Member] is correlated with the middle sandstones of the type Alegria Formation of the Santa Bar- bara embayment and with their Soda Lake Sand- stone of the Cuyama Valley area, eastern San Luis Obispo County. Elliott, W. J., Tripp, Eugene, and Karp, S. E. Lucina acutflineatus [Lucinoma acutilineata] is the pele— cypod for which the Phacoides Sand [Wygal Sand- stone Member] was named. Foss, C. D., and Blaisdell, Robert. Their “Phacoides Sand” member [Wygal Sandstone Member] of the Temblor Formation contains mollusks alleged to be referable to the Turritella inezana zone; seven species are listed. This member is unconformably overlain by their “Lower” Santos Shale Member of the Temblor Formation. Addicott, W. 0. About 35 percent of the still—living molluscan genera from the so-called Phacoides sandstone [Wygal Sandstone Member] are today restricted to warm-water latitudes of the eastern North Pacific. Addicott, W. 0. Fauna] data presented in Addicott (1969) are tabulated. Thirty-one molluscan genera occur in the so—called “Phacoides” reef [Wygal Sandstone Member]. The fauna is of much warmer water aspect than that of the subjacent Refugian Stage of the San Joaquin basin. Mallory, V. S. The “Lower Temblor sandstone” [Wygal Sandstone Member] unconformably over- lies Eocene sandstone at Media Agua Creek. Three mollusks including Lucinoma acutilineata are listed. Addicott, W. 0. A significant increase in the per- centage of molluscan genera of tropical and sub- tropical affinities occurred during the late Oligo— cene in the San Joaquin basin. The increase is based on comparison of the middle Oligocene fauna of the Acila shumardi Zone of the San Emigdio Mountains with the fauna of the so-called “Pha- coides” sandstone [Wygal Sandstone Member] of the type Temblor Formation. Addicott, W. O. Mollusks from the Phacoides Sand Member of Stinemeyer and others (1959) are of post-Refugian and pre—“Vaqueros” age. This fauna is believed to represent a previously unrec- ognized stage in California and is coeval with the lower part of the “Blakely Stage” of western Washington. This summary report was written after the present paper was completed; the sections in it dealing with the lower part of the type Tem- blor Formation are summarized from this report. 1973. Addicott,W. O. The fauna of the so-called “Phacoides reef,” formerly assigned to the “Vaqueros Stage,” represents a previously unrecognized post- 1963. 1968. 1968. 1969. 1970b. 1970. 1970c. 1972. MOLLUSCAN BIOSTRATIGRAPHY, PALEONTOLOGY, TYPE TEMBLOR FORMATION, CALIF. “Lincoln,” California. STRATIGRAPHY The Temblor Formation was named (as Temblor Beds) by Anderson (1905) for exposures between Temblor Ranch and Carneros Creek in the central part of the Temblor Range about 15 miles north- west of McKittrick, Calif. The base of the forma- tion is marked by an unconformity with the under- lying Point of Rocks Sandstone of Eocene age; the top, by a conformable contact with the Monterey Shale. The Temblor Formation consists of several shale and sandstone units totaling about 1,800 feet in thickness in the type area (Woodring and others, 1940, p. 130). These units have been given informal names in the course of petroleum exploration and development in oil fields bordering the eastern flank of the Temblor Range (Cunningham and Barbat, 1932; Gester and Galloway, 1933; Goudkoff, 1941), and these names have been summarized in many publications (Woodring and others, 1940; Weaver and others, 1944; Stinemeyer and others, 1959; Foss and Blaisdell, 1968). A comparison between earlier stratigraphic nomenclature and that for the mem- bers of the Temblor Formation formally named and described by Dibblee (1973) is shown in figure 2. pre-“Vaqueros” molluscan stage in Formation Member Earlier usage (Stine- meyer and others, 1959) Buttonbed Sand Media Shale Carneros Sand Dibblee (1973) and this report Buttonbed Sandstone Media Shale Carneros Sandstone Upper Santos Shale Santos Shale “Agua Sand” interval Agua Sandstone Lower Santos Shale Temblor Formation Santos Shale iPhacoides Sand Wygal Sandstone Salt Creek Shale Cymric Shale FIGURE 2.— Stratigraphic nomenclature of the type Temblor Formation, Temblor Range, Kern County, Calif. WYGAL SANDSTONE MEMBER The Wygal Sandstone Member of the Temblor was named by Dibblee (1973) for exposures on Chico Martinez Creek in the central part of the Temblor Range about 8 miles southwest of South Belridge oil field. This unit was previously known by the informal local name Phacoides sand, or Pha- coides reef, following Kleinpe‘ll (1938, p. 107). The informal name Phacoides was based on the abun— dant articulated valves of the mollusk Lucinoma acutilineata that characterizes this unit throughout 35°30' WYGAL SANDSTONE MEMBER 5 its surface exposure. The Wygal crops out in a folded but generally northeast-dipping marine sequence extending over a 12-mile area from the west fork of Devilwater Creek on the northwest (Glover, 1953) to Temblor Creek on the southeast (fig. 3). This sandstone member is 3 or 4 feet thick on the north side of Media Agua Creek (USGS 10c. M4473) near its northwestern limit and increases to as much as 85 feet on Zemorra Creek (Woodring and others, 1940). Throughout most of the area, however, the fossiliferous part of the member ranges from 20 to 40 feet in thickness. In the thicker sections, there there are usually three or four pelecypod-be-aring, calcareous sandstone beds that average about 2 feet l u} ui m 0 p N n:- . of) T. 27 s. T. 28 S. Antelope Hills oil field McDonald Anticline oil field South Belridge / oil field v.31 5) 5.3.. §. W K; s/ T. 28 s. SE EXPLANATION ' V/m M . M : M4469 ) Temblor Formation /' 5149 M3636 MWMZ- Fault 4‘ (‘J' 3° M4466 _ _—$—_ (\2/ M4457 \_ \Sqlt / Anticlinal fold ‘ / M3579>i UCR 1106 x . ‘ . Cymrlc —*— ’9 9432 Q“. t ' . Synclinal fold 4 C}- fleld Fossil localities 4" fl M3772 O ,- " 5159 g . / U.S. Geological Survey, , M3978 Wygal Sandstone Member \ . +UCR 1235 §/ California Univ., Riverside, . «$- \ Wygal Sandstone Member : Temblor Rancho ( .ucRuoe 9 N _ California Univ., Riverside, 0: n: ) Cymric Shale Member /.. - l ‘ o 5 MILES “9°45 I l l 1 l A FIGURE 3.—Distribution of the Temblor Formation, central Temblor Range, Kern County, Calif, and mega-invertebrate fossil localities in the Wygal Sandstone (X) and Cymric Shale (0) Members (geology from Dibblee, 1968). The northwestern limit of the Wygal Sandstone Member is in sec. 17, T. 28 S., R. 19 E. (Glover, 1953). 6 MOLLUSCAN BIOSTRATIGRAPHY, PALEONTOLOGY, TYPE TEMBLOR FORMATION, CALIF. in thickness, separated by more friable, less cal- careous, poorly exposed sandstone. The basal sand- stone bed is coarse to very coarse grained and is characterized by abundant pelecypod valves and a few mollusk-bored cobbles. In the area between Carneros and Temblor Creeks, the basal sandstone usually overlies a poorly exposed, splintery, gray shale known locally as the Salt Creek Shale but recently renamed as the Cymric Shale Member (Dibblee, 1973); locally, however, the sandstone overlaps the Cymric Shale Member and directly overlies the Eocene Point of Rocks Sandstone (Stinemeyer and others, 1959, p. 6). From Santos Creek (fig. 3) northward the basal sandstone of the Wygal unconformably overlies the Point of Rocks Sandstone ; it progressively truncates about 700 feet of strata in the Point of Rocks Sandstone between Santos Creek and its northwestern limit at Devil- water Creek (Glover, 1953). This contact is well exposed where it crosses sharp ridges on the north and south sides of Media Agua Creek (fig. 3). Although an unconformity is not recognized at the base of the Wygal [Phacoides] Sandstone Mem- ber in basinward oil fields southeast of the Temblor Creek—Chico Martinez Creek area (Weddle, 1966; Foss and Blaisdell, 1968; Anderson and Land, 1969), there is a distinct unconformity at the base of the Wygal in most of the oil fields adjacent to the northeastern flank of the central and northern parts of the Temblor Range (Ritzius, 1954; Bruce, 1956; Park and Weddle, 1959; and Land, 1968). In the subsurface, the Wygal [Phacoides] Sandstone Mem- ber progressively overlaps and truncates Oligocene and upper Eocene formations in a westerly direction. Surface and subsurface evidence indicates, there- fore, that the unconformity at the base of the Wygal Sandstone Member is an even stronger and better defined regional feature than the one at the base of the Cymric [Salt Creek] Shale Member. The basal sandstone of the Wygal Sandstone Member is well exposed about half a mile southeast of Zemorra Creek, in the southern part of the out- crop belt. There it contains abundant Crassostrea and scattered valves of the stratigraphically sig- nificant species Pecten sanctaecruzens-is Arnold (USGS loc. M4467). The stratigraphically higher fossiliferous strata are generally poorly exposed except in gullies and in areas of southern exposure on moderate to steep slopes. Pelecypod valves occur in fine-grained to very fine grained calcareous sandstone lenses or concretions. The valves are so abundant that the Wygal Sandstone Member can usually be recognized and mapped on the basis of fossil float—generally articulated valves of Lucinoma acutilineata and, to a lesser extent, Heteromacoma rostellatw—even in areas of low relief and on north-facing slopes. The top of the Wygal Sandstone Member is marked by a 4- to 5-foot—thick bed of white-weather- ing glauconitic sandstone. This bed is well exposed in trenches between Media Agua Creek and Stone Corral Creek (El/2E1/2 sec. 23, T. 28 S., R. 19 E.), on the south side of Zemorra Creek (NW1/LSE14 sec. 9, T. 29 S., R. 20 E.), and about a mile north- west of Salt Creek (SW. cor. sec. 14, T. 29 S., R. 20 E.). It is a characteristic sub-surface marker bed in the Cymric oil field, the productive limits of which extend for a distance of 10 miles southeast of Temblor Creek (Anderson and Land, 1969). Recent collections from 19 localities in the Wygal Sandstone Member, together with three older col- lections (USGS locs. 5149 and 9432; UCR 1235), have yielded a molluscan fauna of 50 taxa, princi- pally pelecypods (table 1). The largest, best pre- served assemblage is from the divide on the south side of Media Agua Creek (USGS locs. M3280 and M3772; 31 taxa). Here the base of the Wygal is marked by scattered calcareous sandstone concre- tions containing abundant pelecypod valves. In the upper part of this 6-foot-thick unit, Lucinoma acu- tilineata (Conrad) and Heteromacoma rostellata (Clark) are so abundant that this part can be called a Lucinoma-Heteromacoma bed. Gastropods are more common at this locality than elsewhere, but they constitute only a small fraction of the mollusk assemblage. PROVINCIAL AGE AND CORRELATION GENERAL CONSIDERATIONS The lower part of the type Temblor Formation constitutes a unique stratigraphic continuum of marine middle Tertiary strata in the California Coast Ranges. Two molluscan stages of provincial late Oligocene and early Miocene age can be recog- nized from shallow-water faunal assemblages from sandstone members of the formation. These units are directly associated with benthonic foraminiferal assemblages from interbedded shale and siltstone that constitute the type section of the Zemorrian Stage of the provincial microfaunal sequence (Klein- pell, 1938). Elsewhere in the Coast Ranges of cen- tral and southern California, the Oligocene, and perhaps the early part of the Miocene, are usually represented by nonmarine strata—the Sespe Forma- tion and correlative units—that constitute a marked hiatus between the widespread marine deposition of the Eocene and of the Miocene. The significant features of the Temblor Range sequence can be sum- PROVINCIAL AGE AND CORRELATION TABLE 1,—Systematic list of mollusks from the Wygal Sandstone Member of the Temblor Formation, 7 Temblor Range, Calif. Localitiesl '6 Fossil g 5 ON '4 no G ‘9 on Q i la GD 1* no G v-l N ” H a .. a a: a s s a s 5: s s s s s s 3 s s s a: 3 § g one a: «a co 3 co co co a: v w v v E v v v D n: a o- S: S S S E E S S E S S S S E E S S D Gast;opo¢§s: eaua n. sp _______________ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ 7 ___ ___ ___ ___ ___ ___ ___ Cerithiid? _________________ _ ___ ___ ___ ___ i: ___ ___ ___ ___ ___ ___ _,_ ___ ___ ___ ___ _ ___ x Calyptraea dieymma (Conrad) _ _ -_ x ___ __ ___ __- ___ ___ ___ ___ ___ ___ ___ ___ _ ‘ ___ -_- "— Crepidula cf. 0. ungana Dall ______ ___ X ___ ___ ___ ___ x ___ ___ ___ ___ ___ ___ ___ ___ ___ __- "— Nation, (Notice) n. sp ___________ ___ ___ ___ ___ ___ ___ ___ __, ___ ___ ___ ___ ___ ___ ___ ___ ___ "— Neverita, (Glossaulam) thmnaonae "' "— Hickman _____________________ ___ ___ __- ___ ___ x ___ ___ ___ ___ ___ ___ _ Palinicea n. sp __________________ -__ ___ ___ ___ ___ ___ _u ___ ___ ___ _-_ -:: Sinum cf. S. scopulosum (Conrad) _ ___ ___ ___ -__ ___. X ___ ___ ___ ___ ___ _-- _-_ ___ ___ Ficus cf. F. modeata (Conrad) ___- -_- ___ _-- ___ ___ X ___ ___ X ___ _-_ ___ ___ __- _-_ -__ Calicanthama cf. C. bramteri (Clark and Arnold) ___ __- -__ -__ ___ ___ _-_ ___ --.. _-- ___ ___ -.... _-_ -__ ___ ___ ___ X ___ Callcantharus dalli (Clark) ___ ___ ___ X ___ ___ ___ ___. ___ ___ ___ ___ ___ _-_ __- ___ ..__ ___ ___ Kelletz‘a? sp ____________ _ ___ X ___ ___ ___ ___ ___ ___ ___ ___ -__ ___ ___ ___ ___ ___ -__ Siphonalia? sp _________________ ___ X ___ ___ ___ ___ ___ ___ ___ ___ ___ -__ -__ ___ --_ ___ ___ _-_ --_ Bruclarlcia seattlenais Durham ___- X ___ ___ X __- X ___ X X -__ _-_ __- X ___ ___ ___ Pelecypods: Avila (Truncacila) mum Clark _--- __- __- _-_ T ___ ___ 7 ___ -__ __- ___ ___. X ___ ___ ___ ___ X __- ___ _-- Yoldia (Kalauoldia) tenuissima Clark _______________________ ___ _..- X ___ ___ X ___ ‘l ___ X ___ ___ X ___ 1 ___ ___ ___ ___ ___ X Anadara (Amulara) submmttereyana Clark _______ cf. ___ _-_ X X X X ___ X ___. ___ ___. ___ ___ -__ --- ___ __- ___ ___ ___ Crenomytdua? cf. 7 arnoldi (C ark) _____________________ ___. ___ __- ___ --_ X ___ cf. X X _-- __- ___ ___ -__ ___ ___ -..- ___ ___. ___ Crenomytilus expansus (Arnold) __ ___ X ___ _-_ ___ ___ ___ ___ ___ __- ___ ___ ___ _-_ ___ ___ __- --- _-_ ___ ___ Aequipecten sp _________________ ___ ___ _-_ X ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ ___ __- ___ ___ Pecten (Pecten) aanctaecruzemis Arnold _______________________ ___ ___ ___ ___ ___ ___ ___ ___ ___ __- ___ ___ ___- X ___ ___ -__ ___ ___ -__ --- Vertipecten alexvlarki Addicott, n. sp ________________________ ? __- ___ X X -__ X X ___ ___ X X X ___ X --_ X ___ X ___ __- Crassostrea eldridyei ynezam Loel and Corey) _____________ -__ ___ ___ X _-_ X ___ ___ X '! _-_ __- X --_ ___. ___ ___ __- _-.. Crusoatrea'! sp _________________ X ___ ___ ___ ___ -__ ___ ___ ___ _-_ _-_ _-_ -_- X X __- X ___ Here excavate (Carpenter) _______ ___ ___ X X ___ X X ___ X -__ ___ ___ X __- ___ ___ ___. __- ___ Lucimrma. acutilineata (Conrad) .._ X ___ X X X X X X X X X X ___ X X X X Miltha, (Miltha) aanctaecrucis (Arnold) ____________________ X ___ X X --_ X X ___ __- X ___ ___ ___ X X X ___ __- -__ ___ ___. Felaniella. harfordi (Anderson) ___- ___ ___ X X _-_ -__ -__ ___ ___ ___ -__ ___ X ___ ___ ___ __- --_ ___ _-- Pseudocardium? sp ______________ ___ ___ ___ ___ ___ _-_ __.. ___ ___ ___ X -__ ___ __- ___ ___ ___ ___ ___ Spisula cf. S. albaria (Conrad) __ X _-_ ___ ___ --_ ___ X X X T X X X ___ ___ n. sp.? ____________________ ___ ___ ___ X ___ ___ X ___ ___ X ___ ___ ___ ___ _-_ ___ ___ __- ___ mmonenais Packard _________ ___ ___ __- X __- ___ ___ -__ ___ ___. _-_. ___ _-_ ___ of. ___ __- __- ___. ___ ___ cf. S. rushi Wagner and Schilling _________________ ___ -_- ___ X ___ ___ _-- X X X ___ ___ ___ ___ ___ ___ X X f ___ ___ Solen afi. S. yravidus Clark ______ ___ ___ ___ X __- X ___ ___ ___ ___ __- ___ _n- ___ -__ _-_ ___ ___ _-_ _-_ Tellina (Olcesia) piercei (Arnold)- ___ ___ X ___ ___ X -__ ___ ___ ___ _-_ ___ ___ _-- _-_ --_ -__ (Oudardia?) emacerata Conrad- _-_ ___ ___ X ___ ___ ___ ___ __- ___- _-_ ___ ___ ___ ___ -,_ ___ ___ ___ (Tellinella) tenudi‘neata Clark__ ___ ___ X ___ _-_ ___ ___ ___ ___ T ___- _-_ __- ___ ___ ___ ___ _-_ cf. T. towmendenais Clark --__ ___ ___ ___ X ___ ___ ___ ___ ___ -_- ___ ___ ___ ___ ___ ___ __- --_ _-_ Tellina? of. T. vancouverensia Clark and Arnold ___________________ ___ ___ ___ X __- ___ ___ ___ ___ ___ _-_ ___ ___ ___ -__ ___ ___ ___ -__ Macoma arctata (Conrad) ________ ___ X X X Cf. -__ ___ ___ ___ ___ X _-_ X X __- cf. X Heteromacoma rostellam (Clark) __ ___ X X X ——- ——— X X X ——— X ___ X X X X X Amiantis mathewwm' (Gabb) _____ ___ X ___ X X ___ X X Cf- _—— T ——_ 6!. Cf. -__ --_ “— n. sp ______________________ --- T ___. X ~__ ___ —.-— —-— —-- —-- X ——- —-— ——— ——- -—- ——— —-- -—— Clemenlia (Egesta) per-tennis (Gabb) ______________________ X X __- X X ___ X ——— ——— ——— X ——— X X -—— -—— —-— —-- ——— Dosinia (Doaima) whitneyi (Gabb)_ X X __- X ___ __- ___ _-_ -_- X _-- 7 cf. X --- --_ Pitar (Katherinella) cf. P. (K.) califmia (Clark) _____________ ___ __- ___ X __- ___ ___ ___ ___ ___ __- ___ X -__ ___ ___ _-_ -__ X _-.. _-_ sp ________________________ ___ -_.. ___ ___ ___ ___ ___ X -_- ___ _-- -_- ___ ___ ___ __- ___ _-_ __- ___. Securella cf. S. cryptolineata (Clark) _______________________ ___ -—— ——- X ——— ——- —-- -—— -—— -—- --- —-- —-- --- -—— ——- —-- --- —-- ——- —~- Panopea rammensis (Clark) _____ X ___ ___ X ___ X ——— —-— --— --— -—- —-— --- Cf- -—- ——— —-- —-— Scaphopod: Dentalium laneemis Hickman ___- -__ ___ ___ X X X -—— ——— X --- -—— --- ——— --— --- —-— ——— --- -—- —-- --- 1See p. 17 for explanation of symbols. marized as ( 1) the succession of shallow-water Oligocene and early Miocene molluscan assemblages in demonstrable superpositional relationships and (2) the alternation of these molluscan assemblages with thoroughly studied foraminiferal-bearing strata. What is offered, then, is a unique framework for the stratigraphic succession of shallow-water molluscan faunas from the Oligocene into the Mio~ cene and an opportunity to relate these directly to the deeper water benthonic foraminiferal sequences The biostratigraphic units in the lower part of the type Temblor Formation that are taken to rep- resent time—stratigraphic units of stage magnitude are represented by faunal assemblages from (1) the Wygal Sandstone Member and (2) the Agua and Carneros Sandstone Members. A third stage may be represented by a meager faunal assemblage from the Cymric Shale Member near the base of the Temblor Formation. The relationship of these units, and stratigraphically higher members of the Tem- blor Formation, to the standard Pacific coast mega- invertebrate chronology of Weaver and others 8 MOLLUSCAN BIOSTRATIGRAPHY, PALEONTOLOGY, TYPE TEMBLOR FORMATION, CALIF. (1944) is shown in figure 4. The Miocene mega- invertebrate stages are enclosed in quotation marks here and throughout the text to differentiate them from rock-stratigraphic units bearing the same name and to signify that they have not been formally defined. The Refugian Stage (Schenck and Kleinpell, 1936), usually considered to be a microfaunal stage, was originally defined on both foraminiferal and molluscan evidence. It is not, therefore, enclosed in quotation marks as are the other molluscan stages. Formation Member Molluscan stage Foraértigéieferal Buttonbed Relizian Sandstone “Temblor” Media Shale g Carneros Sandstone Saucesian E A S t Sh 1 LE 3 an 05 a e “Vaqueros” ~ e T: Agua Sandstone E g Santos Shale 7777777777777 Zemorrian (type) Wygal Sandstone Unnamed ' Wi , , , 4 ******** ? ****** Cymric Shale “Lincoln” or Refugian FIGURE 4.-—Molluscan and foraminiferal provincial stage classification of the type Temblor Formation. Formational nomenclature is after Dibblee (1972). Molluscan control is indicated by the vertical bars; foraminiferal stage boundaries are from Stinemeyer, Beck, Ortalda, Espen- scheid, Bainton, and O’Keefe (1959). This report is concerned primarily with the fauna of the Wygal Sandstone Member and its bear- ing on the marine Oligocene of the provincial chronology. Heretofore this poorly known fauna had been assigned to the lower Miocene “Vaqueros Stage” of Weaver and others (1944), although the lack of “Vaqueros” index species had been noted by at least a few workers (Woodring and others, 1940). It can now be shown that the Wygal Sand- stone Member, or so-called Phacoides sand, is older than the “Vaqueros Stage” and that it represents an Oligocene interval in the California provincial molluscan sequence that has never been adequately defined. Secondary objectives are (1) delineation of the boundaries of the overlying “Vaqueros Stage” in the Temblor Range and (2) discussion of mollus- can evidence of the provisional reassignment of the lowermost part of the basal Shale, the Cymric Shale Member, to the Refugian Stage. CYMRIC SHALE MEMBER The Cymric Shale Member of the type Temblor Formation (Dibblee, 1972) has been assigned to the “Vaqueros Stage” of the Pacific coast mega— invertebrate chronology (Weaver and others, 1944. as “Barren” shale member of the Temblor Forma- tion). This assignment, together with the subse- quent inference (Kleinpell and Weaver, 1963, fig. 5) that this unit falls within the biozone of Turri- tella inezana s. 1., was based on the assumption that the base of the Zemorrian Stage of the provincial microfaunal sequence is coincident with the lower boundary of the “Vaqueros Stage” (see also Corey, 1954). So far as is known, however, there are no authenticated occurrences of molluscan assemblages referable to the “Vaqueros Stage” found together with Foraminifera referable to the lower part of the Zemorrian Stage. The known range of the Va- queros fauna of Loel and Corey (1932) [= “Va- queros Stage”] corresponds to the upper part of the Saucesian Stage (Addicott, 1967a, p. 09) . Moreover, the subsequent analysis of the fauna of the Wygal Sandstone Member, which overlies the Cymric Shale Member, indicates that it represents a pre- “Vaqueros Stage” time-stratigraphic unit in the California provincial molluscan sequence. Further revision of the age of this member in terms of the provincial sequences is suggested by a ' small collection of mollusks from 10 to 30 feet below the base of the Wygal Sandstone Member in an unnamed creek between Zemorra Creek and Salt Creek (California Univ., Riverside, Ice. 1106). This collection includes Bruclarkia columbitma (Ander— son and Martin) (pl. 9, figs. 22, 23), a well-defined gastropod that is restricted to the upper part of the Refugian Stage in California (Schenck and Kleinpell, 1936) and to the middle and upper part of the Refugian Stage [= Acila shumardi Zone and “Lincoln Stage”] of western Washington (Durham, 1944). B. columbiana was previously reported from this locality as Olequahia lorenzana (Wagner and Schilling) by Addicott (1970b), who based his identification on one by Alex Clark (unpub. rept., 1931, On file in the Geo]. Dept., California Univ., Riverside), but Clark indicated that his two speci- mens might possibly be confused with Bruclarkia columbiana (Anderson and Martin). His unfigured specimen is an internal mold that is generically indeterminate; it seems to have a strongly noded penultimate whorl and could conceivably represent Olequahia. His other specimen, however, clearly is B. columbiana (pl. 9, figs. 22, 23). The implications of this suggested stratigraphic revision are far reaching. Beds containing Bru- clar/cia, columbioma can be confidently traced into the lowermost part of Kleinpell’s (1938) type Ze- morrian section, which is Situated slightly more PROVINCIAL AGE AND CORRELATION 0F CYMRIC SHALE MEMBER 9 than a mile to the northeast (see Elliott and others, 1968, p. 111). The occurrence of this significant index species suggests that part of the type Temblor Formation must also be of Refugian age and that the two stages, as originally defined, are in part coeval. It should be noted that the lower boundary of the Zemorrian Stage was originally placed at the base of the Cymric Shale Member [lower Tem- blor shale member] by Kleinpell (1938). So far as can be determined, this indicated revi- sion does not necessarily contravene foraminiferal control used by Kleinpell (1938, p. 40—45, 54, 105— 108) in defining his Zemorrian Stage. His strati- graphically lowest sample from Carneros Creek (loc. 1a, 10 ft above the base of the member) con- tains only the long-ranging Eocene to Miocene species Plectofrondicularia vaugham’. The only other assemblage from the Cymric Shale Member listed by Kleinpell (1938, 100. CM70) may have been obtained from beds stratigraphically higher than those in which Bruclarkia occurs. Kleinpell’s sample was collected from a position about 50 feet above the base of the member in nearby Chico Martinez Creek (the thickness of the Cymric Shale Member in this area is variously reported as about 35 ft (Stinemeyer and others, 1959) and 75 ft (Kleinpell, 1938)). This critical assemblage that Kleinpell (1938, p. 107, CM70) cited as indicative of the same “zone” as the diverse and stratigraphically diag- nostic lower Zemorrian assemblages of the overly- ing lower part of the Santos Shale Member, for example, contains only seven taxa. Of these, only Buliminella curta seems to be restricted to post- Refugian strata (R. L. Pierce, oral commun., May 1971). Epistomina ramonensis, the zonal index of Kleinpell and Weaver’s (1963, p. 41) lower subzone of the lower part of the Zemorrian Stage, Which presumably is represented by the Cymric Shale Member, has never been recorded from the type Zemorrian. Furthermore, representatives of the stratigraphically diagnostic late Paleogene and early Neogene Uvigem'na, jacksonensis [U. cocoaensis]— Siphogenerina transversa lineage of Lamb (1964) have never been recorded from the Cymric Shale Member or from this lowest subzone (Kleinpell and Weaver, 1963; Lamb, 1964). The Cymric, in fact, represents a critical gap between the highest strati- graphic occurrence of U. tumeyensis Lamb or U. vicksburgensis Cushman and Ellisor of Kleinpell and Weaver (1963) and the lowest stratigraphic occurrence of the presumed antecedent species S. nodifem Cushman and Kleinpell. The highest strati— graphic occurrence of U. tumeyensz's is in strata assigned to the Refugian Stage by Kleinpell and Weaver (1963), whose assignment is based in part on molluscan evidence; the lowest stratigraphic occurrence of S. nodifem is in the basal part of the Santos Shale Member in the lower part of the type Zemorrian section. Accordingly, the Cymric Shale Member, in terms of the Uvigerim-Siphogenerina sequence of Lamb (1964), represents a hiatus be- tween the Refugian and the Zemorrian Stages. The possibility that the lower boundary of the type section of the Zemorrian Stage should be revised and placed at a higher stratigraphic posi- tion—at or near the top of the Cymric Shale Member—is further suggested by analysis of fish- scale assemblages from the lower part of the Tem— blor Formation by R. L. Pierce. Fish scales from the Cymric Shale Member near the base of the lower part of the Zemorrian Stage are considered by Pierce to be “very characteristic of the Refugian Stage of California,” whereas those from the lower part of the Santos Shale Member—near the t0p of lower part of the Zemorrian Stage—are much differ- ent and are considered typically Zemorrian (R. L. Pierce, written commun., Sept. 3, 1971). In summary, the scant molluscan evidence sug- gests that the basal shale of the Temblor Formation may be, at least in part, referable to the earlier defined Refugian Stage of Schenck and Kleinpell (1936) , a stage defined on both molluscan and foram- iniferal data. This correlation is supported by studies of fish scales from the lower part of the Temblor Formation of this area and is not neces- sarily contravened by published data on foramini- fers from exposures of the Temblor Formation of this area. The suggested upward adjustment of the lower boundary of the type Zemorrian Stage would be in keeping with the principle of priority inas- much as the Refugian Stage (Schenck and Klein- pell, 1936) was proposed before the Zemorrian Stage was defined (Kleinpell, 1938), although the name Zemorrian Stage had been introduced a few years earlier (Kleinpell, 1934). Part of the Cymric Shale Member may thus be correlative with Wagner and Schilling’s (1923) faunal assemblages from the upper part of their San Emigdio Formation and the lower part of their Pleito Formation along the southern margin of the San Joaquin basin. These units contain both mol- lusks and benthonic foraminifers referable to the Refugian Stage (DeLi'se, 1967; Kleinpell and Weaver, 1963). A somewhat larger assemblage of Refugian mollusks has been reported from sand- stone in a landslide block about 10 to 15 miles to the southeast of Zemorra Creek near Crocker Flat (Simonson and Kreuger, 1942). Refugian index 10 MOLLUSCAN BIOSTRATIGRAPHY, PALEONTOLOGY, TYPE TE‘MBLOR FORMATION, CALIF. species in this assemblage include Olequahz'a loren- zana, (Wagner and Schilling), Tmm’tella lorenzana (Wagner and Schilling), and Epitonium condom Dal]. Further study of this displaced strata] se- quence is needed because the Refugian mollusks were said to be associated with foraminifers of Zemorrian age. Exposures of the Refugian Wagon- wheel Formation of Smith (1956) in the southern part of the Diablo Range, about 30 miles to the northwest, contain an assemblage of deep-water pelecypods, principally Lucina aff. L. diegoensis Dickerson and Thyasim afi'. T. disjuncta, that does not permit fauna] correlation with shallow-water Refugian assemblages in the southern part of the San Joaquin basin. WYGAL SANDSTONE MEMBER Indirect evidence that the fauna of the Wygal Sandstone Member is of late Oligocene, pre- “Vaqueros” (early Miocene) age is that species re- stricted to the “Vaqueros Stage” occur only in stratigraphically higher sandstones in the type Tem- blor Formation. The shallow, inner sublittoral bathymetric environment indicated by the molluscan assemblages of the Wygal Sandstone Member is similar to that of the Vaqueros Formation, as pre- viously noted by Woodring, Stewart, and Richards (1940). And, as might be expected, the generic composition of the Wygal molluscan fauna is not markedly different from that of the Vaqueros For- mation of Loel and Corey (1932) 1. Accordingly, one would expect to find some of the ubiquitous shallow- water index species of the “Vaqueros Stage” in the Wygal fauna, if it were deposited during the “Va- queros age,” particularly in view of the moderately large, diverse fauna of 50 shallow-water molluscan taxa recorded in table 2. Inferred phylogenetic relationships of two strati- graphically important Oligocene to middle Miocene molluscan genera, Bruclarkia and Vertipecten, further strengthen the contention that the fauna] assemblages of the Wygal Sandstone Member are older than the “Vaqueros.” Bruclarkia is repre- sented by B. seattlensis Durham, a species that seems to be a lineal antecedent of B. barkerriana forma santacmzrma (Arnold), an early and middle Miocene species restricted to the “Vaqueros” and “Temblor Stages” (table 3). B. seattlensis occurs ILoel and Corey’s (1932) Vaqueros Formation is not strictly a litho- stratigraphic unit; rather, it was defined and recognized on a fauna] basis and, therefore, closely fulfills the criteria for a time-stratigraphic unit (Kleinpell and Weaver, 1963; Addicott, 1970a). Accordingly, it is here referred to us "Vaqueros Stage." Quotation marks are used for this, and other stages, to indicate that they have not been formally defined and to distinguish them from rock—stratigraphic units bearing the same name. stratigraphically above the upper Refugian index species B. columbiana (Anderson and Martin), from which it may have evolved both here and in western Washington (Durham, 1944, p. 116). Likewise, the pectinid Vertipecten alexclarki n. sp. from the Wygal Sandstone Member seems to be the lineal antecedent of V. pern’m’ (Arnold), a species that first occurs in the “Vaqueros Stage” and that has its lowest stratigraphic occurrence in the Agua Sandstone Member, higher in the Temblor Formation (table 3). The reported occurrence of V. perm’ni in the upper part of the Refugian Stage (Schenck and Kleinpell, 1936, p. 222) has never been documented in the literature. That occurrence may refer to a large pectinid from Eflinger’s (1935) Refugian Gaviota Formation of the Santa Ynez Mountains subsequently figured by Wilson (1954, pl. 18, figs. 1, 2) as Pecten (Chlamys) cf. P. (C.) sespeensis Arnold (R. M. Kleinpell, oral commun., June 1971). Vertipecten is represented in the region of the type section of the Refugian Stage by V. yneziana (Arnold) (Kleinpell and Weaver, 1963). The occurrence of Pecten sanctaecruzensis Arnold at the base of the Wygal Sandstone Member (USGS loc. M4467) is further evidence of a pre-“Vaqueros” position in the provincial mega-invertebrate chron- ology. As indicated by Loel and Corey (1932), this species does not occur in fauna] assemblages refer- able to their Vaqueros Formation [“Vaqueros Stage”] ; they regard its occurrences in the Salinas Valley area and in the Santa Cruz Mountains as of pre-Vaqueros, Oligocene age. Although this classi- fication was disputed by Schenck (1935, p. 521—522) based on his mapping of the type locality of P. sanctaecmzensis in the Vaqueros Sandstone of the Santa Cruz Mountains and by subsequent workers (Brabb, 1964; Burchfiel, 1964), the fact remains that this species has never been found associated with mollusks restricted to Loel and Corey’s (1932) Vaqueros Formation [their “formation” being, as previously indicated, a time-stratigraphic unit that has subsequently become known as the “Vaqueros Stage”]. This fact was recognized initially by Arnold (1908), who emphasized that P. sanctaecruzensis and the associated molluscan assemblage in the Santa Cruz Mountains were more closely related to the fauna of the underlying San Lorenzo Formation than to his overlying Vaqueros Sandstone [again Arnold’s (1908) Vaqueros Sandstone was essentially a time-stratigraphic unit defined on biostratigraphic criteria, and its lower boundary was defined on es- sentially the same criteria as Loel and Corey’s Vaqueros Formation]. Moreover, the stratigraphic association of P. sanctaecruzensis with foraminifers PROVINCIAL AGE AND CORRELATION OF WYGAL SANDSTONE MEMBER 11 TABLE 2.