7,4 Pée, 7 DAY 3 v/‘I 9. "his Geology of the San Francisco North Quadrangle, California GEOLOGICAL SURVEY PROFESSIONAL PAPER 782 GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE, CALIFORNIA San Francisco Bay area aerial view east, taken November 22, 1967. The Golden Gate channel is at bottom center; San Francisco at bottom right. San Pablo Bay at left center. Geology of the San Francisco North Quadrangle, California By JULIUS SCHLOCKER GEOLOGICAL SURVEY PROFESSIONAL PAPER 782 The distribution and character of the bedrock and surficial deposits in the northern part of the City of San Francisco and southern Marin County, Calif., including a description of the Franciscan Formation in its type area and notes on engineering geology in an urban area 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. 73-600283 fi' U.S. GOVERNMENT PRINTING OFFICE: l97‘ 0—534—039 For sale by the Superintendent of Documents, US Government Printing Office Washington, DC. 20402 — Price $4.80 (paper cover) Stock Number 2401—02490 CONTENTS Page Abstract ........................................................................................ 1 Introduction .................................................................. 2 Purpose .................................................................. 2 Location and topography . 3 Climate .................................................. 5 Early settlement ____________________________ 6 Previous work ............................ 7 Acknowledgments ______ 8 Fieldwork .................... 8 Methods of mapping ........................................................... 8 Stratigraphy ................................................................... 9 Bedrock __________________________________ 9 Franciscan Formation ................................................ 9 Clastic sedimentary rocks ................................ 11 Sandstone ....................... 11 Matrix ................................................... 13 Detrital grains .................................... 14 Shale 18 Conglomerate .............................................. 20 