—Ranges of specifically determined mollusks from the Wygal Sandstone Member of the Temblor Formation in terms of the Pacific coast provincial mega-invertebrate chronology [“Stazes" are modified from Weaver and others (1944), Durham (1954). and Corey (1954)] Series ___________________________ Eocene Oligocene Miocene Stage : Oregon and Washington .................. “Tejon” ? “Keasey” “Lincoln” “Blakeley" \?\ California ................................ “Tejon” Un- named “Va- queros” “Margar- itan” “Tem- Refugian blor” Gastropods: Calyptraea diegoana (Conrad) ......................... Crepidula cf. C. uhga’na Dall ______ Neverita thomsonae Hickman ______________ Sirium cf. Si soopulosum (Conrad) __________ Ficus cf. F. modesta (Conrad) __________________ . _______ Calicantharus cf. C. brarmeri (Clark and Arnold) ___ Calioantharus dalli (Clark) ___________________________ Bruclarkia seattlehsis Durham ________________________ Pelecypods: Acila muta Clark Yoldia tenuissima Clark _____________ Anadara submontereyarla (Clark) Crenomytilus? cf. C? armldi (Clark)____ Crenomytilus expansus (Arnold) ...................... Peeler; sanctaeomzensis Arnold _______________________ Vertipecteri alexclarki n. sp ____________________________ Crassostrea eldm’dgei ynezam (Loel and Corey) _____ Here excavata (Carpenter) ____________________________ Lucinoma aeutilineata (Conrad) _______________________ Miltha sarwtaecmcis (Arnold) _________________________ Felam‘ella harfordi (Anderson) _______________ Spisula cf. S. albaria (Conrad) _________________ Spisula n. sp. ? ____________________ Spisula ramonensis Packard __________________________ Spisula cf. S. rushi Wagner and Schilling ___________ Tellirw piercei (Arnold) ............................... TellimL emacerata Conrad Tellirw, tenuili’heata Clark ________________ Tellimz cf. T. townsenderisis Clark -------------------- Tellina? cf. T. varwouverensis Clark and Arnold Macoma arctata (Conrad) .............................. Heteromacoma rostellata (Clark) ______ Amiomtis mathewsom‘ (Gabb) .......... Clementia pertenuis (Gabb) _______ Dosinia whitneyi (Gabb) ,,,,,,,,,,, Pimr cf. P. califorhica (Clark) ........................ Securella cf. S. cryptolirleata (Clark) ......... Pampea ramorumsis (Clark) __________________________ Scaphopod: Dentalium laneensis Hickman ________________________ of early Zemorrian age in both the Temblor Range and the Santa Cruz Mountains (Burchfiel, 1964) is suggestive of a pre—“Vaqueros” position in the provincial mega-invertebrate sequence because “Vaqueros” index species are not known to be as- sociated with foraminiferal assemblages below the upper part of the Zemorrian Stage (p. 8). Unfortunately, no representatives of the strati- graphically important gastropod genus Turritella have been found in the Wygal Sandstone Member. Nor has it been found in the upper part of the Pleito Formation of Wagner and Schilling (1923) of the nearby San Emigdio Mountains, a unit that is regarded by myself and by at least some workers (Kleinpell and Weaver, 1963) as coeval with the lower sandstone member of the type Temblor For- mation. The absence of Turritella from these forma- tions complicates correlation: Turritella lineages (Merriam, 1941) as well as the pectinids are the most important means of correlation and age deter- mination by mega-invertebrates in the Pacific coast Tertiary. The post-Refugian age of the molluscan fauna of the Wygal Sandstone Member is indicated by (1) the absence of shallow-water species that are re- stricted to Refugian strata in other parts of the San Joaquin basin, (2) a significant number of genera and species that are not known to occur in 12 MOLLUSCAN BIOSTRATIGRAPHY, PALEONTOLOGY, TYPE TEMBLOR FORMATION, CALIF. TABLE 3.—Ranges of some species of Bruclarkia and Verti- pecten referred to in text Provincial stage Species “Temblor” “Vaqueros” Unnamed Refugian Bmctarkia barkertana forma santacruzana seattlenst's colnmbtana Vertipecten pew-int — alexclarki uneziana assemblages as old as Refugian in California (tables 4 and 5), and (3) the abrupt appearance of mollus- can genera (post-Refugian genera and species) indicative of a much warmer water environment than occurred during the Refugian (Addicott, 1969, 1970b). Although there are some species in common with the Refugian Stage (“Lincoln Stage” of Oregon and Washington), as might be expected, most of the Refugian species characteristic of the upper part of TABLE 4.——-Mollnscan species from the Wygal Sandstone Member that are not known to occur in strata of Refugian age [Species restricted to formations of late Oligocene age in California are indicated by an asterisk (‘). The other species range upward in strata of “Vaqueros” and (or) younger age] Gastropods : Calicantharus dalli (Clark) cf. C. branneri (Clark and Arnold) Pelecypods: Anadara submontereyana (Clark) Crassostrea eld'ridgei yneziana (Loel and Corey) Crenomytilus expansus (Arnold) Dosim'a whitneyi (Gabb) Felaniella harfordi (Anderson) Macoma arctata (Conrad) Miltha sanctaecrucis (Arnold) *Pecten sanctaecmzensis (Arnold) Spisula n. sp.? cf. S. albam’a (Conrad) cf. S. rushi Wagner and Schilling Tellina piercei (Arnold) emacerata (Conrad) *Vertipecten alexclarki Addicott n. sp. the San Emigdio and lower part of the Pleito For- mations of Wagner and Schilling (1923) from the southern margin of the San Joaquin basin do not occur in the Wygal Sandstone Member, nor do they range upward into the upper part of the Pleito Formation that is, in part, correlative with the Wygal Sandstone Member. Some of the Refugian species recorded by Wagner and Schilling (1923) and DeLise (1967) that do not occur in the Wygal Sandstone Member are: Acrilla dickersoni Durham, Brnclarkta columbiana (Anderson and Martin), Bullia clarki Wagner and Schilling, Epitom’um condoni Dall s. 1., Molopophorus dalli (Anderson and Martin), Olequahia lorenzana (Wagner and Schilling), Pe’rse lincolnensis (Van Winkle), Si- phonalia merriami Wagner and Schilling, Acila shumardi (Dall), and Macrocallista pittsburgensts Dall. There is ample evidence, therefore, that the fauna of the Wygal Sandstone Member of the Tem- blor Formation occupies a position between the Refugian Stage and the “Vaqueros Stage” of the provincial mega-invertebrate chronology. It is like- wise clear that the molluscan fauna of the Refugian Stage is not directly succeeded by faunal assem- blages of the “Turritella inezana zone” (= the “Vaqueros Stage” of later usage (Weaver and others, 1944)) as believed by Schenck and Kleinpell (1936) and Kleinpell and Weaver (1963). There is no formally or informally proposed time- stratigrap‘hic unit in the California provincial mol- luscan sequence to which the Wygal fauna can be assigned. It is of post-Refugian age and is, in part, coeval with the “Blakeley Stage” of western Wash- ington, presumably the lower part, the Echinophoria rex Zone of Durham (1944). Correlation with the western Washington molluscan sequence, however, is based largely on stratigraphic position and foram- iniferal evidence and only indirectly on molluscan data.2 The post-Refugian age of the type sections of the “Blakeley Stage” is well established both on mollus- can (Durham, 1944) and foraminiferal evidence (Rau, 1964; Fullmer, 1965). There is, however, evidence that at least part of the “Blakeley” is coeval with the “Vaqueros Stage” (Kleinpell, 1938, p. 153; 2The difficulty in post-Refugian molluscan correlation between the southern part of California and western Washington is the result of the intensifica- tion of latitudinal taxonomic gradients, which began during the late Oligocene (Addicott, 1970c). This intensification is reflected in the de- cidedly warm-water aspect of the Wygal assemb‘ages as contrasted with the temperate aspect of those of the type area of the “Blakeley Stage." Moreover, the bathymetric environment indicated by the ”Blakeley” fauna. particularly the lower pal-twthe Echinophmia 'rex Zone of Durham (1944) —is relatively deeper than that represented by the Wygal assemblages and further complicates correlation. 493—696 0 — 73 - 2 PROVINCIAL AGE AND CORRELATION OF SANTOS SHALE MEMBER 13 TABLE 5.-—Molluscan genera and subgenera from the Wygal Sandstone Member that are not known to occur in strata of Refu- gian age Gastropod: Tegula, Pelecypods: Egesta Leptopecten? Miltha Olcesia Oudardia Pecten s. s. Securella Vanderhoof, 1942; Kleinpell and Weaver, 1963; Addicott, 1967b). Although this equivalence can be rather convincingly demonstrated by foraminiferal correlation, the latitudinal taxonomic gradient in molluscan faunas (footnote, p. 12) reduces the number of species common to both sequences to such a point that this supposed equivalence may never be proven or disproven. Many, if not most, of the generic lineages used in the time-stratigraphic subdivision of the Oligocene and early Miocene of western Washington (Durham, 1944, p. 115-116) are either incompletely represented or are not pres- ent at all in the warmer water assemblages of the San Joaquin basin. Among these critical lineages are Aforia, Ancistrolepis, Limcassis [Echinophoria of authors], Molopophoms, Bathybembix [Turcicula of authors], and Turritelta. The one direct line of evidence supporting correlation with the “Blakeley Stage” of western Washington is the occurrence of the gastropod Bruclarkz‘a scattlensis Durham in the Wygal assemblages. This species is restricted to the lower part of the “Blakeley Stage” (the Echino- phom’a rex Zone) of western Washington. It seems probable that the fauna of the Wygal Sandstone Member falls within the biozone of Acila gettysbur- gensz's, but this species which ranges from the late Oligocene to middle Miocene and which is so charac- teristic of late Oligocene and early Miocene strata in Oregon and Washington, has been recorded from only two areas in California: a doubtful locality in the Santa Cruz Mountains (Schenck, 1936) and the San Ramon Sandstone near Mount Diablo (Clark, 1918). The San Ramon Sandstone, however, seems to be somewhat younger than the Wygal Sandstone Member and coeval units in western Washington assigned to the Echinophoria rex Zone, according to a reanalysis of its invertebrate fauna by Primmer (1965, p. 52—53), who assigned it to the upper part of the “Blakeley Stage” (the Echinophoria apta zone). In View of the difficulty in correlating late Oligo- cene molluscan faunas of California with those of western Washington, it seems appropriate to estab- 493-696 0 < 73 < 3 lish a separate time-stratigraphic unit in California. The evidence presented in this report is judged ade- quate for recognition of a unit of stage magnitude, even though it is based on a fauna that has an extremely limited stratigraphic range. But because thoroughgoing biostratigraphic study of the coeval shallow-water molluscan fauna from the upper part of the Pleito Formation (Wagner and Schilling, 1923) of the southern margin of the San Joaquin basin has not been carried out, this time-strati- graphic unit must, for the time being, remain un- named. In fact, the dominantly shallow-water se- quence of Refugian through “Vaqueros” molluscan faunas of the San Emigdio Mountains should pro- vide the best standard section for a late Oligocene provincial stage in California. Other molluscan faunas of this age from the California Coast Ranges are known to occur in the tuf‘f member of the Kirker Formation of Primmer (1965) , in the lower part of the Vaqueros Formation of Burchfiel (1964) and of Brabb (1964), and in the Santa Lucia Range (Edwards, 1940). Reported occurrences in the Alegria Formation of Dibblee (1950) of the western Santa Ynez Mountains (Kleinpell and Weaver, 1963) were later reassigned to the Refugian Stage (Weaver and Franz, 1967). Molluscan data do not seem to support the claim that the Wygal Sandstone Member [lower Temblor sandstone of Curran (1943)] is coeval with the Quail Canyon Sandstone Member of the Vaqueros Formation (Dibblee, 1973) [formerly the Soda Lake Sandstone Member] of the Cuyama Valley area west of the San Andreas fault (Kleinpell and Weaver, 1963), and by inference, of pre-“Vaqueros” age. Molluscan assemblages from exposures of this unit in the eastern Caliente Range collected by J. G. Vedder include Crassostrea sp. and the “Vaqueros” index species Chlamys hertleim‘ (Loel and Corey) (USGS loc. M3511). Accordingly, the Quail Canyon Sandstone Member of the Vaqueros Formation is younger than the Wygal Sandstone Member and should be correlated with the Agua Sandstone Mem- ber or with stratigraphically higher portions of the type Temblor Formation. SANTOS SHALE MEMBER The lower part of the Santos Shale Member, below the Agua Sandstone Member, includes the boundary between the lower part and the upper part of Klein- pell’s (1938) Zemorrian Stage. Only one molluscan assemblage has been collected from the Santos Shale Member (USGS loc. M3983) less than 200 feet stratigraphically above the Wygal Sandstone Mem- ber on Temblor Creek. This small assemblage con- 14 MOLLUSCAN BIOSTRATIGRAPHY, PALEONTOLOGY, TYPE TEMBLOR FORMATION, CALIF. tains some specifically indeterminate gastropods and pelecypods (Calyptraea, Cancellaria, Bruclarkia?, Leptopecten?, and Spisula sp.) as well as the long ranging pelecypods Lucinoma acutilineata and Macoma arctata. None of these genera or species permits unqualified assignment to either the un- named stage represented by the fauna of the Wygal Sandstone Member or the “Vaqueros Stage” repre- sented by mollusks from the stratigr‘aphically higher Agua and Carneros Sandstone Members. The bound- ary between these two molluscan units is arbitrarily placed in the lower part of the Santos Shale Mem- ber so as to correspond to the boundary between the lower and upper parts of the Zemorrian Stage (Kleinpell, 1938, p. 111). The reason for this place— ment is that, elsewhere in California, mollusk assem— blages of the “Vaqueros Stage” have their lowest stratigraphic occurrence in strata referable to the upper part of the Zemorrian Stage (Addicott, 1967a). AGUA SANDSTONE MEMBER The lowest stratigraphic occurrence of mollusks characteristic of Loel and Corey’s (1932) Vaqueros Formation [2 “Vaqueros Stage” of later usage (p. 10)] in the Temblor Formation of the central and northern parts of the Temblor Range is in the Agua Sandstone Member. The initial recognition of the “Vaqueros” index species Crassostrea vaquero- sensis (Loel and Corey) and Macrochlamis magnolia (Conrad) from unspecified localities between Car- neros Creek and Bitterwater Canyon (Clark and Clark, 1935) has been verified by subsequent col- lections: Macrochlamis magnolia occurs near the mouth of Cedar Canyon (USGS 10c. M2631; Stan- ford Univ. loc. 2682, Heikkila and MacLeod, 1951), and Crassostrea vaquerosensis occurs between Stone Corral Canyon and Media Agua Creek (USGS loc. M4464) several miles to the southeast. The “Va- queros” and “Temblor” species Vertipecten perm‘ni (Arnold) also has been collected from USGS locality M2631. These and other molluscan taxa collected from the Agua Sandstone Member are listed in table 6. These species are also represented in faunal assemblages from the so-called Vaqueros Formation of the San Emigdio Mountains of the southern mar- gin of the San Joaquin basin (Loel and Corey, 1932) and from the J ewett Sand of the southeastern mar- gin of the basin (Loel and Corey, 1932; Addicott, 1970a). The Agua Sandstone Member of the type Temblor Formation is assigned to the upper part of the Ze- morrian Stage of the provincial microfaunal se- TABLE 6.—Mollusks from the Agua Sandstone Member of the Temblor Formation USGS loc. No. Pelecypods M2631 M3982 M4464 Crassostrea afi‘. C. titan (Conrad) _______________ X vaquerosensis (Loel and Corey) ___ __________ X ?Crassostrea (fragments) ____________ X x _____ Heteromacoma rostellata (Clark) __________ x _____ ?Kathe*rinella _____________________________ X _____ Lucinoma acutilineata (Conrad) ____________ X _____ Macrochlamis magnolia (Conrad) ______ X __________ Vertipecten pewini (Arnold) __________ X ? _____ quence (Kleinpell, 1938; Stinemeyer and others, 1959; Kleinpell and Weaver, 1963). PALEOBATHYMETRY Larger invertebrates from the Wygal Sandstone Member are indicative of a relatively shallow-water depositional environment, according to present-day bathymetric ranges of those genera that are still living in the eastern North Pacific. Almost without exception modern depth ranges of these genera define an overlap in the upper part of the inner sublittoral zone—between about 10 and 20 fathoms. A few genera are restricted to, or are more char- acteristic of, environments shallower than 10 fathoms: Crassostrea, Crenomytilus (an extinct genus similar to Mytilus), and Tegula. However, of these, Crenomytilus and Tegala are of uncommon occurrence in the Wygal Sandstone Member, and Crassostrea is usually represented by disarticulated, broken valves possibly indicative of displacement from an initially very shallow water site of deposi- tion into somewhat deeper water. Genera of deeper water aspect—Acila, Lucinoma, and Miltha—are generally found living at somewhat greater depths than 10 to 20 fathoms in warm-water latitudes along the Pacific coast. Species of Acila are recorded from as shallow as 10 to 15 fathoms (Smith and Gordon, 1948; Schenck, 1936), but these records are generally from temperate latitudes. Lucinoma, by far the most abundant mollusk in the Wygal assemblages, occurs in middle sublittoral depths, 28 to 45 fathoms, in tropical latitudes of the eastern Pacific (Keen, 1971), but in the upper reaches of the sublittoral in temperate latitudes, 8 to 10 fathoms, (Smith and Gordon, 1948) . The indi- cated present~day equatorward bathymetric gradient based on these records of Lucinoma annulata, a modern analog of the fossil species L. acutilineata, seems to be an example of equatorial submergence. Although there are a few genera of tropical aspect in the Wygal Sandstone Member (p. 16), the com- position of the fauna is more suggestive of warm- temperate to subtropical conditions. Therefore, a ZOOGEOGRAPHY AND PALEOCLIMATOLOGY 15 generalized interpolation between the 8- to 10- fathom records of modern Lucinoma in temperate latitudes and the 30- to 50-fathom records of it in tropical latitudes would suggest an upper limit of bathymetric occurrence within the postulated 10- to 20-fathom depth range of the Wygal faunal assem- blages. Keen (1971) recorded Miltha, as occurring in water depths of about 30 fathoms or greater in the northern part of the Tropical or Panamic mol- luscan province of Mexico. Significantly shallower modern occurrences are suggested, however, by an intertidal record of this genus near the tip of Baja California, Mexico (Pilsbry and Lowe, 1933). The Wygal faunal assemblages bear some simi- larity to late Pleistocene molluscan communities of the Californian Province (Valentine, 1961; Valen- tine and Mallory, 1965). Comparisons are difficult, however, because a number of the Pleistocene genera do not occur in pre-Neogene assemblages of the Pacific coast area and because these Pleistocene assemblages seem to be of cooler water aspect than those of the Wygal Sandstone Member. Neverthe- less, many of the late Oligocene species are repre- sented by analogs in the Dosim’a ponderosa element of Valentine’s (1961) shallow inner sublittoral com- munity, a sandy substrate association representative of depths between 5 and 15 fathoms. Bathymetric inferences based on benthonic foram- iniferal assemblages from shales underlying and overlying the Wygal Sandstone Member are sug- gestive of significantly greater depths. Foraminifers from the underlying Cymric Shale Member listed by Kleinpell (1938, locs. 1a and CM70), for example, are indicative of the middle to lower part of the bathyal zone (about 3,000 to 6,000 feet), according to the classification of San Joaquin basin forami- niferal biofacies by Bandy and Arnal (1969, p. 787— 791). These inferences are based on the deep-water species Cyclamimz incisa, Gyroidina soldam'i, and Plectofrondiculam’a vaughani. Additional forami- niferal assemblages from the Cymric Shale Member studied by R. L. Pierce (written commun., April 1969) are also indicative of middle bathyal or deeper depositional environment, according to the classification of Bandy and Arnal. An assemblage from the basal part of the Santos Shale Member di- rectly overlying the Wygal Sandstone Member in Chico Martinez Creek also studied by Pierce (writ- ten commun., April 1969) includes species listed by Bandy and Arnal (1969) as characteristic of their lower bathyal and abyssal biofacies. Included are Siphogenerimz nodifem, Cibicz'des floridrmus, Gyroi- dim soldamli, and Plectofrondicularia cf. P. cali- fornica. When these depths are contrasted with the indi- cated inner sublittoral depositional environment of the intervening Wygal Sandstone Member, late Oligocene vertical oscillation of the basin floor amounting from 3,000 to as much as 6,000 feet is indicated. Comparable downwarping of the floor of the San Joaquin basin in areas west of Bakersfield and southeast of Taft during the Zemorrian Stage has been inferred by Bandy and Arnal (1969, fig. 5) based on analysis of isopach data and inferred foraminiferal biofacies. However, their inferences of little or no differential vertical movement of the basin floor and continuous deep-water facies in the vicinity of the type section of the Zemorrian Stage at this time (Bandy and Arnal, 1969, fig. 5) are contrary to the evidence indicated by the larger invertebrate fauna of the Wygal Sandstone Mem- ber. Rather, it seems that an extensive oscillatory vertical movement of the basin floor must have occurred during the Oligocene in the vicinity of what is now the northeastern margin of the Temblor Range. The strong erosional unconformity at the base of the Wygal Sandstone Member and the shal- low inner sublittoral depths indicated by molluscan assemblages of this member contrast sharply with the middle to lower bathyal foraminiferal biofacies of the underlying Cymric Shale Member and the overlying Santos Shale Member. The possibility that the Wygal assemblages were displaced downslope and redeposited in bathyal depths indicated by the foraminiferal assemblages seems remote. The extremely abundant and rela- tively well-preserved mollusk shells, for example, generally are not fragmented and disarticulated, as are the scattered shallow-water pectinids and broken valves of Tivela? found in the stratigraphically higher Carneros Sandstone Member, which also occurs between deep-water foraminiferal assem- blages of the Santos and Media Shale Members. ZOOGEOGRAPHY AND PALEOCLIMATOLOGY The molluscan fauna of the Wygal Sandstone Member of the Temblor Formation marks the be— ginning of a late Oligocene to late Miocene period during which molluscan faunas of the San Joaquin basin were characterized by a sizeable element of warm-water genera. These shallow-water genera are termed extralimital in the sense that their modern limits of geographic distribution along the Pacific coast do not reach as far north as the lati- tude of the fossiliferous exposures (lat 35.5° N.). Extralimital, warm-water taxa make up 41 percent , of the still-living molluscan genera represented in the Wygal faunal assemblages. They are almost 16 MOLLUSCAN BIOSTRATIGRAPHY, PALEONTOLOGY, TYPE TEMBLOR FORMATION, CALIF. equally divided, in terms of the northernmost present-day geographic limits of range between the Panamic, Surian, and Californian molluscan prov- inces of Valentine (1966). The shallow-water ma- rine climates of these provinces are characterized, respectively, as inner tropical, outer tropical, and warm temperate (Hall, 1964). During the late Oligocene, the central California area seems to have been a northern outpost for almost all these genera. Only a scattering of warm- water genera reached as far north as western Wash- ington during the late Oligocene where a very small percentage (11 percent) of the still-living genera of the presumably coeval Echinophoria rem Zone of Durham (1944) are considered to be of warm-water aspect (Addicott, 1970c). Warm-water genera represented in the Wygal assemblages and their northernmost limits of dis- tribution in modern shallow-water molluscan prov- inces of the eastern North Pacific are as follows: Panamic—Ficus, Miltha, Clementia; Surian— Natica s. s., A-nadara s. s., Crassostrea, Dosim'a; Californian—Neven'ta, Kelletia, Here, Tellinella, Amiantis. In this late Oligocene fauna a substantial number of the extinct genera, or of genera that are no longer living along the margin of the eastern North Pacific, may also be of warm-water aspect: Siphonalia?, C’r‘enomytilus, Vertipecten, and Olcesia. This judgment is based both on the modern distri- bution of closely related molluscan genera and on the geographic distribution of these extinct genera in warm-water Tertiary assemblages along the Pacific coast. Further evidence of very warm, shallow-water conditions during deposition of the Wygal Sand- stone Member is furnished by the occurrence of the hermatypic coral Fam’tes in collections from USGS locality M3280 near Media Agua Creek. Modern occurrences of this inner sublittoral genus are re- stricted to the tropical latitudes of the Indo-Pacific region and the Red Sea. According to J. Wyatt Durham (oral commun., Jan. 1971), who identified these specimens, and Durham and Allison (1960, p. 72), Favites has not previously been reported from the Cenozoic of the eastern Pacific margin. The zoogeographic relationship of the warm—water molluscan genera in the Wygal fauna to the tem- perate-water middle Oligocene fauna of the San Joaquin basin and to subsequent warm-water Mio- cene faunas of this area is depicted in figure 5. The present study has refined the systematic classifica- tion of the Wygal fauna to the extent that a more accurate zoogeographic analysis than was presented in a recent report (Addicott, 1970a, fig. 4) can be Series i Pliocene 131$? Eocene 7 . cene Oligocene Miocene Provincial stage H‘Do- ,~.. _7 k. 7-- .. . _ E gig; 353:1?" 53;st Refugian 8113;? E “Vaqueros” “Temblor” figfl .Wygal Sandstone Member 50 A 1 7 I— \ L|ZJ \\ 4 5 8 U 40 — \ 6 m \ LIJ \ a \ \ \\ Californian 9 E 30 — \\ (warm (‘ \ \\ temperate) 10 s \ \ 2 Z 20 — \ \ . n g \ \ Surlan 12 \\ \ (subtropical) 13 E \\ \\ 10 M \ \ < \ 14 § \ Panamic \ 4 Lu \ (tropical) o I I J 50 40 30 20 10 0 TIME. IN MILLIONS OF YEARS FIGURE 5.—Zoogeographic profiles for selected Tertiary faunal assemblages from the San Joaquin basin, Cali- fornia (Addicott, 1970a). Cumulative percentages express the modern warm-water zoogeographic provinces in which northern end points of range of extant genera occur: Panamic, 6° S.—23° N.; Surian, 23° N.—28° N.; Californian, 28° N.—34.5° N. The sequence of provincial molluscan stages is modified from Weaver and others (1944) and Corey (1954). Numbers refer to faunal assemblages as follows: (1) Domengine Formation, southern Diablo Range; (2) San Emigdio and Pleito Formations of Wag- ner and Schilling (1923), San Emigdio Mountains; (4) Wygal Sandstone Member of the Temblor Formation, Temblor Range; (5) Jewett Sand, Kern River area; (6) lower part of the Olcese Sand, Kern River area; (7) upper part of the Olcese Sand and lower part of the Round Mountain Silt, Kern River area; (8) Santa Mar- garita Formation, Tejon Hills; (9) Santa Margarita For— mation, Coalinga area; (10-12) Etchegoin Formation, Coalinga area; (13) Etchegoin Formation, Kettleman Hills; (14) San Joaquin Formation, Kettleman Hills. made. Accordingly, cumulative percentages of late Oligocene warm-water genera (Panamic and Sur- ian) are somewhat higher than previously calcu- lated and are almost the same as those for the fauna of the overlying “Vaqueros Stage.” However, zoogeographic relationships of these two successive faunal units—late Oligocene and early Miocene—- are still clouded by the fact that the late Oligocene fauna is relatively poorly known, having only about half as many taxa as the early Miocene fauna. The disparity in size seems to be in the lack of a diverse gastropod element in the Wygal assemblages. SYSTEMATIC DESCRIPTIONS Only those references accompanied by illustra— tions of specimens are included in the following synonymies. Reports that list some of the more SYSTEMATIC DESCRIPTIONS 17 common molluscan taxa from the Wygal Sandstone Member of the Temblor Formation are reviewed in an annotated list of literature (p. 3). The plates include a few illustrations of Oligocene and Miocene mollusks from other stratal units for comparative purposes. Entries in the synonymies are as origi- nally used by the authors; a few contain lapses. Fifty molluscan taxa are treated systematically; all but one of these are figured. One new species and a new tellinid subgenus are described: Vertipecten alexclarki n. sp. and Olcesia n. subgen. Five addi- tional taxa—Tegula n. sp., Natica (Natica) n. sp., Polinices n. sp., Spisula n. sp.?, and Amiantis n. sp.—seem to be undescribed but are represented by material that is too incomplete or too poorly pre- served to permit formal description. The systematic arrangement of gastropods is after Keen (1963); of pelecypods, after McCormick and Moore (1969). Other larger marine invertebrates from the Wygal Sandstone Member include the hermatypic coral Fam'tes, and undetermined solitary, branching coral, and an encrusting bryozoan Antropora? sp. Abbreviations used in tabulating the occurrences of molluscan taxa in the type Temblor Formation are: USGS _____ U.S. Geological Survey, Washington, DC, locality register (locality numbers have no letter prefix). USGS M -__U.S. Geological Survey, Menlo Park, Ca1if., Cenozoic locality register. UCMP _____ California University Museum of Paleon- tology, Berkeley, Calif. UCR _______ California University, Riverside, Calif. Other fossil depositories are: U0 ________ Oregon University, Eugene, Oreg. SU ________ Stanford University, Stanford, Calif. CAS _______ California Academy of Science, San Fran- cisco, Calif. ANSP _____ Academy of Natural Sciences, Philadelphia, Pa. Class GASTROPODA Order ARCHAEOGASTROPODA Family TROCHIDAE Genus TEGULA Lesson, 1835 Tegula n. sp. Plate 8, figures 2, 4 Two incompletely preserved specimens of a mod- erately small, smooth-shelled trochid represent the shallow-water genus Tegula; a fragment in USGS collection M4466 may also be this trochid. The two specimens have somewhat different whorl profiles. The better preserved specimen (pl. 8, fig. 4) has an almost flat whorl profile; the more poorly preserved specimen (pl. 8, fig. 2) has a more strongly convex whorl profile and a faint subsutural tabulation on the body whorl. Because the inner margin of the aperture on one of the specimens shows a fairly strong tooth, the specimen can be assigned to Tegula. This Tegula was first recorded from the Wygal Sandstone Member of the Temblor Formation by Alex Clark (unpub. data, 1931 and 1932, deposited at California University, Riverside), who first recog- nized that it was an undescribed species. Clark’s localities (EC—89 [= UCR 1235] and EC—125A) and USGS locality M3578 are from exposures about % to 11/, miles southeast of Zemorr‘a Creek (fig. 3). One of his specimens is figured on plate 8 (fig. 4). Tegula. n. sp. is similar to the smooth-shelled middle Miocene species T. laem's Addicott (1970a, p. 41—42, pl. 1, figs. 2, 3, 5—7) from the Kern River area of the eastern margin of the San Joaquin basin. The Miocene species differs, however, in having strongly convex whorls and a narrow, but well- defined, subsutural tabulation. This is the oldest record of the genus Tegula. The oldest previously known record was in the Miocene (Knight and others, 1960, p. 254) ; it may have been based on the early Miocene species T. malibuensz's Loel and Corey (1932) from the Vaqueros Forma- tion, of California. Localities—USGS M3578, M4466 ?, UCR 1235. Order MESOGASTROPODA Family CERITHllDAE? Cerithiid? Plate 8, figure 5 This moderately small, very poorly preserved conispiral gastropod which has a flat whorl profile seems to be a cerithiid. The body whorl is broken, and there are no traces of sculpture on the whorls of the spire. Locality.—USGS M3578. Family CALYPTRAEIDAE Genus CALYPTRAEA Lamarck, 1799 Cnlyptrnea diegoana (Conrad) Plate 8, figures 8, 10, 17, 18 Trochita diegoana Conrad, U.S. 33rd Cong, lst sess., H. Ex. Doc. 129, p. 17. Trachita diegoana Conrad. Conrad, U.S. 33rd Cong, 2d sess., S. Ex. Doc. 78, p. 327, pl. 5, fig. 42. Galerus excentm‘cus Gabb, California Geol. Survey, Paleontology, v. 1, sec. 4, p. 136, pl. 20, fig. 95, pl. 29, figs. 232, 232a, Galerus excentricus Gabb. Arnold, U.S. Geol. Survey Bull. 321, pl. 10, figs. 3a, 3b. Galerus excentricus Gabb. Arnold, U.S. Geol. Survey Bull. 396, p. 112, pl. 4, fig. 8 [Imprint 1909.] Galerus excentricus Gabb. Arnold and Anderson, U.S. Geol. Survey Bull. 398, pl. 26, fig. 8. Calyptraea washingtonensis Weaver, Washington Univ. [Seattle] Pubs. Geology, v. 1, p. 44, pl. 3, fig. 44. 1855. 1857. 1864. 1907. 1910. 1910. 1916. 18 MOLLUSCAN BIOSTRATIGRAPHY, PALEONTOLOGY, TYPE TEMBLOR FORMATION, CALIF. 1927. Calyptraea diegoaxna (Conrad). Stewart, Acad. Nat. Sci. Philadelphia Proc., v. 78, p. 340—341, pl. 27, fig. 15. Calyptraea washingtonensis Weaver, California Univ. Pubs. Geo]. Sci. Bull., V. 23, no. 3, p. 137, pl. 14, fig. 25. Calyptraea diegoana (Conrad). Turner, Geol. Soc. America Spec. Paper 10, p. 89-90, pl. 20, figs. 1, 2. Calyptraea diegoana (Conrad). Weaver, Washington Univ. [Seattle] Pubs. Geology, v. 5, p. 351—352, pl. 71, figs. 16, 20, pl. 103, fig. 3. Calyptraea washingtonensis Weaver. Weaver, Wash— ington Univ. [Seattle] Pubs. Geology, v. 5, p. 352— 353, pl. 71, figs. 19, 22. Calyptraea diegoana (Conrad). Weaver and Kleinpell, California Univ. Pubs. Geol. Sci. Bull., v. 43, p. 186, pl. 24, fig. 7. Calyptraea diegoana (Conrad). Hickman, Oregon Univ. Mus. Nat. History Bull. 16, p. 79, 82, pl. 11, figs. 7, 8. Type.—USNM 1856 (designate-d by Keen and Bentson, 1944). Type locality—San Diego, Calif. Eocene. Calyptmea diegoana is a variable species char- acterized by a high, pointed spire and an excentric apex. The lectotype (Keen and Bentson, 1944, p. 204) is an immature specimen almost as high as it is broad. Specimens from the Wygal Sandstone Member of the Temblor Formation have a straight- sided to somewhat convex profile and little, if any, constriction of the shell wall at the suture. In lateral aspect, the narrowly excentric side of the shell is slightly convex whereas the broadly excentric side is relatively much flatter and even somewhat con- cave on certain specimens. Stewart (1927) synonymized several names ap- plied to this early Tertiary taxon, which is wide- spread in and characteristic of strata of Eocene age but which also occurs in Oligocene strata of Oregon and Washington (Hickman, 1969). The highest stratigraphic occurrence of this species is in the upper Oligocene, lower part of the “Blakeley Stage” of western Washington (Tegland, 1933; Durham, 1944), a ‘unit with which the fauna of the Wygal Sandstone Member is believed to be coeval. Calyptraea mammilaris vancouverensis Clark and Arnold (1923, p. 167, pl. 36, figs. 3a, 3b) from the Sooke Formation of southwestern Vancouver Island, British Columbia, differs from C. diegoana by its very thick, evenly and more highly conical shell and more or less centrally located apex. The California early Miocene species C. coreyz’ Addicott (1970a, p. 60—61, pl. 4, figs. 1, 4, 18) that is also character- ized by an eccentric apex and a moderately high spire can be distinguished from C. diegoana by its shouldered whorl profile. 1933. 1938. 1942. 1942. 1963. 1969. Localities—USGS M3280, M3281, M4471. Crepidula cf. C. ungana Dull Plate 8, figures 6, 7, 9, 11, 15, 16 Specimens of thin-shelled Crepidula, in the col- lections are similar to the Alaska Oligocene species C. ungana Dall (1904, p. 118, pl. 10, figs. 8, 9) from the Aleutian Islands. They have been compared with the type specimens (USNM 164918) and with a large suite of topotypes from Unga Island (USGS loc. 2103). The California specimens differ slightly from the types, an incomplete chain consisting of three specimens, in having a relatively finer, less swollen beak. The orientation of the beaks is simi- lar—more or less parallel to the plane of the aper- ture. On some of the topotypes of C. ungana, how- ever, the beak tends to point upwards. A comparable upward orientation has been found on only one of the specimens from the basal sandstone member of the Temblor Formation. As noted by Hickman (1969, p. 82), specimens preserved in colonial form are more highly arched and narrower than those that occur as individual specimens. The configuration and size of the internal septum of Crepidula cf. C. ungana is unknown. Viewed aperturally, the insertion of the septum is near the middle of the left side of the aperture as on C. ungana, which has an open S-shaped septal margin. The point of insertion on the opposite side of C. cf. C. ungoma has not been observed. Crepidula cf. 0. ungoma can be distinguished from the early Miocene species C. sookensis Clark and Arnold (1923) from Vancouver Island, British Columbia, and the northwestern Olympic Peninsula (Durham, 1944) by its much thinner and narrower shell and more delicate, less swollen beak. The septal configuration of specimens of C. sookensis from the Sooke Formation (USGS loc. M4060) is similar to that of the variable early Miocene to Holocene C. princeps Conrad. Occurrences of Crepidula ungana in the conter- minous Pacific Coast States range from the middle Oligocene “Lincoln Stage” (Hickman, 1969) to the late Oligocene lower part of the “Blakeley Stage” (Tegland, 1933; Durham, 1944). Localities—USGS M3280, M3281, M3979. Family NATlClDAE Poorly preserved naticids on which it has not been possible to expose the umbilical area occur in the collections from USGS localities M3280, M3578, M3772, M3978, M3979, M4466, and M4468. The size and whorl profiles of these suggest either Natica or Polim'ces. SYSTEMATIC DESCRIPTIONS 19 Genus NATICA Scopoli, 1777 Subgenus NATICA s. s. Naticn (Naticu) n. sp. Plate 8, figures 12, 13 At least five specimens of a moderately large, heavy-shelled Natica from USGS locality M3280 represent an undescribed species. They differ from known middle Tertiary species of Natica. in their relatively large adult size, widely open umbilicus, and heavy funicular rib. These specimens are too poorly preserved to permit formal description as a new species: most of the surface of the shells is missing, and the spire is incompletely preserved. There are distinct similarities to the middle Ter- tiary species Natica teglandae Hanna and Hertlein (1938, p. 108) and N. vokesi Addicott (1966, p. 638—639, pl. 77, figs. 2—5), but both of these small species have shouldered body whorls and only weakly or very weakly developed funicular ribs. The middle Miocene species N. posuncula Hanna and Hertlein (1938, p. 107—108, pl. 21, fig. 6) from the Kern River area on the southeastern margin of the San Joaquin basin has a similar profile but a narrow umbilical area that is filled by a relatively larger and broader funicular rib. Natica s. s. is a warm-water genus restricted to the modern Surian and Panamic molluscan prov- inces of Valentine (1966) of the Pacific coast. Its occurrence in the fauna from the Wygal Sandstone Member of the Temblor Formation is suggestive of much warmer marine temperatures during the late Oligocene than occur along this part of the Pacific coast today. Locality—USGS M3280. Genus NEVERITA Rilso, 1826 Subgenus GLOSSAULAX Pilsbry, 1929 Neverita (Glossaulax) thomsonae Hickman Plate 9, figure 4 1969. Nevem‘ta (Glossaulax) thomsonae, Hickman, Oregon Univ. Mus. Nat. History Bull. 16, p. 84, pl. 11, figs. 20—23. 1972. Neverita thomsonae Hickman. Addicott, Soc. Econ. Paleontologists and Mineralogists, Pacific Sec., Bakersfield, Calif., 1972, Proc., pl. 1, fig. 10. Types.—Holotype, UO 27366; paratypes, UO 26367—26370, 27371, 27372. Type locality.—UO 2567, northwestern Oregon. Pittsburg Bluff Formation, Oligocene. As noted by Hickman (1969), this large species is very close to the living Nevem'ta recluziana, a warm-water species that ranges from the west coast of Mexico to southern California (Smith and Gor- don, 1948; Keen, 1971). The extremely large body whorl of N. thomsonae, which has a smooth, flat- tened profile, is the principal morphologic feature used to differentiate the two species. N. thomsonae seems most closely related, morphologically, to the large form of N. reeluziana figured by Pilsbry (1929, pl. 6, fig. 1) and differs from it, as noted by Hickman (1969), in having a relatively broader body whorl. Nevem'ta thomsonue differs from N. andersom' (Clark), a common species in early Miocene to Pliocene molluscan assemblages from California (Addicott, 1970a), in having a flattened, nonshoul- dered profile. The range of this species is from the middle to the upper Oligocene. Locality—USGS M3978. Genus POLINICES Monlfort, 1810 Polinices n. sp.? Plate 9, figure 10 A moderately large Polim'ces which has a widely open umbilical area was collected from USGS 10- calities M3280 and M4472. The heavy parietal callus thins and tapers uniformly toward the base of the aperture. The body whorl does not appear to be shouldered. This species differs from Polim'ces victoriana Clark and Arnold (1923, p. 170, pl. 33, figs. la, 1b, 5a, 5b) and P. canalis Moore (1963, p. 28, pl. 2, figs. 18, 20), Miocene species from the Pacific coast, in having a much larger, widely open umbilical area and a relatively narrow, but equally heavy, parietal callus. Natica n. sp.? Hickman (1969, p. 83—84, pl. 10, figs. 15, 16) from the early and middle Oligocene Eugene Formation of Oregon has a much narrower umbilical opening and lacks the heavy parietal- umbilical callus. Localities—USGS M3280, M4472. Genus SINUM Railing, 1798 Sinurn cf. 5. scopulosum (Conrad) Plate 9, figures 16, 17 Two internal molds from USGS locality M3578 seem to represent Sinum scopulosum (Conrad, 1849, p. 727, pl. 19, figs. 6, 6a), an upper Oligocene to Holocene species. The fine spiral sculpture is faintly preserved on both specimens. Sinum scopulosum is distinguished from the Eocene and Oligocene species, S. obliquum (Gabb, 1864, p. 109, pl. 21, fig. 112), by its larger adult size and less flattened shell. The lowest stratigraphic record of this species is from the upper Oligocene Echinophom’a rex Zone (lower part of the “Blakeley Stage”) of the north- ern Olympic Peninsula, Wash. (Durham, 1944), a unit that is believed to be coeval with the fauna of the Wygal Sandstone Member. Localities—USGS M3578, M4471 ?. 20 MOLLUSCAN BIOSTRATIGRAPHY, PALEONTOLOGY, TYPE TEMBLOR FORMATION, CALIF. Family FlClDAE Genus FICUS R'éding, 1798 Ficus cf. F. modesta (Conrad) Plate 9, figures 18, 19 Two crushed specimens of a delicately ornamented Ficus are compared with F. modesta (Conrad, 1848, p. 433, fig. 12), an Oligocene and Miocene species from the Pacific coast. This species differs from the middle Oligocene species Ficus (Tropho- sycon) gesteri Wagner and Schilling (1923, p. 258, pl. 49, figs. 1—3) from the nearby San Emigdio Mountains in lacking nodes on the shoulder of the body whorl. There is no evidence of nodes on the body whorl of the larger of the two specimens (pl. 9, fig. 19). Ficus modesta was recorded by Wagner and Schilling (1923, p. 246, as F. pyriformis Gabb) from the upper part of their Pleito Formation in the San Emigdio Mountains, a unit that is believed to be coeval with the Wygal Sandstone Member of the type Temblor Formation. It is also recorded from the San Ramon Sandstone of the Mount Diablo area, California. Ficus modestu ranges from the middle Oligocene “Lincoln Stage” (Hickman, 1969) to the middle Miocene “Temblor Stage” (Schenck and Keen, 1940; Moore, 1963). Localities—USGS M3578, M3978. Order NEOGASTROPODA Family NEPTUNEIDAE Genus CALICANTHARUS Clark, 1938 Calicautharus, an early Tertiary t0 Quaternary genus, is similar to Seariesia, a genus that occurs in temperate and cool-water molluscan provinces along the Pacific coast. Although reassignment of the following two species to Calicantharus consti— tutes the initial recognition of this genus in the late Oligocene of the Pacific coast, the genus is known to occur in the Eocene Markley Sandstone Member of the Kreyenhagen Formation of central California. Calicautharus is here discriminated from Scarlesia on the basis of its collared or clasping suture. Calicnntharus dalli (Clark) Plate 9, figure 24 1918. Scarlesia dalli Clark, California Univ. Pubs., Dept. Geology Bull., v. 11, no. 2, p. 175, pl. 20, figs. 5, 9, 15. Types.——Holotype, UCMP 11214, 11214a. Type locality.—UCMP 2754, branch of San Pablo Creek on Sobrante Ridge, 13/3 miles northeast of California and Nevada Railroad, lat 37°56’54” N., long 122°14’01” W. Altitude 350 feet. San Ramon Sandstone, early Miocene( ?). UCMP 11215; paratypes, Specimens of Caliccmtharus from USGS locali- ties M3280 and M3772 are conspecific with Clark’s (1918) C. dalli from the Mount Diablo area, Contra Costa County, Calif. C. dalli is larger than the type specimens but has the fine, channeled spiral ribbing, the sutural overlap, and the evenly convex whorl profile characteristic of this species. Moreover, the spire appears to bear very fine nodes at the point of greatest whorl convexity. The principal difference from Clark’s (1918) types is the relatively greater size attained by the late Oligocene specimens from the Temblor Range. Calicantharus dalli is quite distinct from Pacific coast Miocene species of this genus. Compared to specimens of C. bromneri (Clark and Arnold, 1923, p. 159, pl. 30, figs. 3a, 3b), an equally large species from the Sooke Formation of southwestern British Columbia (USGS 10c. M2544), it differs in having a broader body whorl, finer spiral sculpture, and virtually no angulation on the body whorl. C. ker- nensis (Anderson and Martin, 1914, p. 78—79, pl. 4, figs. 6a, 6b), a middle Miocene species from the California Coast Ranges, is characterized by a sinu— ous, rather than relatively straight, growth line and much narrower shell. Another middle Miocene spe- cies from central California, C. woodfordi Addicott (1970a, p. 93—94, pl. 10, fig. 6, pl. 11, figs. 17, 18), is of similar size of C. dalli but has very strongly angulated, noded whorls. Localities—USGS M3280, M3772. Calicanlharus cf. C. branneri (Clark and Arnold) Plate 9, figure 13 An incomplete specimen from USGS locality M4471 is identical to Clark and Arnold’s (1923, p. 159, pl. 30, figs. 3a, 3b) species from the Sooke Formation of southwestern Vancouver Island, British Columbia. The body whorl is characterized by rounded spiral ribs that are separated by rather deep interspaces containing a fine intercalary riblet. As on the holotype, the axial folds are irregularly spaced and almost obsolete on the abapertural side of the body whirl. This taxon differs from specimens of Calican- tharus dalli (Clark) from the Wygal Sandstone Member in having high, rounded spiral ribs rather than relatively broad, straplike ribs separated by channeled interspaces. The subsutural collar of this species indicates assignment to Calicantharus rather than Searlesia to which it was originally assigned (Clark and Arnold, 1923). Locality.—USGS M4471. SYSTEMATIC DESCRIPTIONS 21 Genus KELLETIA Fischer, 1884 Kelletia? 51). Plate 9, figures 7, 11, 12 Two incomplete specimens of a moderately large neptuneid may represent the Eocene (Ruth, 1942) to Holocene genus Kelletia. The spiral sculpture of fine, flattened ribs separated by deeply channeled interspaces and the development of nodes on the whorl angulation are similar to late Cenozoic species of Kelletia described from California. The two speci- mens differ considerably in details of spiral and axial ribbing but are tentatively taken to represent a variable species. One specimen (pl. 9, figs. 11, 12) has relatively broad, straplike ribs that bear a medial groove; the other (pl. 9, fig. 7) has much finer spiral ribs, only a few of which bear a medial groove. The axial folds are very prominent on the penultimate whorl of one specimen (pl. 9, figs. 11, 12) but are very weak on the other one (pl. 9, fig. 7). Both specimens have a relatively broad apical angle and a moderately strong angulation near the middle of the whorls. Localities—USGS M3280, M3772. Genus SIPHONALIA Adams, 1863 Siphonalia? sp. Plate 9, figures 5, 6 A small, crushed specimen from USGS locality M3280 may represent Siphonalia. However, the generic assignment of this neptuneid is doubtful. The body whorl is constricted and apparently twisted to form a distinctive anterior canal very much like specimens of this genus figured by Ruth (1942). However, Siphonalia? sp. differs from Oligocene and Miocene species included in Ruth’s monograph in having only one row of very weakly developed nodes on the whorls of the spire. Locality.—USGS M3280. Genus BRUCLARKIA Stewart, 1927 Bruclarkia seattlensis Durham Plate 9, figures 1, 2, 8, 9, 14 1944. Bmclarkia seattlensis Durham, California Univ. Pubs, Geol. Sci. Bull., v. 27, no. 5, p. 173—174, pl. 16, fig. 15. 1972. Brucla’rkia seattlensis Durham. Addicott, Soc. Econ. Paleontologists and Mineralogists, Pacific Sec., Bakersfield, Calif., 1972, Proc., pl. 1, figs. 8, 13. Type.—UCMP 35397. Type locality.——UCMP A1803, behind the Olympic Foundry, Georgetown, Seattle, King County, Wash. Blakeley Formation, Oligocene. Bruclarkia seattlensis is the most abundant gas- tropod in assemblages from the Wygal Sandstone Member. Many of the specimens are heavily en- crusted with the bryozoan Antropora? sp. (O.L. Karklins, written commun., January 1971). Earlier identifications of B. barkem'ana (Cooper) from ex- posures of the Wygal Sandstone Member (the so- called Phacoides reef) in the vicinity of Zemorra Creek (Kleinpell, 1938; Stinemeyer and others, 1959; Foss and Blaisdell, 1968) are almost certainly of this species. Specimens from the Wygal Sandstone Member differ slightly from the type of Bruclarkia seattlen- sis in having a somewhat stronger subsutural collar and in lacking fine nodes on the whorls of the spire. The form and sculptural detail of the body whorl, however, is the same: a faint angulation devoid of nodes and spiral sculpture of fine ribs that alter- nate in strength. Moore (1963, p. 36) showed that both the strength of the subsutural collar and the presence or absence of nodes are highly variable morphologic characters on Miocene species of Bru- clarkia. Variation within B. seattlensis was pre- viously unknown because this species was known only from the holotype. This species very closely resembles the early and middle Miocene species Bruclarkia santacruzam (Arnold, 1908, p. 379—380, pl. 34, fig. 7), a smooth, strongly collared gastropod that has recently been treated as a form of the common middle Miocene species, B. barkem‘ana (Addicott, 1970a). Bru- clarkia seattlensis differs from B. barkem'ana forma santacruztma (pl. 9, fig. 3) in having a consistently much higher spire, a smooth collar that has no nodes, and, usually, a perceptibly angulated body whorl. Yet the morphologic similarities to B. santacruzam are so striking that B. seattlensis is very likely a direct precursor of the Miocene species. Durham (1944, p. 173—174) commented on the close degree of similarity of B. seattlensis to the middle Oligo- cene B. columbiana, and this similarity may imply an ancestral relationship. Other species of Bruclarkz’a of common occurrence in formations of Oligocene age in the southern part of the San Joaquin basin (Wagner and Schilling, 1923, p. 244, 248) can be readily distinguished from B. seattlensis. B. acumz‘nata (Anderson and Martin, 1914, p. 73, pl. 5, figs. 4a, 4b) has strongly noded, angulated whorls and either no subsutural collar or a very weak one. Bruclarkia gravida (Clark, 1918, p. 182—183, pl. 19, figs. 1, 3, 5) is likewise a strongly sculptured species on which there are three equally spaced rows of heavy, rounded nodes, the uppermost of which is born relatively high on the body whorl (pl. 9, figs. 15, 20, 21). The possibility that B. seattlensis is a weakly sculptured form of B. gravida and might 22 MOLLUSCAN BIOSTRATIGRAPHY, PALEONTOLOGY, TYPE TEMBLOR FORMATION, CALIF. bear a relationship similar to that between B. bar- kem’ana forma santacruzana and B. barkem’ana (Addicott, 1970a) seems unlikely in view of the relatively low position of the angulation on the body whorl of specimens of B. seattlensis from the Wygal Sandstone Member. The position of this angulation is closer to that of the Oligocene B. columbiana (Anderson and Martin, 1914, p. 73, pl. 5, figs. 6a, 6b), an index species for the Refugian Stage. B. columbianu differs, however, in usually having a low spire that has a flat profile, a weak collar, and a long and concave subsutural segment on the body whorl terminating in a strongly noded angulation. Some of the smooth forms of the Early and Mid- dle Miocene species Bruclarkia oregonensis (Con- rad) are of similar size and sculptural aspects to B. seattlensis, but the subsutural collar on this species is very subdued, if developed at all, and the upper part of the body whorl tends to be convex rather than concave (Etherington, 1931, pl. 11, figs. 1, 3, 4, 5, 7). This is the only known record of Bruclarkia seat- tlensis from California. Prior to its discovery in the Wygal (Addicott, 1973), it was known only from the type locality near Seattle, Wash., where it occurs with a diverse molluscan assemblage re- ferable to the upper Oligocene Echinophom'a rex Zone of the “Blakeley Stage” (Durham, 1944). Localities—USGS 5149, M3280, M3578, M3636, M3772, M3978, M3979, M4470, M4471. Class PELECYPODA Order NUCULOlDA Family NUCULIDAE Genus ACILA Adams and Adams, 1858 Subgenus TRUNCACILA Grant and Gale, 1931 Acila (Truncacila) muta Clark Plate 1, figures 4, 15 1918. Acila muta Clark, California Univ. Pubs., Dept. Geology Bull., v. 11, no. 2, p. 119—120, pl. 13, figs. 6,12,13. 1918. Acila muta var. markleyensis Clark, California Univ. Pubs., Dept. Geology Bull., v. 11, no. 2, p. 121, pl. 13, fig. 3. 1936. Acila (Tmncacila) mum B. L. Clark, Schenck, Geol. Soc. America Spec. Paper 4, p. 74—75, pl. 8 (figs. 4, 11; text fig. 7, nos. 20, 21. Types.——Holotype, UCMP UCMP 11199, 11200. Type locality—UCMP 1131, 1/2 mile southwest of Walnut Creek, Contra Costa County, Calif, in creekbed about 100 yards east of Oakland and Anti- och bridge. Altitude 150 feet. San Ramon Sandstone, early Miocene(?). Two specimens of Acila from the Wygal Sand- stone Member are identified as A. muta Clark, a 11196 ; paratypes, species heretofore known only from the San Ramon Sandstone near Mount Diablo, Contra Costa County, Calif. The outline of the valves and details of sur- face sculpture agree closely with Clark’s (1918, pl. 13, figs. 6, 12, 13) types. As indicated by Schenck (1936), the species that most closely resembles Acila muta is the middle Miocene A. comadi from the Temblor Formation (“Temblor Stage”) of California and from the Astoria Formation of Oregon and Washington. Schenck differentiated these two species on the rela- tively greater thickness of the valves of A. muta. They are differentiated here by the significantly smaller angle of bifurcation of A. muta evident from Clark’s (1918) and Schenck’s (1936) illustrated specimens. This difference is reflected by the fact that the primary line of bifurcation on specimens of A. conmdi intersects the ventral margin in a more anterior position than on specimens of A. muta. Two fragmentary specimens of AciLa from USGS localities M3280 and M3579 are doubtfully referred to this species on the basis of similar radial sculp- ture and what appears to be a similarly oriented primary line of bifurcation. One of these specimens (USGS loc. M2380) is intensely deformed. Localities—USGS M3280?, M3579?, M4471. M4466, Family NUCULANIDAE Genus YOLDlA M'éller, 1842 Subgenus KALAYOLDIA Grant and Gales, 1931 Yoldia (Kalayoldia) tenuissimn Clark Plate 1, figures 2, 3, 5, 8 Yoldia cooperii Gabb. Gabb, California Geol. Survey, Paleontology, v. 2, sec. 1, pt. 1, p. 31, pl. 9, fig. 4 [not Yoldia cooperii Gabb, 1865, California Acad. Sci. Proc., 1st ser., v. 3, p. 189]. Yoldia cooperi var. tenuissima Clark, California Univ. Pubs., Dept. Geology Bull., v. 11, no. 2, p. 125—126, pl. 11, fig. 10, pl. 12, figs. 8, 14. Yoldia tenuissima Clark. Clark, California Univ. Pubs. Geol. Sci., v. 15, no. 4, p. 78-79, pl. 8, figs. 5, 9. Yoldia cooperii Gabb subsp. supramontereyensis Arnold. Stewart, Acad. Nat. Sci. Philadelphia Spec. Pub. 3, p. 62—64 [in part], pl. 15, fig. 2. Yoldia (Portlandia) oregona (Shumard). Ethering- ton, California Univ. Pubs., Dept. Geol. Sci. Bull., v. 20, no. 5, p. 67, pl. 1, fig. 8. Yoldia (Portlandia) oregona (Shumard). Weaver, Washington Univ. [Seattle] Pubs. Geology, v. 5, p. 49 [in part], pl. 9, fig. 8. Yoldia tenuissima Clark. Kleinpell and Weaver, Cali- fornia Univ. Pubs. Geol. Sci., v. 43, p. 196, pl. 28, fig. 4. Yuldia (Kalayoldia)tenuissima(Clark). Hickman, Ore- gon Univ. Mus. Nat. History Bull. 16, p. 32—33, pl. 1, fig. 18. 1869. 1918. 1925. 1930. 1931. 1942. 1963. 1969. SYSTEMATIC DESCRIPTIONS 23 Types.—Holotype, UCMP UCMP 11163 and 11170. Type locality—UCMP 798, lat 37.2° N., long 122.1° W. San Ramon Sandstone, early Miocene(?) (no other data in the University of California locality register). Specimens of Yoldia, tenuissima from the Wygal Sandstone Member of the Temblor Formation are of uniform size and morphology; the umfbo is situ- ated posteriorly to the middle of the valve. In the similar Oligocene species from Oregon, Y. oregona (Shumard, 1858), the umbo is centrally located, or nearly so. Hickman (1969, p. 32) carefully consid- ered the morphologic differences between these spe- cies. Another middle Tertiary species that has been included with Y. tenuissima by Stewart (1930, p. 63), Y. supramontereyensis Arnold (1908, p. 382, pl. 35, fig. 9) from middle Miocene strata near Stan- ford University, California, has a medially situated umbo and a deeply concave posterior dorsal margin. These characters seem to permit differentiation from Y. tenm'ssima. The earliest occurrence of this species is in rocks of Oligocene age. It was used to name a molluscan subzone of the Refugian Stage in the western Santa Ynez Mountains by Kleinpell and Weaver (1963). It also commonly occurs in the upper part of the San Emigdio Formation (Wagner and Schilling, 1923; DeLise, 1967) in strata that are at least in large part referable to the Acila shumardi Zone of the Refugian Stage of Schenck and Kleinpell (1936) and the “Lincoln Stage” of Weaver and others (1944). Yoldia tenuissima is also recorded from the northern part of the Diablo Range (Clark, 1918, 1925; Weaver, 1949; Primmer, 1964). Hickman (1969) summarized Oligocene occurrences in Ore- gon and Washington and also assigned to this spe- cies a middle Miocene specimen from the Astoria Formation (Ethering‘ton, 1931, pl. 1, fig. 8) that had previously been assigned to Y. temblorensis by Trumbull (1958). Some specimens of middle Mio- cene age from the Kern River area, California (Addicott, 1956), also seem to represent Y. tenm’s- sima. A specimen from a locality near the base of the Round Mountain Silt is figured in this report (pl. 1, fig. 8). Another specimen from about the same stratigraphic position in the Olcese Sand (pl. 1, fig. 12), characterized by a uniquely short pos- terior dorsal slope and a posteriorly situated beak, may represent Clark’s (1915, p. 446, pl. 48, fig. 6) Y. camarosensis from the San Pablo Formation, Napa County, Calif. Fragments of a large Kalayoldia from the Salt Creek Shale of Foss and Blaisdell (1968) (UCR 11110; paratypes, loc. 1106) seem to represent a different species be- cause they are relatively larger and more coarsely ribbed than specimens of Y. tenm'ssima from the overlying Wygal Sandstone Member. They also have medially located beaks, unlike Y. tenuissima. These fragments might represent Y. oregona. Localities.—USGS 9432, M3578, M3636 ?, M3978, M4466, M4468?; UCR 1235. Order ARCOIDA Funily ARCIDAE Genus ANADARA Guy, 1847 Subgenus ANADARA I. s. Anadnra (Anadan) submontereynnn (Clark) Plate 1, figures 6, 7, 11, 16, 17 Area (Scapharca) submontereyana Clark, California Univ. Pubs., Dept. Geology Bull., v. 11, no. 2, p. 128—129, pl. 16, fig. 2. Anadara (Anada'ra) mediaimpressa var. submonterey- (ma (Clark). Schenck and Reinhart, Mus. Royal Histoire Nat. Belgique Mem., ser. 2, no. 14, p. 38. Anadam (Anadara) mediaimpressa subsp. monterey- ana (Clark). Reinhart, Geol. Soc. America Spec. Paper 47, p. 40—41, pl. 4, fig. 7. Anadara submontereyana (Clark). Addicott, Soc. Econ. Paleontologists and Mineralogists, Pacific Sec., Bakersfield, Calif. 1972, Proc., pl. 1, fig. 7. UCMP 11186; 1918. 1938. 1943. 1972. Types.~——Holotype, UCMP 11186A. Type locality—UCMP 52, near top of first ridge west of Walnut Creek, 114 miles south of Walnut Creek, Contra Costa County, Calif. Long 122°1’39” W., lat 37 °52’44” N. Altitude 350 feet. San Ramon Sandstone, early Miocene?. Representatives of this species from localities near the base of the type section of the Temblor Formation are characterized by about 23 broad, flat-topped ribs that are generally wider than the interspaces, particularly on the posterior part of the valves. The number of ribs varies from as few as 21 or 22 to as many as 25. Both the ribs and interspaces are seen to be very finely noded on a few of the better preserved specimens. The Wygal speci- mens differ from the types principally by their rela- tively larger size. As noted by Reinhart (1943), the beaks are located about one-third of the distance from the anterior extremity of the valves. Although the only reported occurrences of this species is from the type locality in the San Ramon Sandstone near Walnut Creek, Ca1if., poorly pre- served minute specimens of an unnamed species from rocks of probable late Oligocene age in the Santa Lucia Range (SU type colln. 7574) are very similar to this species. They have been compared with Anadam strongi (Loel and Corey) (Reinhart, 1943, p. 43). The unnamed species is from rocks mapped as Vaqueros Formation by Edwards (1940) paratype, 24 MOLLUSCAN BIOSTRATIGRAPHY, PALEONTOLOGY, TYPE TEMBLOR FORMATION, CALIF. that contain Pecten sanctaecruzensis Arnold, a pec- tinid whose presence implies correlation with the Wygal Sandstone Member of the Temeblor Formation and with strata assigned to the lower part of Klein- pell’s (1938) Zemorrian Stage. A. submontereyana differs from A. strongz', a somewhat similar species of early Miocene age, in having a few more radial ribs that are not strongly grooved. Poorly preserved specimens of Anadam from USGS localities M4466 and M4469 in the Wygal Sandstone Member may represent A. submonterey- mm but are specifically indeterminate. Localities—USGS 5149 cf., M3280, M3578, M3579, M3772, M3978. Order MYTlLOlDA Funny MYTILIDAE Genus CRENOMYTILUS Soot-Ryen, 1955 Crenomytilus'.’ cf. C.? arnoldi (Clark) Plate 1, figures 1, 18; plate 3, figure 2 A large mytilid characterized by thin valves and a broad umbonal angle is similar to Crenomytilus amoldi (Clark, 1918, p. 135, pl. 12, fig. 1), an Oligo- cene species from the Kirker Tufi‘ near Mount Diablo, Calif. Unfortunately, all the specimens from the Wygal Sandstone Member are strongly deformed and broken, but they differ from Clark’s species principally in size—they are more than twice the height of the holotype of C. arnoldi. The smaller specimens, however, seem almost indistinguishable from the holotype. They also resemble a form of the large early to middle Miocene species C. expan- sus (Arnold, 1907, p. 528, pl. 43, fig. 2). Although the umbonal angles are similar, the late Oligocene specimens from the Wygal Sandstone Member lack the finely striated surface characteristic of G. ex- pansus. Another large species from the lower Mio- cene “Vaqueros Stage” of California, C. loeli (Grant), differs from these specimens in having a strongly concave rather than straight ventral margin. Poorly preserved, incomplete mytilids from USGS localities M4466, M4469, M4470, and M4472 may represent this taxon. Assignment of these specimens to Crenomytilus, an extinct middle and late Tertiary genus from the North Pacific rim is doubtful because details of the resilial ridge and shell margins are unknown. It is, however, very similar to Tertiary species from the Pacific coast assigned to this genus by Soot-Ryen (1955). Crenomytilus arnoldi is reported from several Oligocene formations in California: the Kirker Tuif and the San Ramon Sandstone of the Mount Diablo M3281, area (Clark, 1918; Weaver, 1949; Primmer, 1964), the upper part of the San Emigdio Formation and the lower part of the Pleito Formation of the San Emigdio Mountains (Wagner and Schilling, 1923), and the San Juan Bautista Formation of the north- ern end of the Gabilan Range (Schenck in Allen, 1945). Localities.—USGS M3578, M3636 cf., M3979, M4468. Crenomylilus explnsus (Arnold) Plate 4, figure 15 1907. Mytilus mathewsom' Gabb var. expansus Arnold, U.S. Natl. Museum Proc., v. 32, no. 1545, p. 528, pl. 43, fig. 2. Mytilus mathewsom‘ Gabb var. nold, U.S. Geol. Survey Bull. Mytilus mathewsoni Gabb var. nold, U.S. Geol. Survey Bull. 396, pl. 5, fig. 3. Mytilus mathewsom' Gabb var. expansus Arnold. Ar- nold and Anderson, U.S. Geol. Survey Bull. 398, pl. 27, fig. 3. Mytilus expansus Arnold. Loel and Corey, California Univ. Pubs., Dept. Geol. Sci. Bull., v. 22, no. 3, p. 205, pl. 34, fig. 1. Type.—USNM 164968. Type locality—Near Torrey Canyon oil wells, southwest of Piru, Ventura County, Calif. Vaqueros Formation, lower Miocene. One specimen was collected from the Wygal Sand- stone Member near Carneros Springs. It differs from C’renomytilus? cf. 0. arnoldi (Clark), a relatively common species in the Wygal, in having a recurved beak and a finely striated surface. This stratigraphic occurrence of C. expansus extends the range into the provincial upper Oligocene, previous records being from strata assigned to the lower Miocene “Vaqueros Stage” (Loel and Corey, 1932; Fritsche, 1967; Adegoke, 1969). Fritsche (1967) suggested that C. expansus might be a synonym of C. coalin- gensis (Arnold, 1909, p. 73—74, pl. 19, fig. 5 and pl. 22, fig. 6), a Pliocene species from near Coalinga, Calif. Although the very strongly recurved and greatly thickened beak of C. coalingensis seem to distinguish it from C. expansus, the relationship of these two species certainly merits further study. Locality—USGS 9427. Order PTERlOlDA Family PECTINIDAE Genus PECTEN Miller, 1776 Suubgenus PECTEN 1. I. Pecten (Pecten) sanctaecruzensis Arnold Plate 2, figures 4, 7 1906. Pecten (Pecten) sanctaecruzensis Arnold, U.S. Geol. Survey Prof. Paper 47, p. 54—55, pl. 3, figs. 12, 13. 1909. Pecten (Pecten) sanctaecmzensis Arnold, U.S. Geol. Survey Folio 163, illus. II, fig. 18. M3978, 1907. expansus Arnold. Ar- 309, pl. 30, fig. 2. 1909. expansus Arnold. Ar- 1910. 1932. SYSTEMATIC DESCRIPTIONS 25 1923. Pecten (Pecten) dickersom' Wagner and Schilling, California Univ. Pubs., Dept. Geol. Sci. Bull., v. 14, no. 6, p. 253, pl. 45, fig. 1. 1940. Pecten dickersom' Wagner and Schilling. Schenck and Keen, California fossils for the field geologist, pl. 28, fig. 1. Pecten sanctaecmzensis Arnold. Addicott, Soc. Econ. Paleontologists and Mineralogists, Pacific Sec., Bakersfield, Calif., 1972, Proc., pl. 1, fig. 2. Types.—Holotype, SU 360; paratype, SU 361. Type locality.—SU locality 111, Two Bar Creek, 14 mile above junction with San Lorenzo River, Santa Cruz County. Vaqueros Formation, about 600 feet above base (Burchfiel, 1964, text fig. 2); late Oligocene. The paratype is from float collected in Bear Creek about 2 miles from the holotype locality. Several disarticulated valves of Pecten sanctua- cruzensis have been collected from exposures of a 3-foot-thick Crassostrea-bearing sandstone bed at the base of the Wygal Sandstone Member in a tribu- tary of Zemorra Creek (USGS loc. M4467). These specimens are similar to the type material from the Vaqueros Formation of the Santa Cruz Mountains (Arnold, 1906, p. 54—55, pl. 3, figs. 12, 13) except for their larger adult size. Right valves of the pectinid range from about 45 to 85 mm (millimeters) in length. They have 12 or 13 strong ribs which have gently rounded upper sur- faces and vertical sides. The interspaces are deep, relatively flat bottomed, and somewhat narrower than the ribs. On two of the larger specimens, there is a suggestion of fine radial sculpture on a few of the ribs. Two left valves in the USGS collection from local- ity M4467 are relatively flat. The interior of the better preserved valve is characterized by strong cardinal crura that extend almost to the ligamental pit. Each valve has 12 prominent radial ribs that are much sharper than those of the right valve and that are separated by relatively wide interspaces. These valves measure about 48 and 73 mm long. Pecten sanctaecruzensis is a potentially valuable species for stratigraphic correlation in California Coast Range basins. Its local stratigraphic range, or teilzone, and position in the provincial forami- niferal chronology are well known (Burchfiel, 1964). Although the record in the Wygal Sandstone Mem- ber is the first reported from east of the San An- dreas fault, this species also occurs at a locality in the Pleito Formation of the San Emigdio Mountains. It was originally described from “a faulted block of the San Emigdio Formation on Salt Creek” by Wagner and Schilling (1923, p. 253, pl. 45, fig. 1) as Pecten dickersom‘. Detailed mapping of this area 1972. (Dibblee, 1961) shows that the type locality is in the Pleito Formation of Wagner and Schilling (1923), and study of several topotypes from USGS locality M3747 indicates that the pattern of ribbing and the number of ribs are identical to P. sanctae- cruzensis. This Pecten occurs with an assemblage that is comparable to the fauna of the Wygal Sand- stone Member and that has no species restricted to the Refugian Stage. The only other known occur- rences are from the type area in the Santa Cruz Mountains (Arnold, 1906; Brabb, 1964; Burchfiel, 1964) and from the Salinas Valley (Edwards, 1940) . Both of these occurrences are in rocks mapped as Vaqueros Sandstone and assigned to the Zemorrian Stage of the foraminiferal chronology. It should be noted, however, that the Santa Cruz Mountains occurrence may be in strata generally older than the Vaqueros Sandstone of other areas in the Cali- fornia Coast Ranges. In fact, Arnold (1908, p. 349) originally noted that the fine-grained sandstones from which P. sanctaecmzensis was collected con- tained a faunal assemblage that appeared to be transitional between that of the San Lorenzo For- mation below and his Vaqueros Sandstone above. He concluded that the assemblage resembled the fauna of the San Lorenzo more closely than that of his Vaqueros Sandstone; his Vaqueros fauna (Ar- nold, 1908, p. 350) consisted of assemblages that are referable to the present—day “Vaqueros” and “Temblor Stages” of Weaver and others (1944). In addition to Arnold’s (1908) molluscan evi- dence, benthonic foraminifers from his transitional fine-grained sandstones include at least one species, Cibicides pseudoungerianus evolutus (Brabb, 1964, p. 677) that is suggestive of an early Zemorrian age (Kleinpell, 1938, p. 140). As discussed in the section on provincial age and correlation, indications are that the fauna of the “Vaqueros Stage” corre- sponds to the upper part of the Zemorrian and lower part of the Saucesian Stages of the provincial foraminiferal chronology. The subsequent mapping of Arnold’s (1908) transitional sandstone in the Vaqueros Sandstone (Brabb, 1964; Burchfiel, 1964) is not, therefore, indicative of an exclusively “Va- queros” age for occurrences of Pecten sanctaecru- zensis in the Santa Cruz Mountains. The indicated cooccurrence of this species with foraminiferal assemblages of early Zemorrian age in the Santa Cruz Mountains and in the Temblor Range (Kleinpell, 1938) suggests that the lower part of the Vaqueros Sandstone of Brabb (1964) and Bur‘chfiel (1964) is correlative with the Wygal Sandstone Member of the type Temblor Formation. Locality—USGS M4467. 26 MOLLUSCAN BIOSTRATIGRAPHY, PALEONTOLOGY, TYPE TEMBLOR FORMATION, CALIF. Genus LEI’TOPECTEN Verrill, 1897 Leptopeclen? 3]). Plate 1, figure 14 1972. Leptopecten? sp. Addicott, Soc. Econ. Paleontologists and Mineralogists, Pacific Sec., Bakersfield, Calif., 1972, Proc., pl. 1, fig. 11. A fragment of a small pectinid from USGS local- ity M3280 is doubtfully identified as Leptopecten sp. on the basis of similarity of the ribbing pattern to L. andersom' (Arnold, 1906, p. 82—83, pl. 26, figs. 5, 7), a Widespread middle Miocene species from the California Coast Ranges. Eleven ribs are pre- served on this incomplete specimen; they are V-shaped in cross section and rise to an angular crest. The interspaces contain microscopic growth lamellae that resemble those developed on specimens of L. andersom' from the so-called Barker’s Ranch fauna of middle Miocene age along the southeastern margin of the San Joaquin basin. A more complete specimen from the lower part of the Santos Shale (Addicortt, 1972, pl. 1, fig. 11; USGS 10c. M3983) seems to represent this species. Locality—USGS M3280. Genus VERTIPECTEN Grant and Gale, 1931 Vertipecten alexclarki Addicotl, n. sp. Plate 1, figures 9, 10, 13; plate 2, figures 1, 3, 5, 8, 9; plate 3, figures 1, 4 1905. Pecten sp. Anderson, California Acad. Sci. Proc., 3d ser., v. 2, no. 2, p. 170. 1906. Pecten (Chlamys) brannem‘ Arnold, U.S. Geol. Survey Prof. Paper 47, p. 55—56 (in part), [not pl. 3, figs. 9—11]. 1972. Vertipecten n. sp. Addicott, Soc. Econ. Paleontologists and Mineralogists, Pacific Sec., Bakersfield, Calif., 1972, Proc., pl. 1, figs. 16, 18. Moderately large, reaching as much as 115 mm in height, higher than long. Apical angle narrow, averaging about 70°. Right valve relatively flat; has 20 to 30 or more flat-topped, scaly ribs that tend to be dichotomous in the central portion of the disk but more irregular in development toward the sides. Scales are fine and closely spaced near beak but be- come coarser and more widely spaced toward base. Secondary riblets usually developed in the inter- spaces between the paired ribs. Anterior auricle long, set off by deep byssal notch. Auricle sculptured by five strong ribs bearing heavy flanges. Left valve convex and has rather uniform sculpture of primary, secondary, and tertiary rounded ribs bearing strong, rather widely spaced flanges toward the base. Medial primary and two lateral primaries that tend to divide the disk into quadrants are somewhat stronger than other primary ribs. This species is assigned to Vertipecten because of the distinct inequality of the valves, the right valve being relatively flat and the left being moderately convex (pl. 1, fig. 10), and because of the relatively greater strength of the central and two lateral pri- mary ribs on the left valve, the three of which tend to divide the valve into quadrants. In sculptural detail and in the size of the right anterior ear, the species resembles Chlamys. The right valve of Vertipecten alexclarkz' n. sp. is characterized by narrow, flat-topped ribs that are highly irregular in their development, whereas the left valve has a more or less orderly development of rounded primary, secondary, and tertiary ribs. The ribs of both valves are scaly. This species seems to be a lineal antecedent of the early and middle Miocene species Vertipecten perm‘m‘ (Arnold). The Miocene specifies differs, how- ever, by having fewer, much coarser ribs on the right valve that also tend to be smooth and generally are not dichotomous. The media] portion of the right valve of V. perm‘m‘ also has a depression coinciding with an interspace and reflecting a relatively much more strongly raised medial rib on the left valve. The sculpture of the right valve of Chlamys brom- neri (Arnold, 1906, p. 55-56, pl. 3, figs. 9—11), a middle Miocene Species from the eastern foothills of the Santa Cruz Mountains, is similar to that of the Vertipecten alexclarki. In fact, some workers (Kleinpell, 1938; Stinemeyer and others, 1959; Foss and Blaisdell, 1968) identified C. brmmeri from the Wygal Sandstone Member. Although the type ma- terial of C. branneri is poorly preserved, it is an equivalved species and, therefore, is properly as- signed to Chlamys and not to the subsequently pro- posed genus Vertipecten, which is characterized by a relatively flat right valve and a convex left valve (see pl. 1, fig. 10). Other differences are the similar sculpture of both valves of C. brmmem', the lack of an accentuated medial rib on the left valve, and the greater number of radial ribs on the auricles of that species—6 to 10 rather than 5. It is possible that a specimen, or specimens, identified by Arnold (1906, p. 56) as C. bromneri from the lower part of what is now mapped as Vaqueros Sandstone (Brabb, 1964) on Twobar Creek in the Santa Cruz Moun- tains, may represent V. alexclurki n. sp. The ma- terial is from rocks of approximately the same age. C. brannem', on the other hand, is from unnamed rocks of middle Miocene age from Coyote Hills near the Stanford University campus (Dibble, 1966). This species is named for Alex Clark, who made extensive unpublished studies of Oligocene and Miocene mollusks of the southern part of the San SYSTEMATIC DESCRIPTIONS 27 Joaquin basin for Shell Oil Co. during the 1930’s (1931-1935, Department of Geology, California University, Riverside). Localities.—-USGS 5149?, M3280, M3281, M3579, M3636, M3772, M3984, M3985, M4466, M4468, M4470, M4472. Family OSTREIDAE Genus CRASSOSTREA Sacco, 1897 Crassoslrea eldridgei ynezana (Loel and Corey) Plate 2, figures 2, 10 1907. Ostrea eldm'dgei Arnold. Arnold and Anderson, U.S. Geol. Survey Bull. 322, pl. 18, figs. 6a, 6b. 1932. Ostrea eld’m’dgei (Arnold) var. ynezana Loel and Corey, California Univ. Pubs., Dept. Geol. Sci. Bull., v. 22, no. 3, p. 188—189, pl. 11, fig. 3; pl. 12, figs. la—lc; pl. 13, figs. 1, 2a, 2b. Type.—UCMP 31745. Type locality.—UCMP A317. At west end of hill that is poorly defined on the topographic map about 0.5 mile northwest of Rancho Atascoso house. 1.2 miles S. 80° E. from 603-foot bench mark on road. Vaqueros Formation, early Miocene. Specimens of Crassostrea eldm‘dgei ynezana are characterized by their relatively thin valves, the left being moderately convex and somewhat larger than the right. The largest specimen is about 100 mm in height. During an early stage of growth the axis of the valve undergoes a gradual rotation through about 45°, the result being posteriorly situated beaks and, on many specimens, a subquadrate outline. This species is similar to Crassostrea sookensis (Clark and Arnold, 1923, p. 138, pl. 17, figs. 1, 2) from the early Miocene( ?) Sooke Formation of southern Vancouver Island, British Columbia. The northern species has a thicker shell in general and aless quadrate outline than specimens of C. eldm‘dgei ynezana from the Wygal Sandstone Member. C. sookensis may represent a prior name for C. el- dm'dgei ynezana, but the holotype is too poorly pre- served to determine possible relationships confi- dently. Fragments of a large, very thick shelled oyster from several localities in the Wygal Sandstone Member seem to represent a different species from Crassostrea eldm’dgei zjnezcma. Localities.—USGS M3280, M3578, M3772, M3978, M3979 ?, M4466 ?. Crassostrea? sp. Fragments of a large, very thick shelled ostreid occur commonly in exposures of the Wygal Sand- stone Member, particularly in the coarse-grained basal sandstone. The fragments range from about 15 to 25 mm thick; many are more than 100 mm long. Judging from these dimensions and the heavy ligamental groove, these fragments represent Cras- sostrea. The listing of occurrences is by no means an accurate index of the occurrence of this ostreid in the Wygal Sandstone Member; specimens were observed in the field at almost every locality but, because of the indeterminate nature of the frag- ments, were not collected. Localities.—USGS 5149, M4467, M4468, M4470, M4472. Order VENEROIDA Family LUCINIDAE Genus HERE Gabi), 1866 Here excavate (Carpenter) Plate 3, figures 6, 8, 9 Lucina excavata Carpenter, Catalogue of the Reigen Collection of Mazatlan Mollusca, in the British Museum, p. 98. Lucina (Here) Richthofem' Gabb, California Geol. Survey, Paleontology, v. 2, sec. 1, pt. 1, p. 29, pl. 8, figs. 49a, 49b. Lucina richthofeml Gabb. Tryon, Structural and sys- tematic conchology, v. 3, p. 210, pl. 119, figs. 46, 53. Phacoides (Here) richthofem' (Gabb). Dall, U.S. Natl. Mus. Proc., v. 23, no. 1237, p. 810, 827, pl. 40, figs. 7, 9. Phacoides richthofem‘ Gabb. Keep, West American shells, p. 58, fig. 45. Lucina richthofem' Gabb. Anderson, California Acad. Sci. Proc., 3d ser., v. 2, no. 2, p. 170. Phacoides richthofem' Gabb. Arnold, U.S. Natl. Mus. Proc., v. 32, no. 1545, p. 543, pl. 45, fig. 4. Phacoides richthofem' Gabb. Arnold, U.S. Geol. Survey Bull. 309, p. 236, pl. 32, fig. 4. Phacoides richthofeni Gabb. Keep, West coast shells, p. 69, fig. 45. Phacoides richthofeni Gabb. Clark, California Univ. Pubs., Dept. Geology Bull. v. 8, no. 22, p. 419, pl. 62, fig. 2. Lucina (Here) excavata Carpenter. Stewart, Acad. Nat. Sci. Philadelphia Spec. Pub. 3, p. 181—182, pl. 15, fig. 3, pl. 17, fig. 5. Lucina (Here) excavata Carpenter. Grant and Gale, San Diego Soc. Nat. History Mem., v. 1, p. 290— 291, pl. 14, figs. 2, 5, 10. Phacoides (Here) richthofeni Gabb. Loel and Corey, California Univ. Pubs., Dept. Geol. Sci. Bull., v. 22, no. 3, p. 210, pl. 36, fig. 4. Lucina (Here) richthofeni Gabb. Keep and Bailey, West coast shells, p. 84, fig. 57. Lucina (Here) excavata Carpenter. Keen, Sea shells of tropical west America, p. 94, fig. 188. Lucina excavata Carpenter. Brann, Illustrations to “Catalogue of the Collection of Mazatlan shells” by Philip P. Carpenter, p. 37, pl. 12, fig. 140. Here (Here) excavata (Carpenter). Chavan, Treatise on invertebrate paleontology, Part N, v. 2, p. N496, fig. E3, 1a, 1b. Type—British Museum (Palmer, 1958). Type locality—Mazatlan, Mexico. Holocene. 1857. 1866. 1884. 1901. 1904. 1905. 1907. 1907. 1911. 1915. 1930. 1931. 1932. 1935. 1958. 1966. 1969. 28 MOLLUSCAN BIOSTRATIGRAPHY, PALEONTOLOGY, TYPE TEMBLOR FORMATION, CALIF. This small, strongly inflated Iucinid is usually represented by articulated specimens. There is some variation in the degree of inflation of the valves and in the umbonal angle on specimens from the Wygal Sandstone Member. They are similar to, but somewhat more inflated than, specimens of middle Miocene age from the upper part of the Olcese Sand of the Kern River area on the southeastern margin of the San Joaquin basin. Here excavat‘a occurs commonly in rocks of Mio- cene and Pliocene age in California. Many of the occurrences are listed by Stewart (1930, p. 181). The species is one of the longest ranging Tertiary bivalves 0n the Pacific coast. Its earliest occurrence is in the middle Oligocene fauna of the San Emigdio Formation of Wagner and Schilling (1923) at the southern margin of the San Joaquin basin (Wagner and Schilling, 1923) where it occurs in the Acila shumardi Zone of the Refugian Stage of Schenck and Kleinpell (1936). Localities—USGS 9432, M3280, M3578, M3579, M3978, M4466. Genus LUCINOMA Ball, 1901 Lucinoma acutilineala (Conrad) Plate 2, figure 6; plate 3, figures 3, 7 Lucina acutilineata Conrad, U.S. Explor. Exped., Geology, v. 10, app. p. 725, atlas pl. 18, figs. 2, 2a, 2b. Pectunculus patulus Conrad, U.S. Explor. Exped., Geology, v. 10, app. p. 726, atlas, pl. 18, figs. 8, 8a. Cyclas permacra Conrad, U.S. 33rd Cong., 2d sess., H. Ex. Doc. 91, no. 7, pt. 2, p. 192, pl. 6, fig. 6. Lucina borealis Linnaeus. Yates, Southern Calif. Acad. Sci. Bull., v. 2, no. 7, pl. 8, fig. 13. Phacoides acutilineatus Conrad. Dall, U.S. Geol. Sur- vey Prof. Paper 59, p. 116—117, pl. 12, fig. 6. Phacoides acutilineatus Conrad. Arnold, U.S. Geol. Survey Bull. 396, p. 122, pl. 8, fig. 4. Phacoides acutilineatus Conrad. Arnold and Ander- son, U.S. Geol. Survey Bull. 398, p. 294, pl. 30, fig. 4. Phacoides acutilineatus (Conrad). Clark, Strati- graphy and faunal horizons of the Coast Ranges of California, pl. 23, fig. 7. Phacoides (Lucinoma) acutilineatus (Conrad). Eth- erington, California Univ. Pubs., Dept. Geol. Sci. Bull., v. 20, no. 5, p. 76—77, pl. 4, fig. 5. Lucina (Myrtea) acutilineata Conrad. Grant and Gale, San Diego Soc. Nat. History Mem., v. 1, p. 286—287 (in part), [not pl. 14, figs. 22a, 22b]. Phacoides (Lucinoma) acutilineatus (Conrad). Loel and Corey, California Univ. Pubs., Dept. Geol. Sci. Bull., v. 22, no. 3, p. 211, pl. 36, fig. 3. Lucinoma acutilineata (Conrad). Stewart, U.S. Geol. Survey Prof. Paper 205—C, pl. 15, fig. 9. [Imprint 1946.] Lucinoma acutilineata (Conrad). Moore, U.S. Geol. Survey Prof. Paper 419, p. 70—71, pl. 15, figs. 7410, 12. [Imprint 1963.] 1849. 1849. ?1857. 1903. 1909. 1909. 1910. 1929. 1931. 1931. 1932. 1947. 1964. 1969. Lucinoma acutilineata (Conrad). Hickman, Oregon Univ. Mus. Nat. History Bull. 16, p. 39—40, pl. 3, figs. 1, 4. 1972. Lucinoma acutilineata (Conrad). Addicott, Soc. Econ. Paleontologists and Mineralogists, Pacific Sec., Bakersfield, Calif. 1972, Proc., pl. 1, figs. 3, 15. Type.——-USNM 3519 (designated by Woodring, 1938). Type locality—Astoria, Oreg. Astoria Formation, Miocene. This is the species most characteristic of the basal sandstone member of the type Temblor Formation. It is very abundant at most localities, notably at USGS M3280. It was initially listed from the Wygal Sandstone Member by Anderson (1905, p. 170) as Lucina borealis. Lucinoma acutz’lz’neata is characterized by regu- larly spaced, sharp concentric lamellae. The inter- spaces are packed with extremely fine concentric threads. The spacing of the primary lamellae is variable, as noted by Moore (1963). On a few speci— mens, these ribs are spaced twice as far apart as on others, almost as if alternate ribs had been omitted as the shell was being deposited (pl. 3, fig. 3). Lucinoma acutilineata has its initial and lowest stratigraphic occurrence in rocks assigned to the middle Oligocene “Lincoln Stage” of western Ore- gon (Hickman, 1969). It ranges at least through the middle Miocene. Pliocene and Quaternary speci- mens that resemble L. acutilineata are generally identified as L. annulata (Reeve) on the basis of a somewhat longer posterior dorsal margin and a less strongly developed hinge which includes weaker anterior teeth (Tegland, 1933; Moore, 1963). The relative length of the posterior dorsal margin on specimens from USGS locality M3280 is variable (pl. 3, figs. 3, 7) owing, at least in part, to post— burial deformation of the valves. Localities—USGS 5149, 9432, M3280, M3578, M3579, M3636, M3772, M3978, M3979, M3984, M3985, M4466, M4468, M4469, M4470, M4471, M4472, M4473. Genus MlLTHA Adams and Adams, 1857 Miltha (Millha) sanctaecrucis (Arnold) Plate 3, figures 5, 10 1910. Phacoides (Miltha) sanctaecmcis Arnold, U.S. Geol. Survey Bull. 396, p. 57—58, pl. 6, fig. 6. [Imprint 1909.] 1910. Phacoides (Miltha) sanctaecrucis Arnold. Arnold and Anderson, U.S. Geol. Survey Bull. 398, pl. 28, fig. 6. 1932. Phacoides (Miltha) sanctacarucis Arnold. Loel and Corey, California Univ. Pubs., Dept. Geol. Sci. Bull., v. 22, no. 3, p. 211, pl. 36, fig‘. 5. SYSTEMATIC DESCRIPTIONS 29 1947. Miltha sanctaecrucis (Arnold). Stewart, U.S. Geol. Survey Prof. Paper 205—0, pl. 17, fig. 8. [Imprint 1946.] Type.—USNM 165569. Type locality—USGS 4861. In “reef beds,” 14 mile south and southeast of Barton’s cabin, in the NWV, sec. 23, T. 25 S., R. 18 E., Devils Den district, Kern County, Calif. “Vaqueros” [Temblor] Forma- tion, middle Miocene. This large, very slightly inflated lucinid is char- acterized by very fine, somewhat irregular con- centric sculpture. A network of extremely fine, irregular, radial riblets is evident on a few well- preserved specimens (pl. 3, fig. 10). This species is very similar to the present-day Miltha xantmi (Dall) from the Cape San Lucas area at the tip of Baja California, Mexico. It is generally distin- guished from the living species by its relatively longer shell: M. xcmtusi tends to be somewhat higher than long. The specimens at hand from the Wygal Sandstone Member of the Temblor Forma- tion also seem to have a somewhat longer posterior dorsal slope. Although the figured specimens are uniformly less inflated than Arnold’s (1909) holotype (USNM 165569) , a suite of 10 specimens from USGS locality M3578 includes two individuals that are at least as inflated as the type; the degree of inflation seems to be a variable morphologic character. The occurrence of Miltha in the basal sandstone of the type Temblor Formation marks the lowest stratigraphic occurrence of the genus in the Cali- fornia Tertiary. The genus ranges through the Mio— cene into the early Pliocene of the San Joaquin basin and the Ventura basin (Grant and Gale, 1931). It is recognized in paleoclimatic analyses as a warm-water indicator because it occurs in the tropical Panamic molluscan province (Durham, 1950; Addicott and Vedder, 1963). Localities—USGS 5149, 9432, M3280, M3578, M3579, M3772, M3979, M4466, M4468, M4469. Family UNGULlNlDAE Genus FELANIELLA Dell, 1899 Felaniella harfordi (Anderson) Plate 4, figures 2, 7 Diplodonta harfordi Anderson, California Acad. Sci. Proc., 3d ser., v. 2, no. 2, p. 197, pl. 17, figs. 88, 89. Diplodonta harfordi Anderson. Arnold, U.S. Geol. U.S. Geol. Survey Bull. 398, pl. 39, fig. 6. [Imprint 1909.] Diplodonta harfordi Anderson. Arnold and Anderson, U.S. Geol. Survey Bull. 396, pl. 39, fig. 6. Diplodonta harfo’rdi Anderson. Loel and Corey, Cali- fornia Univ. Pubs., Dept. Geol. Sci. Bull., v. 22, no. 3, p. 212, pl. 36, figs. 11a, 11b. 1905. 71910. 71910. 1932. 493-696 0 - '73 - 5 Types—CA8 62, 63 (syntypes). Type locality—Three miles west of Coalinga, Fresno County, Calif. Coalinga Beds of Anderson (1905) [Temblor Formation], middle Miocene. The valves of Felamiella harfordi from the basal sandstone member of the type Temblor Formation are somewhat variable in outline and are only slightly inflated. Generally, they are subquadrangu— lar and have a distinctly flattened segment posteri- orly (pl. 4, fig. 2). Although the interior of these specimens is unknown, assignment to Felamiella is indicated by the characteristic inequilateral profile of the valves and the prominent beaks. The stratigraphic range of this species in the San Joaquin basin is from the upper Oligocene to the Pliocene. Prior to recent records from the Etche- goin Formation (Adegoke, 1969), this species has been recognized from only one locality of Pliocene age (Arnold, 1909, pl. 17, fig. 6), a record that had been considered doubtful by Grant and Gale (1931, p. 294). Localities—USGS M3280, M3772, M4466, M4472. Family MACTRlDAE Genus PSEUDOCARDIUM Gabb, 1866 Pseudocardium? 5]). Plate 5, figure 5 An internal mold of a moderately large, inequi- lateral pelecypod from USGS locality M4466 near Zemorra Creek is doubtful identified as Pseudocar- diam on the basis of strongly inflated valves and deeply sunken muscle scars. Locality—USGS M4466. Genus SPISULA Gray, 1837 Spisula cf. 5. albaria (Conrad) Plate 4, figures 3, 4, 8 Several small, poorly preserved specimens (30 to 50 mm long) resemble the Miocene and Pliocene species Spisula albaria (Conrad) more closely than any of the species described from the Oligocene of the Pacific coast. Although somewhat variable in outline, the valves have a trigonal outline and medi- ally to slightly anteriorly situated beaks. They are relatively much higher than the elongate trigonal late Oligocene species S. mmonensis attenuata Clark (1918, p. 158—159, pl. 9, fig. 6) and S. pitts- burgensis frustra Tegland (1933, p. 121, pl. 9, figs. 9—12). Moreover, the posterior dorsal slope is rela- tively longer and less convex than on S. ramonensis attenuata. A figure of a specimen of S. albaria (Conrad) from the middle Miocene Astoria Forma- tion of coastal Oregon is included for comparison (pl. 4, fig. 6). Localities—USGS 5149, M3578, M3636, M3978, 30 MOLLUSCAN BIOSTRATIGRAPHY, PALEONTOLOGY, TYPE TEMBLOR FORMATION, CALIF. M3979, M4466?, M4468, M4469, M4470, M4471, M4472. Spisula n. "1.? Plate 4, figure 1 1932. Spisula afl‘. hemphilli (Dall) n. sp.? Loel and Corey, California Univ. Pubs., Dept. Geol. Sci. Bull., v. 22, no. 3, p. 232, pl. 45, fig. 3. Figured specimen.—UCMP 31791. Locality.—UCMP A562. Pelecypod bed at about the 500-foot contour on the north wall of large east- west bend of Malibu Canyon north of dam. Western Los Angeles County, Calif. Vaqueros Formation, early Miocene. A large, inflated Spisula represented by single specimens from three localities is identical to the poorly known taxon from the Vaqueros Formation in the central Santa Monica Mountains. As recog- nized by Loel and Corey (1932, p. 232), this Spisula seems to be uniquely different from any of the described middle Tertiary species from the Pacific coast. Localities—USGS M3579, M3772, M3979. Spisula ramonensis Packard Plate 4, figure 5 1916. Spisula albaria (Conrad) var. ramonensis Packard, California Univ. Pubs., Dept. Geology Bull., v. 9, no. 15, p. 291—292, pl. 23, fig. 5, pl. 25, figs. 1, 2. 1918. Spisula ramonensis Packard. Clark, California Univ. Pubs., Dept. Geology Bull., v. 11, no. 2, p. 158, pl. 9, figs. 4, 5. Type.—UCMP 11118. Type locality—UCMP 1687. One and one-half mile south of Walnut Creek in valley leading from San Ramon Valley to Tice Valley. On bank of Wal- nut Creek, % mile west of the point where it turns and flows west. Contra Costa County, Calif. Altitude 225 feet. San Ramon Sandstone, early Miocene(?). A few specimens from USGS locality M3280 have the long, nearly straight anterior dorsal slope and posteriorly located beaks characteristic of this species. Spisula mmonensis occurs in several forma- tions of Oligocene age in California (Wagner and Schilling, 1923; Allen, 1945; Primmer, 1964; DeLise, 1967) and has been recorded from the lower Mio- cene part of the Poul Creek Formation in the Gulf of Alaska Tertiary province (Clark, 1932). Localities—USGS M3280, M4468 cf. Spiaula cf. S. rushi Wagner and Schilling Plate 4, figure 14 A few large Spisula from localities in the Wygal Sandstone Member of the Temblor Formation are similar to a species from the Oligocene formations exposed in the foothills of the San Emigdio Moun- tains, S. rushi Wagner and Schilling (1923, p. 256, pl. 47, fig. 2). These specimens are much larger than S. rushi. They are distinguished from Spisula n. sp., with which one specimen occurs in the collection from USGS locality M3979, by the subquadrate out- line of the valves and by the anterior location of the beaks. A similar species from the San Ramon Sandstone of the San Francisco Bay area to the north, Spisula occidentalis (Gabb) (Stewart, 1930, pl. 15, fig. 5), has a more oblique umbonal angle and has a promi- nent sulcus extending from near the beaks to the posterior dorsal extremity. Szn'sula rushi ranges from the middle Oligocene into the late Oligocene in southern part of the San Joaquin basin. It occurs with the Refugian Stage (Schenck and Kleinpell, 1936) index fossil Avila shumardi in the upper part of the San Emigdio and the lower part of the Pleito Formations of Wagner and Schilling (1923) at the southern margin of the San Joaquin basin (Wagner and Schilling, 1923; DeLise, 1967). Localities—USGS M3280, M3636, M3978, M3979, M4470, M4471, M4472 ?. Family SOLENIDAE Genus SOLEN “ME, 1758 Solen a5. 5. gravidus Clark Plate 4, figure 9 Fragments of a large Solen from USGS localities M3280 and M3578 are similar to S. gravidus Clark (1918, p. 156-157, pl. 10, fig. 7) from the San Ramon Formation of the Mount Diablo area, Contra Costa County, Calif. The anterior margin is gently rounded and is bordered by a weak, shallow sulcus that extends from near the beaks to the anterior ventral margin. The strength of the sulcus is vari- able in the few specimens available for study. The height of the valves is proportionately greater than that of the holotype of S. gravidus (UCMP 11133). There are additional fragments of a large Solen in the collections from USGS localities M3772 and M3979 that are specifically indeterminate but pre- sumably represent this taxon. Localities.—USGS M3280, M3578, M4466 ?. Family TELLlNlDAE Genus TELLINA Linni. 1758 Subgenus OLCESIA Addicotl, a. when. Type.——Tellina piercei (Arnold) [= T. nevadensis Anderson and Martin]. Middle Miocene, California. Shell large and heavy. Posterior side longer than anterior side. Posterior portion of valves flexed to the right, flexing much stronger in right valve. Sur- face sculptured by fine concentric lines of growth SYSTEMATIC DESCRIPTIONS 31 and very faint radial striae. Anterior dorsal margin broadly convex, posterior margin long and relatively straight. Hinge plate broad and very heavy. Right valve has a long anterior lateral, a weak anterior cardinal, a heavy, bifid main cardinal, and a weak posterior lateral. Left valve has a weak anterior lateral, a simple cardinal, a weak posterior cardinal, and a strong, slightly curved posterior lateral ter- minating near the base of the hinge plate. Left valve flattened in dorsal aspect, has slight convexity; an- terior part of right valve convex, posterior part concave. Right valve has an elongate, weak lunule. Pallial line deeply impressed, extending to base of the anterior muscle scar. Pallial sinus deep, has slight convexity dorsally, extending anteriorly to a prominent internal rib bordering the anterior muscle scar. Anterior muscle scar elongate, tear shaped; posterior scar subquadrate. Ligament elongate, sunken, base deeply set on hinge plate. Olcesia seems to be most similar to the European Oligocene to Holocene genus Peronaea, Poli (Afshar, 1969, p. 30—31, pl. 5, figs. 6—10). The valves are of very similar shape and sculpture, but there are significant differences internally. The hinge of Olcesia is relatively much heavier, particularly an- terior to the beaks, the pallial sinus is also lower and does not form a prominent peak, and the an- terior margin of the posterior muscle scar is straight rather than highly contorted. The shape of the valves and external sculpture also resemble the Eocene to Holocene northern Pa- cific subgenus Perom’dia Dall (1900), but the deeply sunken ligamental groove, heavy hinge plate, and strong dentition of Olcesia are strikingly different. Olcesia ranges from the upper Oligocene to the upper Miocene in California. It is named for the Olcese land holdings near the type locality of Tellina nevadensis. The name has also been applied to the middle Miocene Olcese Sand from which this species was described. Tallinn (Olcesia) piercei (Arnold) Plate 4, figures 10, 12; plate 5, figures 12, 13 Macoma piercei Arnold, U.S. Geol. Survey Bull. 396, p. 55—56, pl. 7, fig. 6. [Imprint 1909.] Macoma piercez‘ Arnold. Arnold and Anderson, U.S. Geol. Survey Bull. 398, pl. 7, fig. 6. Tellina nevadensis Anderson and Martin, California Acad. Sci. Proc., 4th ser., v. 4, no. 3, p. 61—62, pl. 2, figs. 3a—3c. Tcllina nevadensis Anderson and Martin. Clark, Strati- graphy and fauna] horizons of the Coast Ranges of California, pl. 24, fig. 3. Tellina ocoyana Conrad. Loel and Corey, California Univ. Pubs., Dept. Geol. Sci. Bull., v. 22, no. 3, p. 148 (in part), pl. 43, fig. 4. 1910. 1910. 1914. 1929. 1932. 1932. Macoma nasuta (Conrad). Loel and Corey, California Univ. Pubs., Dept. Geol. Sci. Bull., v. 22, no. 3, p. 228 (in part), pl. 43, fig. 7 (not fig. 8). 1965. Tellina nevadensis Anderson and Martin. Addicott, U.S. Geol. Survey Prof. Paper 525—0, fig. 4e. Type.—-USNM 165595. Type locality.——USGS 4631, Turn'tella bed on east flank of high hill northeast of Oil City, in SEMLNEML sec. 16, T. 19 S., R. 15 E. Vaqueros [Temblor] Formation, middle Miocene. This species has been generally identified as Tellina nevadensis Anderson and Martin (1914) from the middle Miocene Olcese Sand of the Kern River area on the southeast margin of the San Joaquin basin. Arnold’s (1909, pl. 7, fig. 6) T. piercez’ has remained a poorly known taxon because the holotype is incomplete and poorly preserved. Moreover, it was originally compared with the Holocene species Macoma secta. The holotype, a left valve, clearly is not allied to M. secta. It lacks the convex posterior dorsal margin characteristic of that species and has an extremely long, deeply set ligamental groove. Although the hinge is not exposed on the holotype, there can be little doubt that it is idenical to T. nevadensis owing to the long, deeply excavated ligamental groove, the strong ridge below the posterior dorsal margin, the anteriorly located beak, and the surface sculpture of fine concentric ribs. The types of both species are from strat- igraphic units referable to the “Temblor Stage.” Only four specimens of Tellina piercei from the Wygal Sandstone Member of the Temblor Forma- tion are available for study. They are identical in all respects to well-preserved specimens from the upper part of the middle Miocene Olcese Sand from the Kern River area on the east side of the San Joaquin Valley (pl. 5, figs. 12, 13). This large, thick-shelled species is characterized by a very heavy hinge plate that has well-developed lateral teeth. The valves are boat shaped and are pointed posteriorly. On the left valve, the sharp posterior dorsal margin is bordered by a weak ridge, below which is a narrow sulcus; on the right valve, it is bordered by a much stronger sulcus below which occurs a prominent ridge that extends to the posterior extremity. Tellina piercei [as T. nevadensis] has been con- fused with a poorly known middle Miocene species from the Kern River area, T. ocoyamz Conrad (1857d, pl. 8, figs. 75, 75a). The identity of Con- rad’s species may never be satisfactorily determined because the type has apparently been lost (Keen and Bentson, 1944, p. 115) and the original figures appear to be somewhat stylized line drawings that 32 MOLLUSCAN BIOSTRATIGRAPHY, PALEONTOLOGY, TYPE TEMBLOR FORMATION. CALIF. do not indicate the critical details of internal morphology. Conrad’s description (1855, p. 19) and illustrations are of a more trigonal shell that has a pointed, rather than broadly rounded anterior extremity, and the hinge of T. ocoyana does not seem nearly so broad and heavy as that of T. piercei. Most subsequent records of T. ocoyana are pre— sumed to be of T. piercei. This species ranges from the upper Oligocene to the middle Miocene in the San Joaquin basin. It has not previously been recorded from strata of Oligo- cene age. A record of Tellina nevadensis from the upper Miocene Santa Margarita Formation near Coalinga, Calif. (Adegoke, 1969, p. 131), is not of this species. It is based on an internal mold (UCMP 36765) that is much more acute posteriorly than T. piercei and that can only doubtfully be identified as a tellinid. Localities—USGS 9432, M3578, M3978, M4471?. Subgenus OUDARDIA Monterosato, 1884 Tellina (Oudardia) emacerala Conrad Plate 4, figure 11 Tellina oreyonensis Conrad, Am. Jour. Sci., 2d ser., v. 5, p. 432, fig. 5 [nomen dubium, Moore, 1963, p. 79]. Tellina emacerata Conrad, U.S. Explor. Exped., Geology, v. 10, app. p. 725, atlas pl. 18, fig. 4. Tellina clallamensis Reagan, Kansas Acad. Sci. Trans, v. 22, p. 186-187, pl. 2, fig. 18. Tellina oregonensis Conrad. Clark, California Univ. Pubs., Dept. Geology Bull., v. 11, no. 2, p. 152—153, pl. 13, fig. 1, 4. Tellina bodegenst‘s Hinds n. subsp.? Clark and Arnold, California Univ. Pubs., Dept. Geol. Sci. Bull., v. 14, no. 5, p. 149, pl. 22, figs. 7 and ?8. Tellina (Perom'dia) oregonensis Conrad. Etherington, California Univ. Pubs., Dept. Geol. Sci. Bull., v. 20, no. 5, p. 83—84, pl. 10, figs. 6, 7. Tellina emacerata Conrad. Weaver, Washington Univ. [Seattle] Pubs. Geology, v. 5, p. 206, pl. 48, fig. 20, [not pl. 48, fig. 18]. Tellina emacerata Conrad. Moore, U.S. Geol. Survey Prof. Paper 419, pl. 29, figs. 6, 7, 13, 14. [Imprint 1963.] Type.——USNM 3494. Type locality—Astoria, Oreg. Astoria Formation, Miocene. This species is represented by three specimens from USGS locality M3578 that are identical to specimens of middle Miocene age from the Astoria Formation near Newport, Oreg. (USGS loc. M2604). It is characterized by a prominent ridge running from the beaks to the posterior extremity and a flattened area between this ridge and the convex posterior dorsal margin. Tellina emacemta has been perhaps more com- monly known as T. oregonensis Conrad (1848), a 1848. 1849. 1909. 1918. 1923. 1931. 1942. 1964. poorly illustrated species that seems best regarded as a nomen dubium because the type has been lost (Moore, 1963, p. 79). T. emacemta is assigned to the subgenus Oudardia on the basis of the slanting in- ternal rib near the anterior margin of the valves. This rib is only faintly discernible on the specimen from the Wygal Sandstone Member of the Temblor Formation, apparently because of poor preserva- tion. This is the first record of Tellina emacerata, from strata of Oligocene age in the San Joaquin basin. It ranges from the late Oligocene to the late Miocene (Trask, 1922) in California but has a much shorter range in Oregon and Washington, having been re— ported only from strata of early and middle Miocene age (Durham, 1944; Moore, 1963). Locality—USGS M3578. Subgenus TELLINELLA March, 1853 Tellina (Tellinelln) tennilineala Clark Plate 5, figures 2, 8 1918. Tellina tenuilineata Clark, California Univ. Pubs., Dept. Geology Bull., v. 11, no. 2, p. 153—154, pl. 10, figs. 1, 3, 5. 1929. Tellina tenuilineata Clark. Clark, Stratigraphy and fauna] horizons of the Coast Ranges of California, pl. 16, fig. 1. Types. — Holotype, UCMP 11128. Type locality—UCMP 1131, 1/2 mile southwest of Walnut Creek, Contra Cosa County, Calif., in creek bed about 100 yards east of Oakland and Antioch bridge. Altitude 150 feet. San Ramon Sandstone, early Miocene(?). Three specimens of this distinctive tellinid are in the collections from the lowest sandstone member of the Temblor Formation. They are relatively more elongate than Clark’s type specimen from the San Ramon Sandstone, and one specimen (pl. 5, fig. 2) is much more strongly flexed posteriorly than the holotype. This condition is attributed to deforma- tion because the other specimens are only very slightly flexed. In addition to occurrences in the San Ramon Sandstone (Clark, 1918), Tellina. tenm'lt'neata has been reported from the San Juan Bautista Forma- tion (Allen, 1945) and from middle Miocene locali- ties in the Temblor Formation (Loel and Corey, 1932). Localities—USGS 5149, 9432, M3578, M4466 ?. Subgenus 7 UCMP 31169; paratype, Tellina cf. T. lownsendensia Clark Plate 6, figure 7 Two poorly preserved specimens of an elongate tellinid from USGS locality M3578 seem to repre- SYSTEMATIC DESCRIPTIONS 33 sent the Oligocene species Telllina townsendensis Clark (1925, p. 94, pl. 12, figs. 11, 12) from Wash- ington. Records of this species from Washington (Durham, 1944; Clark, 1925) are from the lower part of the “Lincoln Stage”: Durham’s (1944) Molopophorus stephensoni and Molopophoms gabbi Zones. This taxon differs from the “Lincoln Stage" species Tellina adunccmasa Hickman (1969, p. 55— 56, pl. 6, figs. 7—12) in being somewhat longer and in having a broadly convex posterior dorsal margin. Locality.—USGS M3578. Tellina? cf. T. vancouverenais Clark and Arnold Plate 4, figure 13 A very poorly preserved tellinid from USGS 10- cality M3578 appears to be conspecific with Tellina vancouverensis Clark and Arnold (1923, p. 149—150, pl. 22, figs. 5, 6) from the Sooke Formation of southwestern Vancouver Island, British Columbia. Locality.—USGS M3578. Genus MACOMA Leach, 1819 Macoml arctata (Conrad) Plate 5, figures 4, 7 Tellina arctata Conrad, U.S. Explor. Exped., Geology, v. 10, app. p. 725, atlas pl. 18, figs. 3, 3a. Tellina arc-tutu Conrad. Reagan, Kansas Acad. Sci. Trans,v.22,p.184,186,pL 2,figs.16,16a 7%lHna awfiata Conrad van juana Reagan, Kansas Acad. Sci. Trans, v. 22, p. 186, pl. 1, fig. 17. Macoma wynootcheensis Weaver [in part], Washing- ton Geol. Survey Bull. 15, p. 66, pl. 15, fig. 130 [not figs. 128, 129]. Macoma arctata. (Conrad). Clark, Stratigraphy and fauna] horizons of the Coast Ranges of California, pl. 22, figs. 4, 6. Macoma (Psammacoma) arctata (Conrad). Ethering— ton, California Univ. Pubs., Dept. Geol. Sci. Bull., v. 20, no. 5, p. 84—85, pl. 10. Macoma arctata (Conrad). Loel and Corey, Califor- nia Univ. Pubs., Dept. Geol. Sci: Bull., v. 22, no. 3, p. 227 (in part), pl. 43, fig. 2 [not fig. 1]. Macoma arctata (Conrad). Weaver, Washington Univ. [Seattle] Pubs. Geology, v. 5, p. 208—209, pl. 49, figs. 3, 5, 12, pl. 59, fig. 15. Macoma arctata (Conrad) var. wynoocheensis Weaver. Weaver, Washington Univ. [Seattle] Pubs. Geology, v. 5, p. 209, pl. 49, fig. 8. Macoma arctata (Conrad). Hall, California Univ. Pubs., Dept. Geol. Sci. Bull., v. 34, no. 1, p. 55, pl. 8, fig. 4. Macoma arctata (Conrad). Moore, U.S. Geol. Survey Prof Paper419,p.81,pL 2& figs & 7,10,11,13 pl. 29, fig. 8. [Imprint 1964.] Macoma arctata (Conrad). Addicott, Soc. Paleontologists and Mineralogists, Pacific Bakersfield, Calif., 1972, Proc., pl. 1, fig. 14. 1849. 1909. 1909. 1912. 1929. 1931. ?1932. 1942. 1942. 1958. 1964. 1972. Econ. Sec., Type.—USNM 3489. Type locality—Astoria, Oreg. Astoria Formation, Miocene. This species is characterized by moderately large, elongated valves that have posteriorly situated beaks and a fairly strong posterior flexure. It is somewhat similar to the living species, Macoma nasuta (Con- rad), which has a higher shell and more centrally located beaks and which is more acute and more strongly flexed posteriorly. Specimens in the Wygal Sandstone Member of the Temblor Formation constitute a range extension of this species. The lowest previously known strati- graphic occurrence in California is from the early Miocene “Vaqueros Stage” (Loel and Corey, 1932, p. 227). Macoma arctata also occurs in strata of early Miocene age in northwestern Oregon where it is found in the Nye Mudstone of the Newport em- bayment in association with mollusks referable to the upper part of the “Blakeley Stage” (USGS loc. M3630). This species is Widespread in strata of middle Miocene age from southern California to the Gulf of Alaska where there is an unrecorded occur- rence in the lower part of the Yakataga Formation (Saburo Kanno, written commun., July 1970). It is not known to occur in the late Miocene. This is one of the few stratigraphic records of large species of Macoma from pre-Miocene strata of the Pacific coast. None is recorded from the Oligocene of western Washington (Tegland, 1933; Durham, 1944) or western Oregon (Warren and others, 1945; Vokes and others, 1949). There is, however, a middle Oligocene record from the San Emigdio Mountains at the southern margin of the San Joaquin basin in California—Macoma cf. M. nasuta (Conrad) by Wagner and Schilling (1923, p. 248). This taxon, which occurs with Acila shu- mardi in the lower part of the Pleito Formation of Wagner and Schilling (1923), could conceivably represent M. arctata. The holotype of Weaver’s Macoma wynootcheen- sis (1912, p. 66, pl. 15, fig. 130) is M. arctata. The other specimens from the type locality (Weaver, 1912, pl. 15, figs. 128, 129), which is in the Astoria Formation (Gower and Pease, 1965) in the vicinity of Montesano, Wash., have extremely elongate valves and beaks that are situated unusually close to the posterior margin. These may represent an extremely variable form of M. arctata, or an entirely different species. Localities.—USGS 5149, M3280, M3281, M3578, M3579 cf., M3772, M4466, M4468, M4469, M4471 cf., M4472 ?. 34 MOLLUSCAN BIOSTRATIGRAPHY, PALEONTOLOGY, TYPE TEMBLOR FORMATION, CALIF. Genus HETEROMACOMA Hebe, 1952 Heteromacoml rostellata (Clark) Plate 5, figures 1, 3, 6, 9—11; plate 6, figures 2, 4, 14 Metis rostellata Clark, California Univ. Pubs., Dept. Geology Bull., v. 11, no. 2, p. 154—155, pl. 9, fig. 7. 1918. 1923. Metis vancouverensis Clark and Arnold, California Univ. Pubs., Dept. Geol. Sci. Bull., v. 14, no. 5, p. 150, pl. 22, figs. 3, 4. 1932. Macoma sespeensis Loel and Corey, California Univ. Pubs., Dept. Geol. Sci. Bull., v. 22, no. 3, p. 228— 229, pl. 43, figs. 10—12. 1933. Poromya n. sp. Tegland, California Univ. Pubs., Dept. Geol. Sci. Bull., v. 23, no. 3, p. 92. 1942. Poromya teglandae Weaver, Washington Univ. [Seat- tle] Pubs. Geology v. 5, p. 121—122, pl. 25, fig. 23. 1942. Apolymetis vancouverensis (Clark and Arnold). Weaver, Washington Univ. [Seattle] Pubs. Geology, v. 5, p. 221, pl. 50, fig. 17. 1944. Apolymetis twinensis Durham, California Univ. Pubs., Dept. Geol. Sci. Bull., v. 27, no. 5, p. 150, pl. 13, fig. 7. 1951. Macoma sespeensis Loel and Corey. Lutz, California Univ. Pubs., Dept. Geol. Sci. Bull., v. 28, no. 13, p. 387—388, pl. 16, figs. 5, 6. 1963. Apolymetis cf. A. sespeensis (Loel and Corey). Klein- pell and Weaver, California Univ. Pubs., Dept. Geol. Sci. Bull., v. 43, p. 207, pl. 38, fig. 2. 1966. Macoma (Heteromacoma) vancouverensis (Clark and Arnold). Addicott, Jour. Paleontology, v. 40, no. 3, p. 644—645, pl. 76, figs. 1, 4. 1969. Macoma (Heteromacoma) vancouverensis (Clark and Arnold). Hickman, Oregon Univ. Mus. Nat. History Bull. 16, p. 58-59, pl. 7, figs. 3, 5, 7. 1972. Heteromacoma rostellata (Clark). Addicott, Soc. Econ. Paleontologists and Mineralogists, Pacific Sec. Bakersfield, Calif., 1972, Proc., pl. 1, figs. 1, 6. Type.—UCMP 11120. Type locality.—UCMP 1131, 1/2 mile southwest of Walnut Creek, Contra Costa County, Calif., in creek bed about 100 yards east of Oakland and Antioch bridge. Altitude 150 feet. San Ramon Sandstone, early Miocene(?). H eteromacoma rostellata is an unusually variable species, as suggested by the number of specific names included in the synonymy. A large population of mature individuals represented in the collection from USGS locality M3280 clearly demonstrates the range of variation in outline of the valves. The most common forms are of subquadrate (pl. 5, figs. 10 and 11; pl. 6, fig. 2) and elongate shape (pl. 5, fig. 9; pl. 6, fig. 4) ; a less common form is suborbicular in outline (pl. 5, figs. 1, 3, 6; pl. 6, fig. 14). Some of the subquadrate specimens resemble the veneroid genus Pita/r, particularly those in which the posterior part of the valve has been broken off. This form was originally described by Clark (1918, p. 154—155, pl. 9, fig. 7) as Metis rostellata from the San Ramon Sandstone of Contra Costa County, Calif. The elongate form was described by Loel and Corey (1932, p. 228—229, pl. 43, figs. 10—12) as Macoma sespeensis from the Vaqueros Formation of coastal California. The suborbicular form that was named Metis vancouverensis by Clark and Arnold (1923, p. 150, pl. 22, figs. 3, 4) was based on specimens from the lower Miocene( ?) Sooke Formation of southwestern Vancouver Island, British Columbia. An extreme variant of this form, almost suborbicular in outline, from the upper member of the Twin River Forma- tion of western Washington (Addicott, 1966), was named Poromya teglandae by Weaver (1942, p. 121—122, pl. 25, fig. 23). Variation in this taxon was first recognized by Kleinpell and Weaver (1963, p. 207) , who suggested that Poromya teglandae, Mac-oma sespeensis, and Apolymetis twinensis were synonyms. Subsequently, Addicott (1966) and Hickman (1969) extended this treatment to include the suborbicular species Metis vancouverensis Clark and Arnold in this taxon. The present concept of this species further includes Metis rostellata Clark. Primmer (1965) first assigned Macoma sespeen- sis and Metis rostellata to Hetero-momma, a very shallow—water genus that is living off Japan and Korea (Kira, 1962, p. 172), and included these two species in a new subgenus, which he differentiated from Heteromacoma s. s. principally because of a somewhat shorter ligament and related hinge struc- ture. In recognizing these as distinct species, he contrasted the elongate shape of H. sespeensis with the subquadrate shape of H. rostellata and further indicated minor differences in ligamental width and surface of the nymph plate. However, specimens from the middle Oligocene, lower part of the Pleito Formation of Wagner and Schilling ( 1923, their Ice. 3179) of the San Emigdio Mountains seemed to indicate that these differences were not constant and that the two taxa might intergrade (Primmer, 1965, p. 137). The genus Heteromacoma is characterized by a prominent sulcus extending from near the beaks to the posterior ventral region of the valves where it is reflected by a shallow indentation in the margin, an immersed ligament set on a deeply sunken liga- mental groove, and a pseudolunule (Keen, 1962). Along the Pacific coast of North America the genus ranges from the middle Oligocene “Lincoln Stage” or Refugian Stage to the middle Miocene “Temblor Stage.” Its geographic distribution during the Oligocene and the early Miocene extended from SYSTEMATIC DESCRIPTIONS 35 southern California (Kleinpell and Weaver, 1963; Loel and Corey, 1932) to southwestern British Co- lumbia (Clark and Arnold, 1923). The highest stratigraphic occurrence, in strata of middle Mio- cene age, is in northern California where it occurs in the Sobrante Sandstone of the San Francisco Bay area (Lutz, 1951). Localities—USGS 5149, M3280, M3281, M3578, M3772, M3978, M3979, M3984, M4466, M4468, M4469, M4470, M4471, M4472, M4473. Family VENERIDAE Genus AMIANTIS Carpenter, 1864 Amiantis mathewsoni (Gabb) Plate 7, figures 1, 4, 8, 13, 14 Chione mathewsom‘i Gabb, California Geol. Survey, Paleontology, v. 2, sec. 1, pt. 1, p. 23, pl. 5, fig. 39. Antigona. (Artena) mathewsonii (Gabb). Clark, Cali- fornia Univ. Pubs., Dept. Geology Bull., v. 11, no. 2, p. 143—144, pl. 15, figs. 5—7. Antigone: (Artena) neglecta Clark, California Univ. Pubs., Dept. Geology Bull., v. 11, no. 2, p. 144, pl. 8, figs. 3, 4, 6. Antigona (Ventricola) undosa Clark, California Univ. Pubs., Dept. Geology Bull., v. 11, no. 2, p. 145, pl. 8, figs. 1, 2, 5. Amiantis? mathewsonii (Gabb). Stewart, Acad. Nat. Sci. Philadelphia Spec. Pub. 3, p. 245—246, pl. 14, fig. 2. Amiantisfl) mathewsom'i (Gabb). Kleinpell and Weaver, California Univ. Pubs. Geol. Sci., v. 43, p. 204, pl. 35, figs. 8, 9. Lectotype.—ANSP 4532 (designated by Stewart, 1930). Type locality.—Miocene(?) Martinez, Calif. Pre- sumably from the San Ramon Sandstone. This species occurs abundantly in most of the col- lections from the Wygal Sandstone Member of the Temblor Formation. Well-preserved specimens are characterized by coarse, somewhat irregular con- centric lamellae between which are very fine con— centric threads. The coarse lamellae are sharp edged and closely spaced anteriorly, posteriorly, and ven- trally. In the middle part of the valve, these lamellae terminate in a sharp edge that is reflected dorsally. There is a well~developed escutcheon that is bor- dered, on the left valve, by a strong cord. The pallial sinus is deep, and the apex reaches the anterior half of the valve. Assignment to Amitmtis has been questioned by some previous workers (Stewart, 1930, 1946; Klein- pell and Weaver, 1963), possibly because the nature of the pallial sinus and the dentition of the right valve were not known. The pallial sinus is similar to that of the type of Amiantis, A. callosa (Conrad) but differs in having a less sharply pointed apex. There are anterior lateral teeth in both valves; the 1866. 1918. ?1918. ?1918. 1930. 1963. right anterior lateral, however, is rather weak. The right anterior cardinal and the middle cardinal are sharply bifid (pl. 7, fig. 13), differing from A. callosa. A wide range of variation in the shape of the valves is exhibited by the specimens at hand. Most are subtrigonal to subquadrate depending on the convexity of the posterior dorsal slope. Specimens of Amiantis from USGS locality M3578 are especi- ally variable. They can be segregated into two taxa, one of which is relatively much more elongated posteriorly than A. mathewsoml (pl. 7, figs. 6, 7) and may represent a new species. The range of variation in shape and sculpture within individual collections seems sufficiently broad to regard Clark’s (1918) species Antigona, neglecta and A. undosa as probable synonyms of A. mathew- som’ with which they occur in the San Ramon Sand- stone of Contra Costa County, Calif, the probable formation from which the latter species was de- scribed by Gabb (1866). Records of this species from the middle Miocene “Temblor Stage” (Schenck and Keen, 1940; Stew- art, 1946; Adegoke, 1969) are doubtful. Schenck and Keen’s record from the Olcese Sand of the east— ern margin of the San Joaquin basin are based on a relatively smooth, subquadrate taxon Whose denti- tion is entirely different from Amiantis mathewsoni. The middle Miocene taxon (pl. 7, figs. 2, 3, 11) has a much narrower hinge plate and a broad, rather than bladelike middle cardinal in the right valve, and the right posterior cardinal is not sharply bifid as on A. mathewsom’. The specimens from the Olcese Sand are similar to A. diabloensis (Anderson, 1905) but seem to represent an undescribed species be- cause of the much heavier middle and posterior cardinal teeth in the left valve and the more elon- gate, less tumid valves. A doubtful record from the Temblor Formation on Reef Ridge, Fresno County, Calif. (Stewart, 1946, USGS loc. 14402), is of a relatively smooth, quadrate specimen whose partially exposed hinge plate is much narrower anteriorly than on A. mathewsom’. Adegoke’s (1969) unfigured hypotype from the Temblor Formation at Reef Ridge (UCMP 32532) is a poorly preserved specimen that is generically indeterminate. Amiantis mathewsoni ranges from the middle Oligocene to the late Oligocene or early Miocene. Its lowest stratigraphic occurrences are in the mid- dle part of the Gaviota Formation of Kleinpell and Weaver (1963) and the upper part of the San Emigdio Formation of Wagner and Schilling (1923). The highest occurrence is in the San 36 MOLLUSCAN BIOSTRATIGRAPHY, PALEONTOLOGY, TYPE TE‘MBLOR FORMATION, CALIF. Ramon Sandstone of Clark (1918), which may be as young as early Miocene in the sense of the Pa- cific megafaunal chronology of Weaver and others (1944). There are no authenticated records, how— ever, in the lower Miocene Vaqueros Formation of Loel and Corey (1932). Localities—USGS 5149, M3280, M3578, M3579, M3772, M3972, M3978, M3979, M3984 cf., M4466?, M4468 cf., M4469 cf. Aminntis n. 51).? Plate 6, figure 6; plate 7, figures 6, 7 Several poorly preserved specimens in the col- lection from USGS locality M3578 have exception- ally elongate valves and a relatively long posterior dorsal slope. Although the valves appear to be some- what decorticated and are clearly thinner than on A. mathewsom', the sculptural pattern is similar as are details of the escutcheon, lunule, and pallial sinus. There are several other specimens in this collection that are clearly referable to A. mathew- som'. They are variable in outline but do not seem to be intergradational with the elongate specimens that are here doubtfully considered to represent a new species. The hinge of this taxon is not known, but the pallial sinus is typical of Amiantis; it is deep and consists of a linear dorsal limb and slightly concave ventral limb (pl. 7, fig. 6). The apex, as in A. mathewsom', is not sharply pointed. Localities—USGS M3578, M3772?, M4466. Genus CLEMENTIA Gray, 1842 Subgenus EGESTA Conrad, 1845 Clemenlia (Egesta) pertenuis (Gabb) Plate 7, figures 9, 12; plate 8, figures 1, 3, 19 1866. Venus kennerlyi Reeve?. Gabb, California Geol. Sur- vey, Paleontology, v. 2, sec. 1, pt. 1, p. 22, pl. 5, fig. 37. 1869. Venus pertenuis Gabb, California Geol. Survey, Paleon— tology, v. 2, sec. 1, pt. 2, p. 55—56, pl. 5, fig. 37. 1910. Venus pertenuis Gabb. Arnold, U.S. Geol. Survey Bull. 396, pl. 8, fig. 3. [Imprint 1909.] 1910. Venus pertenuis Gabb. Arnold and Anderson, U.S. Geol. Survey Bull. 398, pl. 30, fig. 3. 1926. Clementia (Egesta) perteuuis (Gabb). Woodring, U.S. Geol. Survey Prof. Paper 147—C, p. 40—42, pl. 16, figs. 1—6. 1940. Clementia pertenuis (Gabb). Schenck and Keen, Cali- fornia fossils for the field geologist, pl. 34, figs. 1, 2. 1958. Clementia perteuuis (Gabb). Hall, California Univ. Pubs. Geol. Sci. Bull., v. 34, no. 1, p. 54, pl. 3, fig. 2. Type.—UCMP 12000. Type locality—At Griswold’s [Ranch] on road to New Idria, San Benito County, Calif. Temblor For- mation, Miocene. Specimens of Clementia perteuuis from the Wygal Sandstone Member are subquadrate to subtrigonal in outline. They are sculptured by broad concentric folds on the dorsal part of the valves that grade, ventrally, into fine, irregular lines of growth. This is the initial record of the genus Clementia from strata of Oligocene age along the Pacific coast of North America. The lowest previously recorded stratigraphic occurrence in this region is from strata assigned to the provincial lower Miocene (Woodring, 1926; Cushman and LeRoy, 1938). A late Oligocene occurrence from the Pacific coast of South America has been noted, however, by Wood- ring (1926, p. 31). Localities—USGS 5149, 9432, M3280, M3578, M3579, M3772, M3978, M4466, M4468, M4469. Genus DOSINIA Scopoli, 1777 Subgenus DOSINIA s. s. Dosinia (Dosinia) whitneyi (Gabb) Plate 6, figures 1, 3, 9, 12, 13; plate 7, figures 5, 10 1866. Chione whitneyi Gabb, California Geol. Survey, Paleon- tology, v. 2, sec. 1, pt. 1, p. 23—24, pl. 5, fig. 40. 1866. Dosim’a mathewsonii Gabb, California Geol. Survey, Paleontology, v. 2, p. 57, pl. 15, fig. 16. 1918. Dosiuia (Dosinidia) mathewsom' Gabb. Clark, Cali- fornia Univ. Pubs., Dept. Geology Bull., v. 11, no. 2, p. 141—142, pl. 7, figs. 1, 2, 5, 6, 9. 1918. Dosim'a (Dosinidiu) whitneyi (Gabb). Clark, Cali- fornia Univ. Pubs., Dept. Geology Bull., v. 11, no. 2, p. 143, pl. 7, figs. 3, 4. 1930. Dosiuia mathewsom‘i Gabb. Stewart, Acad. Nat. Sci. Spec. Pub. 3, p. 230—232, pl. 14, fig. 7. 1951. Dosim‘a (Dosinidia) whitneyi (Gabb). Lutz, California Univ. Pubs., Dept. Geol. Sci. Bull., v. 28, no. 13, p. 389—390, pl. 16, figs. 1, 2. 1959. Dosim'a (Dosim'dia) mathewsonii Gabb. Durham, Veli- ger, v. 2, no. 2, pl. 4, figs. 2, 3. 1964. Dosiuia (Dosim'a) whitneyi (Gabb). Moore, U.S. Geol. Survey Prof. Paper 419, p. 73—74, pl. 24, figs. 1—10. [Imprint 1963.] 1972. Dosinia whitneyi (Gabb). Addicott, Soc. Econ. Paleon- tologists and Mineralogists, Pacific Sec., Bakersfield, Calif., 1972, Proc., pl. 1, fig. 4. Type.-—UCMP 11999. Type locality—Near Martintez Creek, Contra Costa County, California. San Ramon Sandstone, early Miocene( ?). Types of Dosinia mathewsonii Gabb.—ANSP 4485 (syntypes). Type locality—Probably from the San Ramon Sandstone near Walnut Creek, Contra Costa County, Calif. (Stewart, 1930, p. 231). Discrimination of Dosiuia mathewsom’ Gabb (1866), also from the San RamonKSandstone, from D. whitneyi principally on the basis of a more angu- lar dorsal margin (Clark, 1918; Adegoke, 1969) seems to have been made on comparison of differ- ently oriented figures. Moore’s (1963) illustrations SYSTEMATIC DESCRIPTIONS 37 of the type of D. whitneyi and of one of Clark’s (1918) specimens of D. mathewsom’ indicate that one, rather than two, species is represented. Dosimla whitneyi is represented by poorly pre- served specimens from the so-called Phacoides Sand. These vary somewhat in the degree of inflation of the valves and in the angularity of the dorsal mar- gin. The exterior is well preserved only on a few of the smaller specimens (pl. 7, fig. 5). This species ranges from the upper Oligocene to the middle Miocene in California. During the middle Miocene, it ranged as far north as northwestern Washington (Arnold and Hannibal, 1913). Localities—USGS 9432, M3280, M3578, M3772, M3979, M4466, M4468 ?, M4469 cf., M4470 ?, M4471, M4472. Genus PITAR Rfimer, 1857 Pitar 3]). Plate 6, figure 10 An internal mold from USGS locality M3978 may represent Pitar. It is very strongly inflated and has a subquadrate outline that is more nearly equidi- mensional than P. (Katherinella) cf. P. (K.) cali- form'ca. It might represent P. lorenztma (Clark, 1918, p. 147—148, pl. 10, figs. 2, 4), a comparably large species from the San Ramon Sandstone, but it appears to have a much less convex posterior dorsal slope and to be more produced anteriorly. Locality—USGS M3978. Subgenus KATHERINELLA Tegland, 1929 Pitar (Katherinella) cf. P. (K.) californica (Clark) Plate 6, figures 8, 11 Several poorly preserved specimens from locali- ties in the Wygal Sandstone Member are similar to Pitar califomica. (Clark, 1918, pl. 11, figs. 2, 3, 4, 11) from the San Ramon Sandstone of Contra Costa County, Calif. The figured specimens differ, how- ever, in being relatively more produced anteriorly. Both specimens are more elongate than any species described from the middle Tertiary of the Pacific Coast States. Assignment of this species to Katherinella is sug- gested by the pouting lunule that is set off by an incised line (Clark, 1918, p. 148), a feature similar to that on specimens of Pitar (Katherinella) angus- tifrons (Conrad) from the Astoria Formation of coastal Oregon (USGS loc. M2116). A similar, faintly incised line occurs on one of the specimens from USGS locality M3280. Localities—USGS 5149?, M3280, M3772?, M4466, M4472. 493-696 0 — 73 - 4 Genus SECURELLA Parker, 1949 Securella cf. S. cryptolineata (Clark) Plate 6, figure 5 A poorly preserved specimen of Securella in the collection from USGS locality M3280 resembles. very closely, some of the topotypes of S. crypto- lineata (Clark) described and figured by Parker (1949, p. 589—590, pl. 94, figs. 10, 11, 13) from the San Ramon Formation of the Mount Diablo area, California. The specimen is heavy shelled and has a relatively broad hinge plate. Although the surface is deeply eroded and the consequent pattern of widely spaced ribs does not appear to be that of Securellu, there is a small area near the anterior margin of the shell on which the original surface sculpture of closely spaced concentric lamellae char- acteristic of this genus is preserved. This species has been doubtfully identified from the presumably correlative upper part of the Pleito Formation of Wagner and Schilling (1923) of the San Emigdio Mountains at the southern margin of the San Joaquin basin (Wagner and Schilling, 1923, p. 245—246) as Chione cf. C. lineolata Clark. Locality.———USGS M3280. Order MYOIDA Family MYlDAE Genus PANOPEA Menard, 1807 Panopea ramonensis (Clark) Plate 8, figure 20 Panope ramonensis Clark, California Univ. Pubs., Dept. Geol. Sci. Bull., v. 15, no. 4, p. 106, pl. 10, figs. 2, 3. Panope ramonensis Clark. Clark, Stratigraphy and fauna] horizons of the Coast Ranges of California, pl. 15, figs. 5, 7. Panope ramonensis Clark. Weaver, Washington Univ. [Seattle] Pubs. Geology, v. 5, p. 263—264, pl. 59, fig. 11. Panopea (Panopea) ramonensis Clark. Hickman, Ore- gon Univ. Mus. Nat. History Bull. 16, p. 65—66, pl. 8, figs. 8, 12. 1925. 1929. 1942. 1969. Types. — Holotype, UCMP 30331. Type locality—UCMP 1131, 1/2 mile southwest of Walnut Creek, Contra Costa County, Calif. in creek bed about 100 yards east of Oakland and Antioch bridge. Altitude 150 feet. San Ramon Sandstone, early Miocene(?). An elongate Panopea from the Wygal Sandstone Member of the Temblor Formation differs from the common middle and late Cenozoic species P. abrupta (Conrad) by having consistently longer valves. Some of the specimens have medially situated beaks as on the holotype from the San Ramon Sandstone. On other specimens (pl. 8, fig. 20), the beaks are UCMP 30330; paratype, 38 MOLLUSCAN BIOSTRATIGRAPHY, PALEONTOLOGY, TYPE TEMBLOR FORMATION, CALIF. located anterior to the middle of the valves. Speci- mens from USGS locality M3280 have a faint sul- cus extending from near the beaks to the posterior ventral margin similar to the sulcus on the para- type of P. ramonensis (Clark, 1925, pl. 10, fig. 2). Panopea estrellana (Conrad, 1857b, p. 194, pl. 7, fig. 5), a late Miocene species from the California Coast Ranges, is less elongate than P. ramonensis. Although recognized by some workers as a separate and distinct species (Durham, 1944; Adegoke, 1969), it seems to be best treated as a synonym of P. abrupta as concluded by Moore (1963). This species ranges from strata of middle Oligo- cene (Primmer, 1964) to middle Miocene (Weaver, 1942) age on the Pacific‘coast. It may be repre- sented in the Pleito Formation of Wagner and Schilling (1923, check list 2) of the southern mar- gin of the San Joaquin basin by their Panope cf. P. estrellana. Localities—USGS 5149, M3280, M3578, M3772, M3978, M4466?, M4469, cf., M4473. Class SCAPHOPODA Family DENTALHDAE Genus DENTALIUM Linng, 1758 Dentalium laneensis Hickman Plate 8, figure 14 1969. Dentalium (?Fissidentalium) laneensis Hickman, Ore- gon Univ. Mus. Nat. History Bull. 16, p. 74, pl. 9, figs. 1—6. 1972. Dentalium laneensis Hickman. Addicott, Soc. Econ. Paleontologists and Mineralogists Pacific Sec., Bakersfield, Calif., 1972, Proc., pl. 1, fig. 9. Types.—Holotype, UO 27332; paratypes, U0 27333—27339. Type locality.—UO 2538, roadcut at 30th Ave. and Agate St., Eugene, Oreg. (NW1/4 sec. 8, T. 18 S., R. 3 W., Eugene 15’ quad) Eugene Formation, Oligocene. Dentalium laneensis is represented by a few in- complete specimens from the Wygal Sandstone ‘lember. They are characterized by a relatively arge, thick shell sculptured by numerous broad ongitudinal ribs betWeen which are intercalated iner secondary ribs. D. radiolineata Clark (1918, p. 191, pl. 22, fig. 12) from the early Miocene(?) San Ramon Sandstone of central California differs from D. laneensis in having fewer radial ribs that are narrow and crested in cross section. This is the only record of this species from Cali- fornia as well as the only record from strata of late Oligocene age, previous records being from the lower and middle Oligocene Eugene Formation of western Oregon. The species also occurs in the middle Oligocene Pittsburg Blufi‘ Formation of western Oregon (E. J. Moore, written commun., Oct. 1971). Localities.—USGS M3280, M3281, M3578, M3772, M3978. FOSSIL LOCALITIES USGS localities (Washington, D.C., register) Locality Description 4941. Three-fourths of a mile north of Miller Brothers ranch house beside their grade road, 6 miles east of Annette [south side of road between Franciscan Creek and Miller Flats, about 200 ft south of ridge crest; SWV;r SW14 sec. 36, T. 26 S., R. 17 E., Packwood Creek 71/2' quad] VaquerOS Sandstone [Carneros Sandstone Member(?) of the Temblor Formation, about 1,000 ft stratigraphically below the base of the Media Shale Member.] Same as USGS 10c. M2633. Collected by Ralph Arnold, 1905. Three-fourths of a mile southwest of Carneros Spring, SW14 sec. 5, T. 29 S., R. 20 E. Vaqueros Sandstone [Wygal Sandstone Member of the Temblor Formation]. Same as USGS loc. M4469. Collected by Ralph Arnold, 1908. North bank of Kern River, 10—11 miles northeast of Bakersfield. Float rock at base of bluff below where fossils were collected at USGS Ice. 6623. Middle part of lower Miocene. These fossils are unquestionably from beds in the cliffs at whose foot they lie and are from the same horizon as those of USGS 10c. 6623 [lower part of the Round Mountain Silt, middle Miocene]. Collected by R. W. Pack and J. D. Northrop, October 1911. Twelve miles N. 30° E. of Bakersfield. In center Wl/z sec. 36, T. 27 S., R. 28 E. In small arroyo tributary to Adobe Canyon from west. About 1 mile above its mouth. In first arroyo upstream from 1,070-ft hill. Middle part of lower Miocene [lower part of the Olcese Sand, middle Miocene]. Collected by R. W. Pack, Sept. 2, 1911, and A. T. Schwennessen, Aug. 31, 1911. Basal beds of the Vaqueros [Temblor Formation], on hill slope just south of Carneros Spring near center of sec. 5, T. 29 S., R. 20 E. Collected by Walter English, 1916. On east side of creek near streambed, about 125 ft east of bluff of Point of Rocks Sandstone. Near center of west line of SE14 sec. 18, T. 29 S., R. 20 E. Basal Vaqueros Sandstone [Wygal Sand- stone Member of the Temblor Formation]. Col- lected by W. S. W. Kew, 1916. USGS Cenozoic localities (Menlo Park, Calif., register) M1596. Pelecypod biostrome on east side of southerly trend- ing canyon, 1,400 ft north, 100 ft west of SE. cor. sec. 24, T. 28 S., R. 28 E., Oil Center quad. (1954 ed.). Near the top of the Olcese Sand. Same as UCMP Ice. 1603. Collected by W. O. Addicott, 1962. M1599. Bottom of southeasterly trending gully 900 ft south, 150 ft east of NW. cor. sec. 33, T. 28 S., R. 29 E., Rio Bravo Ranch quad. (1954 ed.). Near the top of the Olcese Sand, 17 ft stratigraphically below USGS loc. M1600. Same as UCMP loc. B1599. Collected by W. 0. Addicott, 1962. 5149. 6622. 6627. 9427. 9432. Locality M1802. M2544. M2631. M2632. M2633. M3280. M3281. M3511. M3578. M3579. FOSSIL LOCALITIES Description Seaclifl" exposure north of Marine Gardens, a locally named point east of Gull Rock, 350 ft south, 3,900 ft west of NE. cor. sec. 32, T. 9 S., R. 11 W., Cape Foulweather 15’ quad. Oregon. Astoria Formation. Collected by W. 0. Addicott, 1963. Intertidal and spray zone exposure at mouth of Kirby Creek about 7,200 ft west of provincial highway 14 bridge over Muir Creek, Sooke 92 B/5 quad. Renfrew District, Vancouver Island, British Columbia. Sooke Formation. Collected by W. O. Addicott, 1965. About one-fourth of a mile south of the mouth of Cedar Canyon on east side of road, 1,800 ft south, 1,550 ft east of NW. cor. sec. 28, T. 27 S., R. 18 E., Packwood Creek 71/2' quad. Agua Sandstone Member of the Temblor Formation, basal con- glomeratic sandstone. Collected by W. O. Addi- cott, 1965. West-trending ridge south of Bitterwater Creek, 600 ft south, 1,400 ft east of NW. cor. sec. 27, T. 27 S., R. 18 E., Packwood Creek 714’ quad. Alt. 1,820 ft. Buttonbed Sandstone Member of the Temblor Formation. Collected by W. O. Addi- cott, 1965. On south side of Franciscan Creek—Miller Flats road, 600 ft north, 1,000 ft east of SW. cor. sec. 36, T. 26 S., R. 17 E., Packwood Creek 714' quad. Carneros(?) Sandstone Member of the Temblor Formation, about 1,000 ft stratigraphically below the base of the Media Shale Member. Same as USGS 10c. 4941. Collected by T. W. Dibblee, Jr., and W. O. Addicott, 1965. Minor saddle in northeast-trending ridge on south side of Media Agua Creek, 900 ft south, 500 ft east of NW. cor. sec. 23, T. 28 S., R. 19 E., Las Yeguas Ranch 71/2' quad. Wygal Sandstone Mem- ber of the Temblor Formation, 6-ft—thick fos- siliferous sandstone un'conformably overlying the Eocene Point of Rocks Sandstone. Collected by Howard Sonneman and W. O. Addicott, 1967; H. C. Wagner, Saburo Kanno, and W. O. Addicott, 1969; W. O. Addicott, 1971. On south side of hill 2259 just below summit, 2,200 ft north, 1,000 ft west of SE. cor. sec. 22, T. 28 S., R. 19 E., Las Yeguas Ranch 71/! quad. Base of Wygal Sandstone Member of the Temblor Formation. Collected by Howard Sonneman and W. O. Addicott, 1967. 8,900 ft south, 2,900 ft west of lat 35°00' N., long 119°32’30” W. East of Cuyama Ranch 71/2’ quad. Alt about 3,250 ft. Quail Canyon Sandstone Mem- ber of the Vaqueros Formation. Collected by J. G. Vedder, 1960. About 30—40 ft above unnamed stream flowing south- westward through the SW14 sec. 15, T. 29, S., R. 20 E., 1,100 ft north, 2,200 ft west of SE. cor. sec. 15, T. 29 S., R. 20 E., Carneros Rocks 7%' quad. Wygal Sandstone Member of the Tem- blor Formation, collection from about 200 ft along strike in southwest-trending gullies. Col- lected by W. O. Addicott, May 1967. Northeast bank of southeast-flowing stream about 15 ft above bottom, 2,200 ft south 850 ft east of Locality M3636. M3747. M3772. M3978. M3979. M3981. M3982. M3984. M3985. 39 Description NW. cor. sec. 15, T. 29 S., R. 20 E., Carneros Rocks 7%’ quad. Wygal Sandstone Member of the Temblor Formation. Collected by W. O. Addicott, 1967. North side of Chico Martinez Creek, approximately 2,100 ft west, 250 ft south of NE. cor. sec. 8, T. 29 S., R. 20, E., Carneros Rocks 71/2’ quad. Wygal Sandstone Member of the Temblor Forma- tion. Collected by R. L. Pierce and T. W. Dibblee, Jr., 1967; W. O. Addicott, 1971. East-west-striking pelecypod biostrome on hillside west of Salt Creek, 850 ft north, 1,850 ft east of SW. cor. sec. 21, T. 10 N., R. 20 W., Pleito Hills 71/2’ quad. Pleito Formation of Wagner and Schil- ling (1923). Collected by W. O. Addicott, J. G. Vedder, and T. W. Dibblee, Jr., 1966. On southeast side of ridge paralleling Media Agua Creek about 200 feet above dirt road, 1,025 ft south, 200 ft east of NW. cor. sec. 23, T. 28 S., R. 19 E., Las Yeguas Ranch 71/2’ quad. Wygal Sandstone Member of the Temblor Formation. Same as USGS loc. M3280. Collected by W. 0. Addicott, 1968. In bed of Temblor Creek, 1,000 ft south, 1,900 ft east of NW. cor. sec. 25, T. 29 S., R. 20 E., Carneros Rocks 71/2' quad. Alt about 1,350 ft. Fossiliferous concretions at the top of the Wygal Sandstone Member of the Temblor Formation. Collected by Ii. S. Sonneman and W. O. Addicott, 1968. Top of prominent ledge above southeast-trending segment of first creek south of Zemorra Creek, 750 ft north, 2,750 ft west of SE. cor. sec. 16, T. 29 S., R. 20 E., Carneros Rocks 7%’ quad. Alt about 2,450 ft. Wygal Sandstone Member of the Temblor Formation, collection from about 300 ft along the strike of a concretionary stratum about 20 ft above base of the Wygal. Collected by H. S. Sonneman and W. O. Addicott, 1968. On southwest side of northwest-trending ridge, 2,300 ft north, 800 ft west of SE. cor. sec. 25, T. 28 S., R. 19 E., Carneros Rocks 71/2' quad. Alt about 1,460 ft. Lower part of the Carnerog Sandstone Member of the Temblor Formation. Collected by H. S. Sonneman and W. 0. Addi- cott, 1968. In bottom of north fork of Santos Creek, 1,300 ft north, 300 ft west of SE. cor. sec. 25, T. 28 S., R. 19 E., CarnerOS Roeks 71/2' quad. Alt about 1,300 ft. Agua Sandstone Member of the Temblor Formation. Collected by H. S. Sonneman and W. O. Addicott, 1968; W. O. Addicott, 1971. Southwest side of small knoll encircled by 1,400-ft contour, 1,600 ft north of SE. cor. sec. 23, T. 28 S., R. 19 E., Las Yeguas Ranch 71/2' quad. Alt about 1,330 ft. Basal 10 ft of the Wygal Sandstone Member of the Temblor Formation; collection from along 300 to 400 feet of strike of beds. Collected by H. S. Sonneman and W. O. Addicott, 1968. On south side of Stone Corral Creek, 600 ft south, 1,900 ft east of NW. cor. sec. 25, T. 28 S., R. 19 E., Las Yeguas Ranch 71/! quad. Alt about 1,330 40 Locality M3986. M3987. M3988. M4060. M4464. M4466. M4467. M4468. M4469. MOLLUSCAN BIOSTRATIGRAPHY, PALEONTOLOGY, TYPE TEMBLOR FORMATION, CALIF. Description ft. Wygal Sandstone Member of the Temblor Formation, basal 6—10 ft. Collected by H. S. Son- neman and W. 0. Addicott, 1968. In bottom of gulch northeast of old side of Mac- Donalds Ranch (see USGS Bull. 406), 2,600 ft north 750 ft east of SW. cor. sec. 32, T. 27 S., R. 19 E., Shale Point 71/2' quad. Alt about 1,235 ft. About 40 ft stratigraphically above the base of the Buttonbed Sandstone Member of the Tem- blor Formation. Collected by H. S. Sonneman and W. O. Addicott, 1968. Hillside west-southwest of stream junction, 500 ft north, 2,700 ft east of SW. cor. sec. 30, T. 27 S., R. 19 E., Shale Point 71/2' quad. Alt about 1,350 ft. Buttonbed Sandstone Member of the Temblor Formation, brown sandstone concretions occurring about 30 ft above the base of the member. Col- lected by H. S. Sonneman and W. 0. Addicott, 1968. Hillside west-southwest of stream junction, 550 ft north, 2,600 ft east of SW. cor. sec. 30, T. 27 S., R. 19 E., Shale Point 71/2’ quad. Alt about 1,375— 1,950 ft. Buttonbed Sandstone Member of the Temblor Formation, basal 5 ft. Collection from about 150 ft along strike of beds. Collected by H. S. Sonneman and W. O. Addicott, 1968. Abundantly fossiliferous conglomeratic sandstone exposed in low sea cliff and isolated rock on east side of mouth of Kirby Creek, in land parcel 52, Sooke 92 B/5 quad., Renfrew District, Vancouver Island, British Columbia. Sooke Formation. Col- lected by W. O. Addicott, 1968. On northwest plunge of small northwest-trending ridge, 1,250 ft north, 1,350 ft east of SW. cor. sec. 24, T. 28 S., R. 19 E., Las Yeguas Ranch 71/2’ quad. Agua Sandstone Member of the Tem- blor Formation, Crassostrea bed. Collected by H. S. Sonneman and W. O. Addicott, 1968. East wall of north-northwest—trending tributary to Zemorra Creek, about 50 ft above gully bottom, 400 ft north, 1,200 ft west, of SE. cor. sec. 9, T. 29 S., R. 20 E., Carneros Rocks 71/2' quad. Wygal Sandstone Member of the Temblor Forma— tion, about 10—30 ft above the base. Collected by W. 0. Addicott, 1971. In north-northwest-trending tributary to Zemorra Creek, 250 ft south, 950 ft west of NE. cor. sec. 16, T. 29 S., R. 20 E., Carneros Rocks 71/2' quad. Wygal Sandstone Member of the Temblor Forma- tion, basal 3 ft. Collected by W. O. Addicott, 1971. Southwest-facing hillside on east flank of Cameros Rocks anticline, 3,100 ft north, 800 ft west of SE. cor. sec. 5, T. 29 S., R. 20 E., Cameras Rocks 71/2' quad. Wygal Sandstone Member of the Tem- blor Formation, lowest 20 ft. Collected by W. O. Addicott, 1971. Crest of east-trending ridge on north side of Chico Martinez Creek, 350 ft north, 1,200 ft east of SW. cor. sec. 5, T. 29 S., R. 20 E., Carneros Rocks 71/2' quad. Alt about 1,960 ft. Wygal Sandstone Member of the Temblor Formation. Same as USGS loc. 5149. Collected by W. O. Addicott, 1971. Locality Description M4470. East-northeast-trending ridge about one-fourth mile northwest of Carneros Creek, 450 ft north, 2,500 ft west of SE. cor. sec. 31, T. 28 S., R. 20 E., Carneros Rocks 71/2' quad. Alt about 1,580 ft. Wygal Sandstone Member of the Temblor Forma- tion. Collected by W. O. Addicott, 1971. M4471. South side of road along Carneros Creek, 2,200 ft south, 150 ft west of SE. cor. sec. 36, T. 28 S., R. 19 E. [in sec. 6, T. 29 S., R. 20 E.], Carneros Rocks 71/2’ quad. Wygal Sandstone Member of the Temblor Formation. Collected by W. O. Addicott, 1971. M4472. Near crest of 1,480-ft knoll on north side of Santos Creek, 550 ft north, 1,100 ft west of SE. cor. sec. 25, T. 28 S., R. 19 E., Carneros Rocks 71/2' quad. Wygal Sandstone Member of the Temblor Forma- tion. Collected by W. 0. Addicott, 1971. M4473. Crest of south-trending ridge on north side of Media Agua Creek (about 200 ft south of hill 1814), 550 ft north, 2,750 ft east of SW. cor. sec. 15, T. 28 S., R. 19 E., Las Yeguas Ranch 71/2’ quad. Wygal Sandstone Member of the Temblor Formation, basal 2—3 ft. Collected by W. O. Addi- cott, 1971. California University Museum of Paleontology (Berkeley) localities (UCMP) B1598. At mouth of short Y-shaped gully approximately 200 yds north of abandoned sand and gravel plant, 1,200 ft south, 300 ft west of of NE. cor. sec. 5, T. 29 S., R. 29 E., Oil Center quad. (1950 ed.). Gray clean fine to very fine sand. Upper part of the Olcese Sand. Collected by W. O. Addi- cott, 1953, 1954. B1599. In second southeast-trending gully due east of hill 933 (Bakersfield quad.), NW1/4NW1/i sec. 33, T. 28 S., R. 29 E., Caliente quad. (1914 ed.). Upper part of the Olcese Sand, approximately 80 ft stratigraphically below the base of the Round Mountain Silt. Same as USGS loc. M1599. Col— lected by W. O. Addicott, 1953, 1954. California University, Riverside, localities (UCR) 1106. About 150 ft east of creekbed on east side of small shoulder which forms a bend in the creek, NE. cor. SW14 sec. 15, T. 29 S., R. 20 E. Dark-brown- weathering, blue-gray limestone concre-tions in soft siltstone [Cymric Shale Member of the Tem- blor Formation]. About 10—30 ft stratigraphically below the “basal Miocene” [Wygal Sandstone Member of the Temblor Formation]. Collected by Eric Craig (field 10c. EC 89A), 1931. 1235. 150 ft north, 50 ft east of SW. cor. sec. 14, T. 29 S., R. 20 E. Wygal Sandstone Member of the Temblor Formation. Collection from a 50-ft strati- graphic interval directly overying soft gray silt- stone of the Cymric Shale Member. Collected by Eric Craig (field 10c. EC 89), 1931. REFERENCES Addicott, W. 0., 1956, Miocene stratigraphy northeast of Bakersfield, California: California Univ. (Berkeley), Ph. D. dissert., 207 p. 1966, New Tertiary marine mollusks from Oregon REFERENCES 41 and Washington: Jour. Paleontology, v. 40, no. 3, p. 635—646, 3 pls. 1967a, Age of the Skooner Gulch Formation, Mendo- cino County, California: U.S. Geol. Survey Bull. 1254—0, p. 01—011, 4 figs. 1967b, Zoographic evidence for late Tertiary lateral slip on the San Andreas fault, California: U.S. Geol. Survey Prof. Paper 593—D, p. D1—D12. 1969, Tertiary climatic change in the marginal north- eastern Pacific Ocean: Science, v. 165, no. 3893, p. 583— 586. 1970a, Miocene gastropods and biostratigraphy of the Kern River area, California: U.S. Geol. Survey Prof. Paper 642, 174 p., 21 pls. 1970b, Tertiary paleoclimatic trends in the San Joa- quin basin, California: U.S. Geol. Survey Prof. Paper 644—D, p. D1—D19. 1970c, Latitudinal gradients in Tertiary molluscan faunas of the Pacific coast: Palaeogeography, Palaeo- climatology, Palaeoecology, v. 8, no. 4, p. 287—312. 1972, Provincial middle and late Tertiary molluscan stages, Temblor Range, California: Soc. Econ. Paleon— tologists and Mineralogists, Pacific Sec., Pacific Coast Miocene Biostratigraphic Symposium, Bakersfield, Calif., 1972, Proc., p. 1—26, pls. 1—4, 1973, Oligocene and Miocene molluscan stages, Tem- blor Range, California [abs.]: Am. Assoc. Petroleum Geologists Bull. (In press.) Addicott, W. 0., and Vedder, J. G., 1963, Paleotemperature inferences from late Miocene mollusks in the San Luis Obispo—Bakersfield area, California: U.S. Geol. Survey Prof. Paper 475—C, p. CG3—C68. Adegoke, O. S., 1969, Stratigraphy and paleontology of the marine Neogene formations of the Coalinga region, California: California Univ. Pubs. Geol. Sci., v. 80, 241 p., 13 pls. Afshar, Freydoun, 1969, Taxonomic revision of the super- specific groups of the Cretaceous and Cenozoic Tellini- dae: Geol. Soc. America Mem. 119, 215 p., 45 pls. Allen, J. E., 1945, Geology of the San Juan Bautista quad- rangle, California: California Univ. (Berkeley), Ph. D. dissert. Anderson, D. N., and Land, P. E., 1969, Cymric oil field [California]: California Oil Fields, 'v. 55, no. 1, p. 5—21. Anderson, F. M., 1905, A stratigraphic study in the Mount Diablo Range of California: California Acad. Sci. Proc., 3d ser., v. 2, no. 2, p. 155—248, 23 pls. Anderson, F. M., and Martin, Bruce, 1914, Neocene record in the Temblor Basin, California, and Neocene deposits of the San Juan district, San Luis Obispo County: Cali- fornia Acad. Sci. Proc., 4th ser., v. 4, no. 3, p. 15—112, pls. 1—10. Arnold, Ralph, 1906, The Tertiary and Quaternary pectens of California: U.S. Geol. Survey Prof. Paper 47, 264 p., 53 pls., 2 figs. 1907, New and characteristic species of fossil mollusks from the oil-bearing Tertiary formations of Southern California: U.S. Natl. Mus. Proc., v. 32, no. 1545, p. 525—546, pls. 38—51. 1908, Descriptions of new Cretaceous and Tertiary fossils from the Santa Cruz Mountains, California: U.S. Natl. Mus. Proc., v. 34, no. 1617, p. 345-390, pls. 31—37. 1909, Paleontology of the Coalinga district, Fresno and Kings Counties, California: U.S. Geol. Survey Bull. 396, 173 p., 30 pls. [1910] Arnold, Ralph, and Hannibal, Harold, 1913, The marine Ter- tiary stratigraphy of the north Pacific Coast of America: Am. Philos. Soc. Proc., v. 52, no. 212, p. 559—605. Arnold, Ralph, and Johnson, H. R., 1910, Preliminary report on the McKittrick-Sunset oil region, Kern and San Luis Obispo Counties, California: U.S. Geol. Survey Bull. 406, 225 p. Bandy, O. L., and Arnal, R. E., 1969, Middle Tertiary basin development, San Joaquin Valley, California: Geol. Soc. America Bull., v. 80, no. 5, p. 783—820. Brabb, E. E., 1964, Subdivision of San Lorenzo Formation (Eocene—Oligocene), west—central California: Am. Assoc. Petroleum Geologists Bull., v. 48, no. 5, p. 670—679. Bruce, D. D., 1956, North Antelope Hills oil field [Cali- fornia]: California Oil Fields, v. 42, no. 2, p. 39—42. Burchfiel, B. G., 1964, Stratigraphic reassignment of four species in the lower Miocene rocks of the Bear Creek area, Santa Cruz County, California: Jour. Paleontology, v. 38, no. 2, p. 401—405. Clark, B. L., 1915, Fauna of the San Pablo Group of middle California: California Univ. Pubs., Dept. Geology Bull., v. 8, no. 22, p. 385—572, pls. 42—71. 1918, The San Lorenzo series of middle California: California Univ. Pubs., Dept. Geology Bull., v. 11, no. 2, p. 45—234, pls. 3—24. 1925, Pelecypoda from the marine Oligocene of west- ern North America: California Univ. Pubs., Dept. Geol. Sci. Bull., v. 15, no. 4, p. 69—136, pls. 8—22. 1932, Fauna of the Poul and Yakataga Formations (upper Oligocene) of southern Alaska: Geol. Soc. America Bull., v. 43, no. 3, p. 797—846, pls. 14—21. 1941, Notes on California Tertiary correlation: Cali- fornia Div. Mines Bull. 118, p. 187—191. Clark, B. L., and Arnold, Ralph, 1923, Fauna of the Sooke Formation, Vancouver Island, with description of a new coral by T. Wayland Vaughan: California Univ. Pubs., Dept. Geol. Sci. Bull., v. 14, no. 5, p. 123—234, pls. 15—42. Clark, L. M., and Clark, Alex, 1935, The Vaqueros in the Temblor Range [California] [abs.]: Am. Assoc. Petro- leum Geologists Bull., v. 19, no. 1, p. 137. Conrad, T. A., 1848, Fossil shells from Tertiary deposits on Columbia River, near Astoria [Oregon]: Am. Jour. Sci., 2d ser., v. 5, p. 432—433, 14 figs; reprinted in Dall, W. H., 1909, Contributions to the Tertiary paleontology of the Pacific coast—[Part] 1, The Miocene of Astoria and Coos Bay, Oregon: U.S. Geol. Survey Prof. Paper 59, p. 150—151. 1849, Fossils from northwestern America, in Dana, J. D., Geology, U.S. Exploring Expedition, during the years 1838, 1839, 1840, 1841, 1842, under the command of Charles Wilkes, U.S.N.: Philadelphia, v. 10, app. p. 722—728, pls. 17—21; reprinted in Dall, W. H., 1909, Contributions to the Tertiary paleontology of the Pacific coast—[Part] 1, The Miocene of Astoria and Coos Bay, Oregon: U.S. Geol. Survey Prof. Paper 59, p. 152—156. 1855, Report of Mr. T. A. Conrad on the fossil shells collected in California by W. P. Blake, geologist of the expedition, under the command of Lieutenant R. S. Williamson, U.S. Topographical Engineers, 1853, in appendix to Preliminary Geological Report of W. P. Blake: U.S. 33d Cong, lst sess., H. Ex. Doc. 129, p. 5—20; reprinted in Dall, W. H., 1909, Contributions to 42 MOLLUSCAN BIOSTRATIGRAPHY, PALEONTOLOGY, TYPE TEMBLOR FORMATION, CALIF. the Tertiary paleontology of the Pacific coast—[Part] 1, The Miocene of Astoria and Coos Bay, Oregon: U.S. Geol. Survey Prof. Paper 59, p. 163—171. 1857a, Descriptions of the fossil shells, in Williamson, R. S., Report of explorations in California for railroad routes to connect with the routes near the 35th and 32d parallels of north latitude: U.S. 33d Cong., 2d sess., S. Ex. Doc. 78 and H. Ex. Doc. 91, v. 5, pt. 2, app. art. 2, p. 317—329, pls. 2—9. [Pacific RR. Repts.] 1857b, Report on the paleontology of the survey, in Parke, J. G., Report of explorations for railroad routes from San Francisco Bay to Los Angeles, California, west of the Coast Range, and from the Pimas Villages on the Gila to the Rio Grande, near the 32d parallel of north latitude: U.S. 33d Cong., 2d sess., S. Ex. Doc. 78 and H. Ex. Doc. 91, v. 7, pl. 2, p. 189—196, pls. 1—10; reprinted in Dall, W. H., 1909, Contributions to the Tertiary paleontology of the Pacific coast—[Part] 1, The Miocene of Astoria and Coos Bay, Oregon: U.S. Geol. Survey Prof. Paper 59, p. 180-185. [Pacific RR. Repts.] Corey, W. H., 1954, Tertiary basins of southern California [Part] 8, in Chapter 3 of Jahns, R. H., ed., Geology of southern California: California Dept. Nat. Res., Div. Mines Bull. 170, p. 73—83, 9 figs. Cunningham, G. M., and Barbat, W. F., 1912, Age of pro- ducing horizon at Kettleman Hills, California: Am. Assoc. Petroleum Geologists Bull., v. 16, no. 4, p. 417— 421. Curran, J. F., 1943, Eocene stratigraphy of Chico Mar- tinez Creek area, Kern County, California: Am. Assoc. Petroleum Geologists Bull., v. 27, no. 10, p. 1361—1386. Cushman, J. A., and LeRoy, L. W., 1938, A microfauna from the Vaqueros Formation, lower Miocene, Simi Valley, Ventura County, California: Jour. Paleontology, v. 12, no. 2, p. 117—126. Dall, W. H., 1900, Synopsis of the family Tellinidae and of the North American species: U.S. Natl. Mus. Proc., v. 23, no. 1210, p. 285—326. 1904, Neozoic invertebrate fossils, a report on col- lections made by the expedition, in Harriman Alaska Expedition, Volume IV (Geology and Paleontology): New York, Doubleday, Page and 00., p. 99—122, pls. 9, 10; reprinted by Smithsonian Inst., 1910. DeLise, K. G., 1967, Biostratigraphy of the San Emigdio Formation, Kern County, California: California Univ. Pubs. Geol. Sci., v. 68, 67 p. Dibblee, T. W., Jr., 1950, Geology of southwestern Santa Barbara County, California—Point Arguello, Lompoc, Point Conception, Los Olivos, and Gaviota quadrangles: California Dept. Nat. Res., Div. Mines Bull. 150, 95 p. 1961, Geologic features of the San Emigdio Mountains, Kern County, California, in Geology and paleontology of the southern border of the San Joaquin Valley, Kern County, California, 1961 Spring Field Trip Guidebook: [Whittier, Ca1if.] Soc. Econ. Paleontologists and Min- eralogists, Pacific Sec., p. 2—25. 1966, Geology of the Palo Alto quadrangle, Santa Clara and San Mateo Counties, California: California Div. Mines Map Sheet 8. 1968, Regional geologic map of San Andreas and re- lated faults in Carrizo Plain, Temblor, Caliente, and La Panza Ranges and vicinity, Santa Barbara, San Luis Obispo, Kern, and Kings Counties, California: U.S. Geol. Survey open-file map, scale 1:125,000. 1973, Stratigraphy of the southern Coast Ranges near the San Andreas fault, from Cholame to Maricopa, California: U.S. Geol. Survey Prof. Paper 764 (In press.) Durham, J. W., 1944, Megafaunal zones of the Oligocene of northwestern Washington: California Univ. Pubs., Dept. Geol. Sci. Bull., v. 27, no. 5, p. 101—211, pls. 13—18. 1950, Cenozoic marine climates of the Pacific coast: Geol. Soc. America Bull., v. 61, no. 11, p. 1243—1264. 1954, The marine Cenozoic of southern California, [Part] 4, in Chapter 3 of Jahns, R. H., ed., Geology of southern California: California Dept. Nat. Res., Div. Mines Bull. 170, p. 23—31. Durham, J. W., and Allison, E. C., 1960, The geologic history of Baja California [Mexico] and its marine faunas, in Symposium on The biogeography of Baja California and adjacent seas—Part 1, Geologic history: Systematic Zoology, v. 9, no. 2, p. 47-91. Edwards, K. L., 1940, Stratigraphic occurrence of Pecten sanctaecrunnsis in California [abs.]: Geol. Soc. America Bull., v. 51, no. 12, pt. 2, p. 1983. Eflinger, W. L., 1935, Gaviota formation of Santa Barbara County [California] [abs.]: Pan—Am Geologist, v. 64, no. 1, p. 75—76. Elliott, W. J., Tripp, Eugene, and Karp, S. E., 1968, Road guides, west side oil fields and Temblor Range outcrop area, in Geology and oil fields, west side southern San Joaquin Valley—AAPG, SEG, and SEPM, Pacific Secs., 43d Ann. Mtg., Bakersfield, Calif., 1968, Guidebook: p. 104—130. English, W. A., 1921, Geology and petroleum resources of northwestern Kern County, California: U.S. Geol. Sur- vey Bull. 721, 48 p., 2 pls., 2 figs. Etherington, T. J., 1931, Stratigraphy and fauna of the Astoria Miocene of southwest Washington: California Univ. Pubs., Dept. Geol. Sci. Bull., v. 20, no. 5, p. 31—142, 14 pls. Foss, C. D., and Blaisdell, Robert, 1968, Stratigraphy of the west side, southern San Joaquin Valley, in Geology and oil fields, west side southern San Joaquin Valley— AAPG, SEG, and SEPM, Pacific Secs., 43d Ann. Mtg., Bakersfield, Calif., 1968, Guidebook: p. 33—43. Fritsche, A. E., 1967, Taxonomy and distribution of some southern California Miocene mollusks [abs.]: Geol. Soc. America Ann. Mtg., Cordilleran Sec., 63d, Santa Barbara, Calif., March 22—25, 1967, Program, p. 35—36. Fullmer, C. V., 1965, Stratigraphy and paleontology of the type Blakeley Formation [Washington] [abs.]: Pacific Petroleum Geologist, v. 19, no. 10, p. 4. Gabb, W. M., 1864, Description of the Cretaceous fossils: California Geol. Survey, Paleontology, v. 1, sec. 4, p. 55—243, pls. 9—32. 1866, Tertiary invertebrate fossils: California Geol. Survey, Paleontology, v. 2, sec. 1, pt. 1, p. 1—38, pls. 1—13. Gester, G. C., and Galloway, John 1933, Geology of Kettle- man Hills oil field, California: Am. Assoc. Petroleum Geologists Bull., v. 17, no. 10, p. 1161—1193. Glover, J. J. E., 1953, Studies in petrology and mineralogy of sedimentary rocks—[Part] I, Stratigraphy and struc- ture of the Chico Martinez-Bitterwater Creek area, Kern County, California, and [Part] II, Investigation REFERENCES 43 of authigenic albite with the universal stage: California Univ. (Berkeley), Ph. D. disse‘rt. Goudkofi‘, P. P., 1941, Correlation of oil field formations on west side of San Joaquin Valley [California]: California Div. Mines Bull. 118, p. 247—252. Gower, H. D., and Pease, M. H., Jr., 1965, Geology of the Montesano quadrangle, Washington: U.S. Geol. Survey Geol. Quad. Map GQ—374, scale 1:62,500. Grant, U. S., 4th, and Gale, H. R., 1931, Catalogue of the marine Pliocene and Pleistocene Mollusca of California: San Diego Soc. Nat. History Mem., v. 1, 1036 p., 32 pls., 15 figs. Hall, C. A., Jr., 1964, Shallow-water marine climates and molluscan provinces: Ecology, v. 45, no. 2, p. 226—234. Hanna, G D., 1936, Interesting whale jaw from Kern County, California [abs.]: Geol. Soc. America Proc. for 1935, p. 419. Hanna, G D., and Hertlein, L. G., 1938, New Tertiary mol- lusks from western North America: Jour. Paleontology, v. 12, no. 1, p. 106—110, pl. 21. Heikkila, H. H., and MacLeod, G. M., 1951, Geology of Bit- terwater Creek area, Kern County, California: Califor- nia Dept. Nat. Res., Div. Mines Spec. Rept. 6, 21. Hickman, C. J. S., 1969, The Oligocene marine molluscan fauna of the Eugene Formation in Oregon: Oregon Univ. Mus. Nat. History Bull. 16, 112 p., 14 pls. Keen, A. M., 1962, Reinstatement of the specific name Macoma inquina-ta (Deshayes): Veliger, v. 4, no. 3, p. 161. 1963, Marine molluscan genera of western North America—An illustrated key: Stanford, Calif., Stan- ford Univ. Press, 126 p. 1971, Sea shells of tropical West America, marine mollusks from Baja California to Peru: Stanford, Calif., Stanford Univ. Press, 1,064 p. Keen, A. M., and Bentson, Herdis, 1944, Check list of Cali- fornia Tertiary marine Mollusca: Geol. Soc. America Spec. Paper 56, 280 p. Kira, Tetsuaki, 1962, [Coloured illustrations of the shells of Japan]: Osaka, Japan, Hoikusha, 239 p., 71 pls. (In Japanese.) Kleinpell, R. M., 1934, Difficulty of using cartographic ter- minology in historical geology: Am. Assoc. Petroleum Geologists Bull., v. 18, no. 3, p. 374—379. 1938, Miocene stratigraphy of California: Tulsa, Okla., Am. Assoc. Petroleum Geologists, 450 p. Kleinpell, R. M., and Weaver, D. W., 1963, Oligocene bio- stratigraphy of the Santa Barbara embayment, Cali- fornia: California Univ. Pubs., Dept. Geol. Sci. Bull., v. 43, 250 p., 38 pls. Knight, J. B., Cox, L. R., Keen, A. M., Batten, R. L., Yochel- son, E. L., and Robertson, Robert, 1960, Systematic descriptions, in Part I, Mollusca 1, of Moore, R. 0., ed., Treatise on invertebrate paleontology: New York and Lawrence, Kans., Geol. Soc. America and Kansas Univ. Press, p. 1169—1331. Lamb, J. L., 1964, The stratigraphic occurrences and rela- tionships of some mid-Tertiary Uvigerinas and Sipho- generinas: Micropaleontology, v. 10, no. 4, p. 457—476. Land, P. E., 1968, East Temblor oil field [California]: Cali- fornia Oil Fields, v. 54, no. 1, p. 57—59. Loel, Wayne, and Corey, W. H., 1932, The Vaqueros forma- tion, Lower Miocene of California—[Part] 1, Paleon- tology: California Univ. Pubs., Dept. Geol. Sci. Bull., v. 22, no. 3, p. 31—410, pls. 4—65. Lutz, G. C., 1951, The Sobrante sandstone: California Univ. Pubs., Dept. Geol. Sci. Bull., v. 28, no. 13, p. 367—406, pls. 15—18, 3 text figs. McCormick, Lavon, and Moore, R. G., 1969, Outline of classi- fication [of Bivalvia], in Part N, Volume 1, Mollusca 6, of Moore, R. 0., ed., Treatise on invertebrate paleon- tology: New York and Lawrence, Kans., Geol. Soc. America and Kansas Univ. Press, p. N218—N222. Mallory, V. S., 1970, Lower Tertiary Foraminifera from the Media Agua Creek drainage area, Kern County, Cali- fornia: Thomas Burke Memorial Washington State Mus. Res. Rept. 2, 181 p., pls. 1—15. Merriam, C. M., 1941, Fossil Turritellas from the Pacific coast region of North America: California Univ. Pubs., Dept. Geol. Sci. Bull., v. 26, no. 1, p. 1—214, pls. 1—41. Moore, E. J., 1963, Miocene mollusks from the Astoria For- mation in Oregon: U.S. Geol. Survey Prof. Paper 419, 109 p., 33 pls. [1964] Palmer, K. V. W., 1958, Type specimens of marine Mollusca described by P. P. Carpenter from the West Coast (San Diego to British Columbia): Geol. Soc. America Mem. 76, 376 p. Park, W. H., and Weddle, J. R., 1959, Correlation study of southern San Joaquin Valley [California]: California Oil Fields, v. 45, no. 1, p. 33—34, pls. 1—3. Parker, Pierre, 1949, Fossil and Recent species of the pelecy- pod genera Chione and Securella from the Pacific Coast: Jour. Paleontology, v. 23, no. 6, p. 577—593, pls. 89—95. Pilsbry, H. A., 1929, Nevem'ta reclusitma (Desh.) and its allies: Nautilus, v. 42, no. 4, p. 109—113, pl. 6. Pilsbry, H. A., and Lowe, H. N., 1933, West Mexican and Central American mollusks collected by H. N. Lowe, 1929—1931: Acad. Nat. Sci. Philadelphia Proc., v. 84, p. 33—144, pl. 1—17. Primmer, S. R., 1964, Preliminary report of the type Kirker Formation [California], in Geol. Soc. Sacramento Guide- book, Mount Diablo field trip, 1964: p. 60—69. 1965, The type Kirker Formation [California]: Cali- fornia Univ. M.A. thesis, 243 p. Rau, W. W., 1964, Foraminifera from the northern Olympic Peninsula, Washington: U.S. Geol. Survey Prof. Paper 374—G, p. G1—G33, pls. 5—7. Reinhart, P. W., 1943, Mesozoic and Cenozoic Arcidae from the Pacific slope of North America: Geol. Soc. America Spec. Paper 47, p. 1—117, pls. 1—15, 3 figs. Ritzius, D. E., 1954, McDonald Anticline oil field [Califor- nia]: California Oil Fields, V. 40, n0. 1, p. 5—13. Ruth, J. W., 1942, The molluscan genus Siphonalia of the Pacific coast Tertiary: California Univ. Pubs., Dept. Geol. Sci. Bull., v. 26, no. 3, p. 287—306, pls. 47—48. Schenck, H. G., 1935, What is the Vaqueros Formation of California and is it Oligocene?: Am. Assoc. Petroleum Geologists Bull., v. 19, no. 4, p. 521—536. 1936, Nuculid bivalves of the genus Acila: Geol. Soc. America Spec. Paper 4, 149 p., 18 pls., 15 figs. Schenck, H. G., and Keen, A. M., 1940, California fossils for the field geologist (prelim. ed.): Stanford, Calif., pri- vately printed, 86 p., 56 pls.; reprinted 1950, Stanford, Calif., Stanford Univ. Press, 88 p., 56 pls. Schenck, H. G., and Kleinpell, R. M., 1936, Refugian stage of the Pacific coast Tertiary: Am. Assoc. Petroleum Geologists Bull., v. 20, no. 2, p. 215—225. 44 MOLLUSCAN BIOSTRATIGRAPHY, PALEONTOLOGY, TYPE TE‘MBLOR FORMATION, CALIF. Shumard, B. F., 1858, Descriptions of new fossils from the Tertiary formations of Oregon and Washington Terri- tories and the Cretaceous of Vancouver Island: St. Louis Acad. Sci. Trans., v. 1, no. 2, p. 120—123. Simonson, R. R., and Kreuger, M. L., 1942, Crocker Flat landslide area, Temblor Range, California: Am. Assoc. Petroleum Geologists Bull., V. 26, no. 10, p. 1608—1631. Smith, A. G., and Gordon, Mackenzie, Jr., 1948, The marine mollusks and brachiopods of Monterey Bay, California, and vicinity: California Acad. Sci. Proc., 4th ser., v. 26, no. 8, p. 147—245, pl. 3, 4. Smith, H. P., 1956, Foraminifera from the Wagonwheel Formation, Devils Den district, California: California Univ. Pubs. Geol. Sci., v. 32, no. 2, p. 65—126. Soot-Ryen, Tron, 1955, A report on the family Mytilidae (pelecypods): Allan Hancock Pacific Exped., v. 20, no. 1, 175 p., pls. 1—10, figs. 1—78. Stewart, R. B., 1927, Gabb’s California fossil type gastro- pods: Acad. Nat. Sci. Philadelphia Proc., v. 78, p. 287— 447, pls. 20—32, 5 text figs. 1930, Gabb’s California Cretaceous and Tertiary type lamellibranchs: Acad. Nat. Sci. Philadelphia Spec. Pub. 3, 314 p., 17 pls., 5 text figs. 1946, Geology of Reef Ridge, Coalinga district, Cali- fornia: U.S. Geol. Survey Prof. Paper 205—0, p. 81—115, pls. 9—17 [1947] Stinemeyer, E. 8., Beck, R. S., Ortalda, R. A., Espenscheid, E. K., Bainton, J. D., and O’Keefe, M. S., 1959, San Joaquin Geol. Soc. Guidebook, Chico Martinez Creek area field trip [California] 1959: 15 p. Tegland, N. M., 1933, The fauna of the type Blakeley upper Oligocene of Washington: California Univ. Pubs., Dept. Geol. Sci. Bull., v. 23, no. 3, p. 81—174, pls. 2—15. Trask, P. D., 1922, The Briones formation of middle gain- fornia: California Univ. Pubs., Dept. Geol. Sci. ull., v. 13, no. 5, p. 133—174, pls. 1—8. Trumbull, E. J., 1958, Shumard’s type specimens of Tertiary mollusks from Oregon and other types at Washington University, St. Louis: Jour. Paleontology, v.32, no. 5, p. 893—906, pls. 115—117. Valentine, J. W., 1961, Paleoecologic molluscan geography of the Californian Pleistocene: California Univ. Pubs. Geol. Sci., v. 34, n0. 7, p. 309~442. 1966, Numerical analysis of marine molluscan ranges on the extratropical northeastern Pacific shelf: Limnol- ogy and Oceanography, v. 11, no. 2, p. 198—211. Valentine, J. W., and Mallory, Bob, 1965, Recurrent groups of bonded species in mixed death assemblages: Jour. Geology, v. 73, no. 5, p. 683—701. Vanderhoof, V. L., 1942, Bearing of sea-cows on age of Vaqueros, in Schenck, H. G., and Childs, T. S., Jr., Significance of Lem’docyclina (Lepidocyclina) califor- mica, new species, in the Vaqueros formation (Tertiary), California: Stanford Univ. Pubs. Geol. Sci., v. 3, no. 2, p. 40—42. Vokes, H. E., Norbisrath, Hans, and Snavely, P. D., Jr., 1949, Geology of the Newport-Waldport area, Lincoln County, Oregon: U.S. Geol. Survey Oil and Gas Inv. Prelim. Map 88, scale 1:62,500, geol. map and text. Wagner, C. M., and Schilling, K. H., 1923, The San Lorenzo Group of the San Emigdio region, California: California Univ. Pubs., Dept. Geol. Sci. Bull., v. 14, no. 6, p. 235— 276, pls. 43—50. Warren, W. G., Norbisrath, Hans, and Grivetti, R. M., 1945, Geology of northwestern Oregon west of Willamette River and north of latitude 45°15’: U.S. Geol. Survey Oil and Gas Inv. Prelim. Map 42, geol. map and text. Weaver, C. E., 1912, A preliminary report on the Tertiary paleontology of western Washington: Washington Geol. Survey Bull. 15, 80 p., 15 pls. 1942, Paleontology of the marine Tertiary formations of Oregon and Washington: Washington Univ. [Seat- tle] Pubs. Geology, v. 5, pts. 1—3, 789 p., 104 pls. 1949, Geology of the Coast Ranges immediately north of the San Francisco Bay region, California: Geol. Soc. America Mem. 35, 242 p. Weaver, C. E., chm., and others, 1944, Correlation of the marine Cenozoic formations of western North America: Geol. Soc. America Bull., v. 55, no. 5, p. 569—598. Weaver, D. W., and Frantz, J. D., 1967, Re-evaluation of the type Refugian Stage of the Pacific Coast Tertiary [abs.]: Geol. Soc. America Ann. Mtg., Cordilleran Sec. 63d, Santa Barbara, Calif. March 22—25, 1967, Program p. 69—70. Weddle, J. R., 1966, Carneros, Phacoides, and Oceanic pools, McKittrick Front area of Cymric oil field [California]: California Oil Fields—Summ. Operations 1966, v. 52, no. 2, pt. 2, p. 23—30. [1967] Wilson, E. J., 1954, Foraminifera from the Gaviota Forma- tion east of Gaviota Creek, California: California Univ. Pubs. Geol. Sci., v. 30, no. 2, p. 103-170, pls. 12—18. Woodring, W. P., 1926, American Tertiary mollusks of the genus Clementia: U.S. Geol. Survey Prof. Paper 147—C, p. 25—47, pls. 14—17. 1938, Lower Pliocene mollusks and echinoids from the Los Angeles Basin, California, and their inferred en- vironment: U.S. Geol. Survey Prof. Paper 190, 67 p., 9 pls. Woodring, W. P., Stewart, Ralph, and Richards, R. W., 1940, Geology of the Kettleman Hills oil field, California: U.S. Geol. Survey Prof. Paper 195, 170 p., 57 pls. [1941] U. 5. GOVERNMENT PRINTING OFFICE : 1973 O - 493-696 INDEX [Italic page numbers indicate major references and descriptions] Page A abrupta, Panopea. __________________ 37, 38 Avila, ___________ __ 14, 22 com‘adi __ 3, 22 gettysburgensis _________________ 13 mum _________________________ 22 markleyensia _______________ 22 ahumurdi ____________ 4, 12, 23, 28, 30, 33 Zone ______________________ 8 (Truncacila) muta. ___________ 22; pl. 1 Acknowledgments __________________ 3 Acn'lla. dickersoni __________________ 12 acuminata, Bruclarkia ______________ 21 acutilineata, Lucinu ________________ 4 Lucina (Myrteu) ______________ 28 Lucinoma ________ 3, 4, 6, 14, 28; pls. 2, 3 acutilineatus, Phacoides _____________ 3, 28 Phacoides (Lucinoma) ___________ 28 aduncanasa, Tellinu ________________ 33 Aforia ___________________________ 13 Age, Agua Sandstone Member _______ 14 Cymric Shale Member __________ 8 Temblor Formation _____________ 1 molluscan evidence for ______ 9 Wygal Sandstone Member ______ 2, 8, 10 Agua Sandstone Member ___- 2, 7, 10, 13, 14 albaria, Spisula _______________ 29, 30; pl. 4 Alegria Formation _________________ 13 alexclarki, Vertipecte’n _ 4, 10, 17, 26; pls. 1, 2, 3 Amiantis _______________ 16, 17, 35, 36; pl. 6 callosa ________________________ 35 diabloensis ____________________ 35 mathewsom’ _______________ 35, 36; pl. 7 Amussiopecten sp __________________ 3 Anadara _______________________ 16, 23, 24 strongi _______________________ 23, 24 subma’ntereyana ________________ 23, 24 (Anadara) mediaimpreasa montereyana ____________ 23 mediaimpressa submantereyana- 23 submontereyana ___________ 23; pl. 1 (Anada'ra) mediaimpressa montereyana, Anadam. _ _ _ 23 mediaimpressa submontereyana, Anadara _______________ 23 submontereyana, Anadara __ _ 23; pl. 1 Ancistrolepis _______________ _ 13 andersom', chtopecten ____________ 26 N everim ______________________ 19 angustifrons, Pitar (Katherinella) ____ 37 annulata, Lucinomu ________________ 14, 28 Antigona neglectu __________________ 35 undosa ________________________ 35 (A'rtrma) mathewsonii __________ 35 neglecta ___________________ 35 ( Ventricula) undosa ____________ 35 Antropom ________________________ 17 sp ___________________________ 21 Apolymetis sespeensis _______________ 34 twinensis _____________________ 34 vancouverensis __________________ 34 Area, (Scapharca) submontea'eyanu ___- 23 Archaeogastropoda _________________ 17 Arcidae ___________________ . _______ 23 Arcoida ___________________________ 23 Page a’rctuta, Maco’ma _____________ 14, 38; pl. 5 Macama (Paammacoma) __ _ 33 Tellina. ________________ _ 33 wynoocheensis, M acoma urnoldi, Crenomytilus ___________ (Artena) mathewsornii, Antigona _____ 35 neglecta, Antigona __________ _ 35 attenuata, Spisula ramoncnsis _ _ 29 B barkeriana, Brucla'rkia __________ 3, 4, 21, 22 santacruzana, Bruclarkia _____ 10, 21, 22 “Barren” shale member of Temblor Formation ______ 8 Bathybembix ______________________ 13 bodegensis, Tellina _________________ 32 borealis, Lucinu ___________________ 28 branneri, Calicantharus ___________ 20; pl. 9 Chlamys ______________________ 26 Pecten ________________________ 4 (Chlamys) _________________ 26 Bruclarkia ___________________ 9, 10, 14, 21 acuminata _____________________ 21 barkeriana _________________ 3, 4, 21, 22 santacruzana ____________ 10, 21, 22 columbiana ____________ 8, 10, 12, 21, 22 gravida _______________________ 21 oregonensis ____________________ 22 santacruzana ___________________ 21 seattlensis _______ 3, 4, 10, 13, 21, 22; pl. 9 Buliminclla curta __________________ 9 Bullia clarki ______________________ 12 C Calicantharus 20 b’ra‘rmeri _ ______ 20; pl. 9 dalli ___ ______ 20; pl. 9 Icernensis __ 20 woodfordi ___ ___ 20 California, Oliva _ _ 3 califo’rnica, Pitar _______ 37 Pitar (Katherinella) _ _ 37, pl. 6 Plectof'rondicularia. ___ _ 15 callosa, Amiantis ___- _ 35 Calyptraeu ______ _ 14, 17 coreyi __ 18 diegoana ___________ __ 17; pl. 8 mammilaris vancouverensis .. 18 washingtonensis _____________ _ 17, 18 Calyptraeidae __________________ 17 canalis, Polinices __________________ 19 Cancellaria ________________________ 14 carnarosensis, Yoldia _______________ 23 Carneros Sandstone Member ______ 7, 14, 15 Cerithiid _______________________ 17; pl. 8 Cerithiidae ________________________ 17 Chione lineolata ___________________ 37 mathewsrmii ___________________ 35 panzana ______________________ 3 whitneyi ______________________ 36 Chlamys __________________________ 26 branneri ______________________ 26 hertleini ______________________ 13 seapeensis _____________________ 3 Page (Chlamys) b'rmmeri, Pecten _________ 26 sespeensis, Pecten _ _ 10 Cibicides floridanus __________ _ 15 pseudoungefianus evolutus __ _ 25 clallamensis, Tell-int; _________ _ 32 clarki, Bullia __ __ 12 Clementia _,__ __ 16, 36 pertenm’s __ __ _____ 36 (Egesta) pertenuis __ _ .76, pls. 7, 8 caalingensis, Crenamytilus __ _______ 24 cocoaensis, Uvigerina ___- ________ 9 columbiana, Brucla'rlcia. ______ 8, 10, 12, 21, 22 condom, Epitonium ________________ 19, 12 canmdi, Acila _____________________ 3, 22 cooperi tenuiasima, Yoldia ___________ 22 cooperii, Yoldia ____________________ 22 supramantereyensis, Yoldia _______ 22 coreyi, Calyptraea _________________ 18 Correlation of Cymric Shale Member__ 9 crassicardo, Lyropecten _____________ 3 Crassostrea ________________ 6, 14, 16, 25, 27 eld’ridgei ynezana _____________ 27; p]. 2 sookensis ______________________ 27 vaquerosensis __________________ 14 sp ___________________________ 13, 27 Crenomytilus ___________________ 14, 16, 24 a'rnoldi _____________________ 24; pl. 1 coalingensis ___________________ 24 expansus ___________________ 24; pl. 4 loeli __________________________ 24 Crepidula _________________________ 18 princeps ______________________ 18 sookensis _____________________ 18 ungana _____________________ 18; pl. 8 cryptolineata, Securella ___________ 37; pl. 6 curta, Buliminella _________________ 9 Cyclamina incisa __________________ 15 Cyclas permacru ___________________ 28 Cymric oil field ___________________ 6 Cymric Shale Member _________ 2, 6, 7, 8, 15 D dalli, Culicantharus _______________ 20; pl. 9 Molopophorus __________________ 12 Searlesia ______________________ 20 Dentaliidae _______________________ 38 Dentalium ________________________ 38 laneensis ____________________ .98; pl. 8 radiolineata ___________________ 38 (Fissidentalium) lanecnsis _ _ 38 diabloensis, Amiantis _ 35 dickersoni, Acrilla _1 _ 12 Pecten _________ _ 25 (Pecten) ___ __ 25 dieyoana, Culyptraea _ 17; pl. 8 True/Lita __________ _ 17 diegoensis, Lucina _________ _ 10 Diplodonta harfordi ________ _ 29 disjuncta, Thyasira 10 Dasinia __________________________ 16, 36 mathcwsonii ___________________ 36, 37 ponderosa _____________________ 15 whitneyi ______________________ 36, 37 (Dosim'a) whitneyi ________ 36; pls. 6, 7 45 46 Page Dosinia—Continued (Dosinidia) mathewsom' _________ 36 whitneyi ___________________ 36 (Dosim'a) whitneyi, Dosimla _____ 36; pls. 6, 7 (Dosinidia) malhewszmi, Dosinia _____ 36 whitneyi, Dosinia. ______________ 36 E Echinophoria __________ - 13 aptu zone ________ _ 13 Tea: _____________ __ 19, 22 Zone ______ 12, 13, 16 E gesta __________________________ 36 (Egesta) pertenuis, Clementia 36; pls. 7, 8 eldridgei, Ostrea ________________ 27 unezana, Crassostrea __ __ 27; pl. 2 Ostrea ._ _____ 27 emace'rata, Tellina ___________ 32 Tellina (Oudurdia) ___________ 32; pl. 4 Epistomina ramtmensis _____________ 9 Epitom'um condoni _________________ 10, 12 estrellama, Panopea _________________ 38 cvolutus, Cibicides p8eud0ungerianus--- 25 excavata, Here __________________ 27; pl. 3 Here (Here) __________________ 27 Lucina _______________________ 27 (H are) ___________________ 27 excenlricus, Galerus ________________ 17 expansus, Crenomytilus ___________ 24; pl. 4 Mytilus ____________________ 24 malhewsoni _____________ 24 F Fauna, Wygal Sandstone Member“ 2,5,6, 15 Favites __________________________ 16, 1‘7 Fela'niella. _________________________ 29 harfordi ______________ _ 29; pl. 4 Ficidae __________________ __ 20 Ficus _____________ -_ 16, 20 modesta ________ __ 20; pl. 9 pyriformis ___________ __ 20 (Trophasycon) yesten‘ ______ __ 20 Fish scales, Cymric Shale Member ___ 9 Santos Shale Member ___________ 9 Temblor Formation 9 (Fissidentalium) laneensis, Dentalium__ 38 floridanus, Cibicides ________________ 15 Foraminifera, biostratigraphic study of- 1 Cymric Shale Member ___________ 9 depth range for ___________ 15 Santog Shale Member, depth range for ______________ 15 Temblor Formation _____________ 9 frustra, Spisula pittsbu‘rgensis _______ 29 G gabb'i, Molopophorus ________________ 33 Galerus excentricus ________________ 17 Gastropoda _______________________ 17 Gastropods _______________ _ 6, 14 Gaviota Formation __________ _ 10 gesteri, Ficus (Trophosycon) _ _ 20 gettysburgensis, Acila __ _ 13 Glossaulaa; ________________________ 19 (Glossaulax) thomsonae, Neverita __ 19; pl. 9 gravida, Bruclarlcia ________________ 21 gravidus, Solen __ Gyraidina, aoldanii harfardi, Diplodonla _______________ 29 Felaniella ___________________ 29; pl. 4 hemphilli, Spisula __________________ 30 INDEX Page Here __________________ __ 16, 27 excavata ____________ _ 27; pl. 3 (Here) excavam ____ _ 27 (Here) excavata, Here __ _ 27 excavate, Lucina ___- _ 27 Richthofeni, Lucina _ _ 27 richthofeni, Phacoides _ _ 27 hertleini, Chlamys _______ _ 13 Hetemmacomu __ _____ 34 rostellum __ _ 6, 34; pls. 5, 6 sespeensis _________________ 34 ( H eteramacoma ) vaneowverensis, M acoma _______________ 34 Introduction Invertebrates, depth ranges of ________ 14 J jacksonensis, Uvigerina, _____________ 9 J ewett Sand ______________________ 14 K Kalayoldia ________________________ 22, 23 (Kalayoldia) tenuissima, Yoldm ___- 22; pl. 1 Katherinella _______________________ 37 (Katherinella) angustifrons, Pitar ____ 37 califo’r‘n‘iCa, Pita/r _____________ 37; pl. 6 Kelletia __________________________ 16, 21 sp ___________________________ 21 kennerlyi, Venus ___________________ 36 kernensis, Calicantharus 20 kernianum, Trophasyczm ____________ 3 L laem‘s, Tegula ____________________ 17 laneensis, Dentalium _____________ 38; pl. 8 Dentalium (Fissidentalium) _____ 38 Leptopecten ______________________ 14, 26 andersoni _ 26 sp _________________________ 26; pl. 1 lincolnensis, Perse _________________ 12 lineolata, Chione ___________________ 37 Liracassis ________________________ 13 Lithology, Temblor Formation ________ 1, 4 Wygal Sandstone Member _______ 5, 6 Locality, Wygal Sandstone Member ___ 4 loeli, Crenomytilus _________________ 24 lorenzana, Olequahiu _____________ 8, 10, 12 Pitar _________________________ 37 Turritella _____________________ 10 Lucina. acutilineam ________________ 4 borealis _______________________ 28 diegoensis _____________________ 10 excavata ______________________ 27 richthofeni ____________________ 27 (Here) excuvata _______________ 27 Richthofcni ________________ 27 (Myrtea) acutilineala ___________ 28 Lucinidae _________________________ 27 Lucinoma _____________________ 14, 15, 28 acutilineata ______ 3, 4, 6, 14, 28; pls. 2, 3 annulam ______________________ 14, 28 (Lucinoma) acutilineatus, Phacaides __ 28 Lyropecten Crassica’rdo _____________ 3 M Macoma ______________ 33 urctata _________ _ 14, 33; pl. 5 wynoocheensis -_ _______ 33 nasuta _______________________ 31, 33 Page Manama—Continued piercei ________________________ 31 secta _________________________ 31 seapeensis _____________________ 34 wynootcheensis ______________ 33 (Heteromacoma) vancmwerensis 34 (Paammacoma) arctam ______ _ 33 Macrocallista pittsburgensis ___ _ 12 Macrochlamis magnolia ___ _ 14 Mactridae ______________ _ 29 magnolia, Macrochlamis _ _ 14 malibuensis, Tegula _________________ 17 mammilaris vancouverensis, Calyptraea- 18 markleyenais, Acila, mum ___ _______ 22 _. 35, 36: pl. 7 ._-__ 36 mathewstmi, Amiantia ___ Dosinia (Dosinidia) _ expansus, Mytilus _________ _ 24 mathewsonii, Antigona (Artena) _ 35 Chione ________________________ 35 Dosinia ______________________ 36, 37 medim'mpressa mantereyana, Anadara (Anadara) _____________ 23 submontereyama, Anadara (Anadara) _____________ 23 merriami, Siphonalia _______________ 12 Mesogastropoda ____________________ 17 Men's rastellata ____________________ 34 vanco‘uverensis _________________ 34 Miltha. ___________________ 14, 15, 16, 28, 29 sanctaecrucis __________________ 29 mantusi _______________________ 29 (Miltha) sanctaecrucis _________ 28; pl. 3 (Milthu) sanctaecrucis, Miltha ______ 28; pl. 3 sanctaecrucis, Phacoides _________ 28 modesta. Ficus __________________ 20; pl. 9 Mollusks _________________________ 8 Agua Sandstone Member _________ 14 biostratigraphic study of ________ 1 Cymric Shale Member ___________ 9 provinces _____________________ 16 Santos Shale Member ____________ 13 Temblor Formation _____________ 1, 6 annotated references to ______ .9 tufi member of Kirker Formation __ 13 Wygal Sandstone Member ______ 7, 10, 15 Molopophorua ________________ 13 dalli __________________ 12 gabbi ________ 33 stephensoni _____ 33 Monterey Shale ____________________ 4 montereyana, Anadara (Anadara) mediaimpressu __________ 23 mum, Avila ______________ _ 22 Avila (Truncacila) -22; pl. 1 markleycnsis, Acila. __ __ 22 Myidae ________________ __ 37 Myoida _________ 37 (Myrtea) acutilineata, Lucina ________ 28 Mytilidae _________________________ 24 Mytiloida __ 24 Mytilus __________________________ 14 expansus ______________________ 24 mathewsom' expansus ____________ 24 N nasuta, Manama. ___________________ 31, 33 Nutica ____________________ 16, 18, 19; pl. 8 pasuncula _____________________ 19 teglandae _____________________ 19 vokesi ________________________ 19 (Nation) ______________________ 17, 19 (Nation), Natica. ___________________ 17, 19 Naticidae _________________________ 18 neglecta, Antigona _________________ 35 Antigona (Artena) ______________ 35 Neogastropoda _____________________ 20 Neptuneidae ______________________ 20 Page nevadensis, Tellina ______________ 30, 31, 32 N everita _________________________ 16, 19 andersoni _____________________ 19 recluziana _____________________ 19 thomsonae _____________________ 19 (Glossaulax) thomaonae ________ 19; pl. 9 nodifera, Siphogenen’na ______________ 9, 15 Nuculanidae ______________________ 22 Nuculidae ________________________ 22 Nuculoida ________________________ 22 0 obliquum, Sinum ___________________ 19 occidentalis, Spisula ________________ 30 ocoyana. Tellina ___________________ 31, 32 Olcesia. ________________________ 16, 17, 80 (Olcesia) piercei, Tallinn _______ 31; pls. 4, 5 Olequahia ________________________ 8 lo’re’nzana ___________________ 8, 10, 12 Oligocene vertical oscillation of San Joaquin basin __ -___ 15 Olioa california, __________ -_ 3 o’regona, Yoldia. ___ _ __ 23 Yoldia. (Pa'rtlandia) _ __ 22 o'rego'nensis, Brucla‘rkia __ __ 22 Tellina __________ __ 32 (Peronidia) ___ __ 32 Ostrea, ______________ __ 3 eldridgei __________________ _.. 27 ynezana. _______________ __ 27 sp _______________________ __ 4 Ostreidae _____________________ 27 Oudurdia _________________________ 32 (Oudardia) emacerata, Tellina ___- 32; 111.4 P Paleobathymetry ___________________ 14 Panope mmonensis _________________ 37 Panopea __________________________ 37 abrupta _______________________ 37, 38 “ estrellana _____________________ 38 mmonensis __________________ 37; pl. 8 (Panopea) rumonensis ___________ 37 (Panopea) rammensis, Panapea ______ 37 panzana, Chione ___________________ 3 patulus, Pectunculua ________________ 28 Pecten ___________________________ 24, 25 branneri ______________________ 4 dickersoni _____________________ 25 sanctaecruzensis ___________ 6, 10, 24, 25 (Chlamys) brunneri _____________ 26 sespeensia __________ 10 (Pecten) dickersoni _ _____ 25 sanctaecruzensis -24; pl. 2 sp _________________ ._ 26 (Pecten) dicke’rsoni, Pecten _ 25 sanctaecruzensis, Pecten _ _ 21;; pl. 2 Pectinidae _______________ __ 24 Pectunculus patulus __ _ 28 Pelecypoda _______ ___ 22 Pelecypods ______________________ 6, 10, 14 permacra, Cyclas __________________ 28 Peronaea _________________________ 31 Peronidia _________________________ 31 (Perom'dia) oregonensis, Tellina ______ 32 perrz'ni, Vertipecten _____________ 10, 14, 26 Perse lincolnensis __________________ 12 pertenuis, Clementia ________________ 36 Clementia (Egesta) ________ 36; pls. 7, 8 Venus ________________________ 36 Phacaides _______________________ 3, 4, 21 acutilincatus ___________________ 3, 28 richthofeni ____________________ 27 (Here) richthofeni ______________ 27 (Lucinama) acutilineatus _________ 28 (Miltha) sanctaecrucis ___________ 28 INDEX Page Phacoides reef ____________ _ _ 4 Phacoides sand ___ _. 2, 4, 8 piercei, Macoma. __________ .._ 31 Tallinn _______________ __ 30, 31, 32 (Olcesia) ____________ 31; pls. 4, 5 Pitar ________________________ 34, 37 caliform'ca _____________________ 37 lo'renzana _____________________ 37 ( K atherinella ) unguatifrona ______ 37 californica _______________ 37; pl. 6 sp _________________________ 37; pl. 6 pittsburgensis, Macrocullista _________ 12 frust'ru, Spisula ________________ 29 Plectafrondicularia californica ________ 15 'uuugham‘ ______________________ 9, 15 Pleito Formation ______________ 9, 11, 12, 13 Point of Rocks Sandstone ___________ 4, 6 Polim'ces __________________ 17, 18, 1.9; pl. 9 canalis _______________________ 19 victoriana _____________________ 19 ponderosa, Dosinia _________________ 15 Poromya _________________________ 34 teglandae _____________________ 34 (Portlandia) oregano, Yoldia _________ 22 poauncula, Natica _________________ 19 princeps, Crepidula _________________ 18 (Psammacoma) arctata, Mavco’ma _____ 33 Paeudocardium ____________________ 29 sp _________________________ 29; pl. 5 pseudoungerianus evolutus, Cibicides -_ 25 Pterioida _________________________ 24 pyrifmmis, Ficus __________________ 20 Q Quail Canyon Sandstone Member of Vaqueros Formation _____ 13 R radiolineata, Dentalium _____________ 38 rammensis attenuatu, Spisula _______ 29 Epistomina ____________________ 9 Panope _______________________ 37 Panopea ____________________ .97; pl. 8 (Panopea) __________________ 37 Spisula _____________________ .90; pl. 4 recluzianu, Neven‘ta _____________ ,__ 19 rem, Echinophoria _________________ 19,22 Richthofeni, Lucina. (Here) __________ 27 richthofem', Lucina ________________ 27 Phacoides _____________________ 27 (Here) ____________________ 27 rostellata, Heteromacomu _____ 6, 34; DIS. 5, 6 .Metis ________________________ 34 rushi, Spisula ___________________ 30, pl 4 S Salt Creek Shale ___________________ 6 San Emigdio Formation _ ________ 9 San Lorenzo Formation _ ________ 10 San Ramon Sandstone __ _ 13 sunctaec’rucis, Miltha, _ __ 29 Miltha, (Miltha) _ _ 28; pl. 3 Phacoides (Miltha) ___- 28 sanctaecruzensis, Pecten __ _ 6, 10, 24, 25 Pecten (Pecten) ________ ___- 24; pl. 2 santacruzana, Bruclarkia ___- ___- 21 B’ruclarkiu barkericma ________ 10, 21, 22 Santos Shale Member __________ 2, 9, 13, 15 (Scapharca) submontereyana, Area -__ 23 Scaphopoda _______________________ 38 scopulosum, Sinum _______________ 19; pl. 9 Searlesia __________________________ 20 dalli _________________________ 20 seattlensis, Brucla‘rkia- 3, 4, 10, 13, 21, 22; pl. 9 Page secta, Macoma ____________________ 31 Securella _________________________ 37 cryptolineata ________________ 87; pl. 6 Sespe Formation __________________ 6 sespeensis, Apolymetis ______________ 34 Chlamys ______________________ 3 Heteromacoma _________________ 34 Manama ______________________ 34 Pecten (Chlamus) ______________ 10 shumardi, Acila _________ 4, 12, 23, 28, 30, 33 Sinum ___________________________ 19 obliquum ______________________ 19 scopulosum __________________ 19; pl. 9 Siphogenerina nodife'ra _____________ 9, 15 transversa _____________________ 9 Siphonalia ________________________ 16, 21 merriami ______________________ 12 sp _________________________ 21; pl. 9 Soda Lake Sandstone Member ________ 13 soldam’i, Guroidina _________________ 15 Salem ___________ ___- 30 gravidus __ __ 30; pl. 4 Solenidae ____________ ___- 30 sookensis, Crassastrea __ __-_ 27 Crepidulu ________________ 18 South Belridge oil fields ________ 4 Spisula ______________ 17, 29, 30; pl. 4 albaria __- __. 2.9, 30; pl. 4 hemphilli -__ ______ 30 occidentalis _________ _ 30 pittsbu‘rgensis frustra _ 29 mmonensis _________ __30; pl. 4 attenuato--- _-_ 29 rush; ____________ __30; pl. 4 sp ________________ _ 14 stephensani, Molopophorus _ _ 33 strongi, Anadara ________ _ 23,24 submontereyana, Anadam ____ _ 23, 24 Anadam (Anadara) __ 23; pl. 1 (Anada'ra) mediaimp‘ressa _____ 23 Area. (Scapharca) ______________ 23 suprumontereyensis, Yoldia __________ 23 Yoldia cooperii ________________ 22 Systematic descriptions _____________ 16 T teglandae, Natica __________________ 19 Poromya ______________________ 34 Tegula. ______________________ 14, 17; pl. 8 laevis ________________________ 17 malibueneis ___________________ 17 Tellina ___________________________ 30 aduncanasa ____________________ 33 arctata _______________________ 33 bodeyensis _____________________ 32 clallamensis ___________________ 32 emucerata _____________________ 32 nevadensis _________________ 30, 31, 32 ocoyana ______________________ 31, 32 oregonensis ____________________ 32 piercei _____________________ 30, 31, 32 tenuilineata ____________________ 32 townsendensis _______________ 32; pl. 6 vancouverensis _______________ 83; pl. 4 (Olcesia) piercei ___________ 81; pl. 4, 5 (Oudardia) emacerata ________ 32; pl. 4 (Peronidia) oregonensis _________ 32 (Tellinellu) tenuilineata _______ 32; pl. 5 Tellinelia _________________________ 16, 32 (Tellinella) tenuilineata, Tellina_ __ 32; pl. 5 Tellinidae _________________________ 30 Temblor Formation, microfaunal sequence _______________ 6 provinicial age and correlation 6 stages of ______________ 7 stratigraphy 4 48 Page Temblor Range ____________________ 4 Temblor Stage ____________________ 1 Temblor Zone _____________________ 1 temblorensis, Yoldia ________________ 23 tenuilineata, Tellina, ________________ 32 Tellina (Tellinella) ____________ 32; pl. 5 tenuiss‘imu, Yoldia _________________ 22,23 Yoldia cooperi __________________ 22 (Kalayaldia) _____________ 22; pl. 1 Thickness, Cymric Shale Member _____ 9 Temblor Formation 1, 4 Wygal Sandstone Member ___ _ 5 tho’msonae, Neverita _________ __ 19 Neverita (Glossaulax) _ _ 19; pl. 9 Thyaaira disjuncta. _______ -_ 10 Tivela. _______________ __ 15 townsendensis, Tellina ___- _ 32; pl. 6 transverse, Siphogencrina, __ 9 Trochidae _______________ _ 17 Trachita, diegoa/na _____ _ 17 Trophosyctm kernianum ____ _ 3 (Trophosycon) gesten', Ficus _ 20 Truncacila _______________ 22 (Truncacila) muta, Acila _ pl. 1 tumeyensis, Uvige'rina ___- Turcicula _ Turritella __ inezama __ zone _ _________________ lorenzana, _ ___________________ twinensis, Apolymetis ______________ 34 Type area, Temblor Formation _______ 1 Type locality, Temblor Formation ___- 4 U Unconformity, Temblor Formation ___ 4 Wygal Sandstone Member ________ 6, 15 INDEX Page undosa, Antigona __________________ 35 Antigona (Ventricola) __________ 35 ungana. Crepidula _______________ 18; pl. 8 Ungulinidae ______ 29 Uvigerina cocoaensis _______________ 9 jacksonensis ___________________ 9 tumeyensis ____________________ 9 vicksburgensis _________________ 9 V vancouve’rensis, Apalymetis __________ 34 Calyptraea mummilaris __________ 18 Macama (Heteromacoma) _______ 34 Metis ________________________ 34 Tellina _____________________ 33; pl. 4 Vaqueros Formation _______________ 10, 14 Vaqueros Sandstone ________________ 10 Vaqueros Stage ___________________ 8, 10 vuque‘rosensis, Crassostrea ___________ 14 vaugham', Plectof’rondicularia, ________ 9, 15 Veneridae ________________________ 35 Veneroida ________________________ 27 (Ventricula) undosa, Antigana ______ 35 Venus kennerlyi ___________________ 36 pertenuis _____________________ 36 Vertipecten ____________________ 10, 16, 26 ulexclarki ________ 4, 10, 17, 26; p15. 1,2, 3 per'rini ____________________ 10, 14, 26 yneziana ______________________ 10 vicksburgensis, Uvigerina ___________ 9 victoriuna, Polim'ces ________________ 19 vokesi, Natica _____________________ 19 W Wagonwheel Formation _____ 10 washvingtone’nsis, Calyptraea _________ 17, 18 Page whitneyi, Chione ___________________ 36 Dosimla, _______________________ 36, 37 (Dasinia) _____________ 36; pls. 6, ‘7 (Dosinidia) _______________ 36 woodfordi, Calicanthm‘us ____________ 20 Wygal Sandstone Member _________ 2, 4, 10 corals ________________________ 16 genera of tropical aspect ________ 14 Lucinoma-Heteromacama bed __ - _ 6 wynoocheensis, Momma. urctata ______ 33 wynootcheensis, Macoma ____________ 33 X xantusi. Miltha ____________________ 29 Y ynezana, Crassostrea eldridgei _____ 27; pl. 2 Ost'rea cldridgei ________________ 27 yneziana, Vertipecten 10 Yoldia ___________________________ 22 carnarosensis _____ __ 23 cooperi tenuissima __ __ 22 coope'rii ____________ __ 22 supra/monte'reyensis __ a- 22 oregano ______________ __ 23 supramontereyensis __ __ 23 tembloreneis _______ _- 23 tenuissima _____ _ _ __ 22,23 (Kalayoldia) tenuissima _ _ 22; pl. 1 (Portlandia) oregonu _____ ___- 22 Z Zoogeography and paleoclimatology_-__ 15 PLATES 1-9 [Contact photographs of the plates in this report are available, at cost, from U.S. Geological Survey Library, Federal Center, Denver, Colorado 80225] PLATE 1 [Specimens are from the Wygal Sandstone Member of the Temblor Formation unless otherwise indicated] FIGURES 1, 18. C’renomytilus? cf. C.? arnoldi (Clark) (p. 24). 1. Length 83 mm, height 35 mm. USNM 646511. USGS 10c. M3578. 18. Length 120 mm, height 68 mm. USNM 646512. USGS 10c. M3979. 2, 3, 5, 8. Yoldia (Kalayoldia) tenuissima Clark (p. 22). 2. Length 45 mm, height 24.5 mm. USNM 646513. USGS 10c. M3578. 3. Length 48 mm, height 25.5 mm. USNM 646514. USGS 10c. M3578. 5. Length 43 mm, height 23.5 mm. USNM 646515. USGS 10c. M3578. 8. Length 31 mm, height 16 mm. USNM 646516. USGS Ice. 6622. Round Mountain Silt, Kern River area, California, middle Miocene. 4, 15. Acila (Truncacila) muta Clark (p. 22). 4. Length 12 mm, height 10.5 mm. USNM 646517. USGS 10c. M4471. 15. Length 12.2 mm, height 10.5 mm. USNM 646518. USGS loc. M4466. 6, 7, 11, 16, 17. Anadam (Anadara) submontereyana (Clark) (p. 23). 6. Length 30 mm, height 20 mm. USNM 646519. USGS 10c. M3978. 7. Length 32 mm, height 22.5 mm. USNM 646520. USGS 10c. M3978. 11. Length 20 mm, height 13 mm. USNM 646521. USGS 10c. M3978. 16, 17. Length 26 mm, height 17 mm. USNM 646522. USGS 10c. M3772. 9, 10, 13. Vertipecten alexclarki Addicott, n. Sp. (p. 26). 9. Length 24 mm, height 28 mm. USNM 646523. USGS 10c. M3280. 10. Height 42 mm, thickness 13.5. USNM 646524. USGS 10c. M3636. 13. Height of fragment 33 mm. USNM 646525. USGS loc. M3280. 12. Yoldia (Kalayoldia) cf. Y. (K.) carna'rosensis Clark (p. 23). Length 32 mm. USNM 646526. USGS 10c. M1596. Olcese Sand- stone, Kern River area, California, middle Miocene. 14. Leptopecten? sp. (p. 26). Height 24 mm. USNM 646527. USGS loc. M3280. PROFESSIONAL PAPER 791 PLATE 1 GEOLOGICAL SURVEY $1~m . « 'P'u mung}, 18 l7 ACILA, ANADARA, CRENOMYTILUS ?, LEPTOPECTEN, VERTIPECTEN, YOLDIA PLATE 2 [All specimens are from the Wyga] Sandstone Member of the Temblor Formation] FIGURES 1, 3, 5, 8, 9. Vertipecten alexcla'rki Addicott, n. sp. (p. 26). 1. Length 89 mm, height 80 mm. USNM 646528. USGS 10c. M3285. 3. Length 63 mm, height 82 mm. USNM 646529. USGS 10c. M3772. Specimen deformed. 5. Height 40 mm. USNM 646530. USGS 10c. M3579. 8. Length 66 mm, height 69 mm. USNM 646531. USGS loc. M3281. 9. Height 68 mm. USNM 646532 USGS 10c. M3280. 2, 10. Crassostrea eldridgei ynezana (Loel and Corey) (p. 27). 2. Immature specimen. Length 41 mm, height 45 mm. USNM 646533. USGS 10c. M3280. 10. Length 60 mm, height 101 mm. USNM 646534. USGS 10c. M3280. 4, 7. Pecten (Pecten) sanctaecruzensis Arnold (p. 24). 4. Length 65 mm. USNM 646535. USGS 10c. M4467. 7. Height 36 mm. USNM 646536. USGS 10c. M4467. 6. Lucinoma acutilineata (Conrad) (p. 28). Length 35 mm, width 33 mm. USNM 646537. USGS 10c. M3280. GEOLOGICAL SURVEY PROFESSIONAL PAPER 791 PLATE 2 CRA SSOSTREA, L U CIN 0M A, PEC TEN, VER TIPEC TEN PLATE 3 [All specimens are from the Wygal Sandstone Member of the Temblor Formation] FIGURES 1, 4. Vertipecten alexclarki Addicott, n. sp. (p. 26). 1. Height 84 mm. USNM 646538. USGS 10c. M3281. 4. Length of fragment 66 mm. USNM 646539. USGS 10c. M3985. Crenomyt'ilus? cf. 0.? arnoldi (Clark) (p. 24). Length 81 mm, height 37.5 mm. USNM 646540. USGS loc. M3978. Lucinoma acutilineata (Conrad) (p. 28). 3. Length 33 mm, height 31.5 mm. USNM 646541. USGS 10c. M3280. 7. Length 34 mm, height 29.5 mm. USNM 646542. USGS 10c. M3280. Miltha (Miltha) sanctaecrucis (Arnold) (p. 28). 5. Length 65 mm, height 64 mm. USNM 646543. USGS 10c. M3579. 10. Length 68 mm, height 65 mm. USNM 646544. USGS Ice. 9432. Here excavata (Carpenter) (p. 27). 6. Length 16 mm, height 15.5 mm. USNM 646545. USGS 10c. M3578. 8. Length 17 mm, height 17 mm. USNM 646546. USGS 10c. M3578. 9. Length 17.5 mm, height 17.5 mm. USNM 646547. USGS 10c. M3578. GEOLOGICAL SURVEY PROFESSIONAL PAPER 791 PLATE 5 9 ' 10 CRENOMYTIL US?, HERE, LUCINOMA, MILTHA, VERTIPE'CTEN PLATE 4 [Specimens are from the Wygal Sandstone Member of the Temblor Formation unless otherwise indicated] FIGURE 1. Spisula n. sp.? (p. 30). Length 91 mm, height 68 mm. USNM 646548. USGS 10c. M3579. 2, 7. Felaniella harfordi (Anderson) (p. 29). 2. Length 19.5 mm, height 17.5 mm. USNM 646550. USGS 10c. M3280. 7. Length 24 mm, height 21.5 mm. USNM 646551. USGS loc. M3280. 3, 4, 8. Spisula cf. S. albam'a (Conrad) (p. 29). 3. Length 47 mm, height 35.5 mm. USNM 646552. USGS 10c. M3578. 4. Length 37 mm, height 28.5 mm. USNM 646553. USGS loc. M3578. 8. Length 36 mm, height 28 mm. USNM 646554. USGS 10c. M3979. 5. Spisula ramonensis Packard (p. 30). Length 49 mm, height 37 mm. USNM 646555. USGS 10c. M3280. 6. Spisula albaria (Conrad) (p. 29). Length 46.5 mm, height 36.5 mm. USNM 646556. USGS loc. M1802, Astoria Formation, middle Miocene, Oregon. 9. Solen aff. S. gravidus Clark (p. 30). Height 25.5 mm. USNM 646557. USGS loc. M3280. 10, 12. Tellina (Olcesia) piercei (Arnold) (p. 31). 10. Height 49 mm. USNM 646558. USGS 10c. M3978. 12. Length 63 mm, height 43.5 mm. USNM 646559. USGS 10c. M3578. 11. Tellina (Oudardia) emacerata Conrad (p. 32). Length 35 mm, height 20 mm. USNM 64.6560. USGS loc. M3578. 13. Tellina? cf. T. vaneouve’rensis Clark and Arnold (p. 33). Height 19 mm. USNM 646561. USGS loc. M3578. l4. Spisula cf. S. rushi Wagner and Schilling (p. 30). Length 83 mm, height 63.5 mm. USNM 646562. USGS 10c. M3979. 15. Crenomytilus expansus (Arnold) (p. 24). Length 116 mm, height 60 mm. USNM 646549. USGS Ice. 9427. GEOLOGICAL SURVEY PROFESSIONAL PAPER 791 PLATE 4 14 15 CRENOMYTILUS, FELANIELLA, SOLEN, SPISULA, TELLINA, TELLINA? PLATE 5 [Specimens are from the Wygal Sandstone Member of the Temblor Formation unless otherwise indicated] FIGURES 1, 3, 6, 9-11. Heteromacoma rostellata (Clark) (p. 34). 1. Suborbicular form. Length 55.5 mm, height 47 mm. USNM 646563. USGS loc. M3280. 3. Suborbicular form. Length 51 mm, height 46 mm. USNM 646564. USGS 10c. M3280. 6. Suborbicular form. Length 53.5 mm, width 45 mm. USNM 646565. USGS 10c. M3281. 9. Elongate form. Length 57.5 mm, height 42 mm. USNM 646566. USGS 10c. M3772. 10. Subquadrate form. Length 65 mm, height 44 mm. USNM 646567. USGS 10c. M3280. 11. Subquadrate form. Length 67 mm, height 55 mm. USNM 646568. USGS loc. M3280. 2, 8. Tellina (Tellinella) tenuilineata Clark (p. 32). 2. Length 53.5 mm, height 29 mm. USNM 646569. USGS 10c. M3578. 8. Length 53 mm, height 28.5 mm. USNM 646570. USGS Ice. 9432. 4, 7. Macoma arctata (Conrad) (p. 33). 4. Length 56.5 mm, height 37 mm. USNM 646571. USGS 10c. M3280. 7. Length 56 mm, height 38 mm. USNM 646572. USGS 10c. M3280. 5. Pseudocardium? sp. (p. 29). Length 45 mm, height 42.5 mm. USNM 646573. USGS 10c. M4466. 12, 13. Tellina (Olcesia) piercei (Arnold) (p. 31). 12. Length 74.5 mm, height 55 mm. USNM 646574. USGS loc. M1599. Olcese Sand, Kern River area, California, middle Miocene. 13. Length 76 mm, height 51 mm. USNM 646575. USGS 10c. M1599. Olcese Sand, Kern River area, Cali- fornia, middle Miocene. GEOLOGICAL SURVEY PROFESSIONAL PAPER 791 PLATE 5 12 13 HETEROMACOMA, MACOMA, PSEUDOCARDIUM?, TELLINA PLATE 6 [All specimens are from the Wygal Sandstone Member of the Temblor Formation] FIGURES 1, 3, 9, 12, 13. 2, 4, 14. Dosim'a, (Dosim'a) whitneyi (Gabb) (p. 36). 1. Length 71 mm, height 61 mm. USNM 646576. USGS Ice. 9432. 3. Length 44.5 mm, height 48.5 mm. USNM 646577. USGS 10c. M3280. 9. Length 48 mm, height 46 mm. USNM 646578. USGS 10c. M3280. 12, 13. Length 58 mm, height 58.5 mm. USNM 646579. USGS 10c. M3280. Heteromacoma rostellata (Clark) (p. 34). 2. Subquadrate form. Height 54 mm. USNM 646580. USGS loc. M3280. 4. Elongate form. Length 74 mm, height 47 mm. USNM 646581. USGS 10c. M3280. 14. Suborbicular form. Length 74 mm, height 60 mm. USNM 646582. USGS 10c. M3280. Securella cf. S. cryptolineata (Clark) (p. 37). Length 40 mm, height 32 mm. USNM 646583. USGS 10c. M3280. Amiantis n. sp.? (p. 36). Length 50 mm, height 38.5 mm. USNM 646584. USGS 10c. M3578. Tellina cf. T. townsendensis Clark (p. 32). Length 32 mm, height 17.5 mm. USNM 646585. USGS 10c. M3578. Pitar (Katherinella) cf. P. (K.) caliform'ca (Clark) (p. 37). Length 67 mm, height 41.5 mm. USNM 646586. USGS 10c. M3280. Pitar sp. (p. 37). Length 58 mm, height 45.5 mm. USNM 646587. USGS 10c. M3978. ER 791 PLATE 6 u GEOLOGICAL SURVEY PROFESSIONAL PAP 12 14 AMIANTIS, DOSINIA, HETEROMACOMA, PITAR, SECURELLA, TELLINA PLATE 7 [Specimens are from the Wygal Sandstone Member of the Temblor Formation unless otherwise indicated] FIGURES 1, 4, 8, 13, 14. Amiantis mathewsoni (Gabb) (p. 35). 1. Length 54 mm, height 61.5 mm. USNM 646588. USGS 10c. M3280. 4, 8. Length 64.5 mm, height 60.5 mm. USNM 646589. USGS loc. M3978. 13. Hinge, natural size. USNM 646590. USGS loc. M3978. 14. Length 67 mm, height 60.5 mm. USNM 646591. USGS 10c. M3772. 2, 3, 11. Amiantis n. sp.? aff. A. diabloensis (Anderson) (p. 35). 2, 3. Length 54.5 mm, height 46.5 mm. USNM 646592. USGS 10c. 6627. Olcese Sand, Kern River area, California, middle Miocene. 11. Hinge, natural size. UCMP 14090. UCMP loc. B1599. Olcese Sand, Kern River area, California, middle Miocene. 5, 10. Dosim'a, (Dosim'a) whitneyi (Gabb) (p. 36). 5. Length 45.5 mm, height 42 mm. USNM 646593. USGS loc. M3280. 10. Hinge, natural size. USNM 646594. USGS 10c. M3772. 6, 7. Amiantis n. sp.? (p. 36). ‘ 6. Length 59 mm, height 43 mm. USNM 646595. USGS 10c. M3578. 7. Length 54 mm, height 41 mm. USNM 646596. USGS 10c. M3578. 9, 12. Clementia (Egesta) pertenuis (Gabb) (p. 36). 9. Length 65.5 mm, height 54 mm. USNM 646597. USGS 10c. M3280. 12. Length 72.5 mm, height 53.5 mm. USNM 646598. USGS Ice. 9432. GEOLOGICAL SURVEY AMIANTIS, CLEMEN TIA, DOSINIA PLATE 8 [All specimens are from the Wygal Sandstone Member of the Temblor Formation] FIGURES 1, 3, 19. 5. 6, 7, 9, 11, 15, 16. 8, 10, 17, 18. 12, 13. 14. 20. Clementia (Egesta) pertenm‘s (Gabb) (p. 36). 1. Length 70.5 mm, height 62 mm. USNM 646599. USGS loc. M4466. 3. Length 71 mm, height 62.5 mm. USNM 646600. USGS loc. M4466. 19. Length 71.5 mm, height 60 mm. USNM 646601. USGS 10c. M3578. Tegula n. sp. (p. 17). 2. Height 17 mm, width 19.5 mm. UCR 1235. UCR Ice. 1235. 4. Height 20.5 mm, width 21.5 mm. USNM 646602. USGS 10c. M3578. Cerithiid? (p. 17). Height 19 mm. USNM 646603. USGS 10c. M3578. Crepidula cf. C. ungtma Dall (p. 18). 6, 15. Width 25.5 mm. USNM 646604. USGS 10c. M3280. 7. Length 28.5 mm, width 20 mm. USNM 64660.5. USGS 10c. 3979. 9, 11, 16. Length 37.5 mm, width 20 mm. USNM 646606. USGS 10c. M3281. Calyptraea diegoana (Conrad) (p. 17). 8, 10. Height 10.5 mm, width 22 mm. USNM 646607. USGS 10c. M3280. 17, 18. Height 19 mm, diameter 48 mm. USNM 646608. USGS 10c. M3281. Natica (Natica) n. sp. (p. 19). 12. Height 21 mm, width 20.5 mm. USNM 646609. USGS 10c. M3280. 13. Height 28 mm, width 25.5 mm. USNM 646610. USGS loc. M3280. Dentalium lameensis Hickman (p. 38). Length 23.5 mm, width 8 mm. USNM 646611. USGS loc. M3772. Panopea ramonensis (Clark) (p. 37). Length 92 mm, height 52.5 mm. USNM 646612. USGS loc. M3280. GEOLOGICAL SURVEY PROFESSIONAL PAPER 791 PLATE 8 19 CAL YPTRAEA, CERITHIID?, CLEMEN TIA, CREPIDULA, DEN TALI UM, NA TICA, PANOPEA, TEGULA PLATE 9 [Specimens are from the Wygal Sandstone Member of the Temblor Formation unlms otherwise indicated] FIGURES 1, 2, 8, 9, 14. Bruclarkia seattlensis Durham (p. 21). 1. Height 46 mm, width 33 mm. USNM 646613. USGS loc. M3979. 2. Height 45.5 mm, width 30.5 mm. USNM 646614. USGS loc. M3772. 8, 9. Height 30.5 mm, width 24.5 mm. USNM 626615. USGS 10c. M3772. 14. Height 33 mm, width 26 mm. USNM 646616. USGS loc. M3280. 3. Bmclarkia barkerioma forma sanctacruzana (Arnold) (p. 21). Height 46 mm, width 28.5 mm. UC‘MP 36556. UCMP 10c. B1598. Olcese Sand, Kern River area, California, middle Miocene. 4. Neverita (Glossaulax) thomsonae Hickman (p. 19). Height 32.5 mm, width 38.5 mm. USNM 646617. USGS 10c. M3978. 5, 6. Siphonalia? sp. (p. 21). Height 25 mm, width 18 mm. USNM 646618. USGS 10c. M3280. Specimen deformed. 10. Polinices n. sp.? (p. 19). Height 29 mm, width 29 mm. USNM 646619. USGS 10c. M3280. Specimen deformed. 7, 11, 12. Kelletia? sp. (p. 21). 7. Height 32 mm, width 30 mm. USNM 646620. USGS loc. M3280. 11, 12. Width 28 mm. USNM 646621. USGS 10c. M3772. 13. Calicantharus cf. C. branne'ri (Clark and Arnold) (p. 20). Height 31.5 mm, width 23 mm. USNM 646622. USGS 10c. M4471. 15, 20, 21. Bruclarkia gravida (Gabb) (p. 21). 15. Lectotype. Height 32 mm, width 21 mm. ANSP 4345. San Ramon Sandstone, Contra Costa County, Calif., early Miocene(?). 20, 21. Height 42 mm, width 29 mm. ANSP 4345a. San Ramon Sandstone, Contra Costa County, Calif., early Miocene(?). 16, 17. Sinum cf. S. scopulosum (Conrad) (p. 19). 16. Height 20 mm, width 22 mm. USNM 646623. USGS 10c. M3578. Specimen deformed. 17. Height 17 mm, width 20 mm. USNM 646624. USGS loc. M3578. 18, 19. Ficus cf. F. modesta (Conrad) (p. 20). 18. Height 32 mm, Width 22 mm. USNM 646625. USGS 10c. M3578. Specimen deformed. 19. Height 33.5 mm, width 28.5 mm. USNM 646626. USGS loc. M3978. Specimen deformed. 22, 23. Bruclarkia columbiana, (Anderson and Martin) (p. 8). Height 43.5 mm, width 32.5 mm. UCR 1106. UCR Ice. 1106. Cymric Shale Member of the Temblor Formation, Oligo- cene. 24. Calicantha’rus dalli (Clark) (p. 20). Height 53.5 mm, width 36 mm. USNM 646628. USGS 10c. M3772. GEOLOGICAL SURVEY PROFESSIONAL PAPER 791 PLATE 9 20 24 BR UCLARKIA, CALICANTHAR US, FICUS, KELLETIAZ NEVERITA, POLINICES, SINUM, SIPHONALIA?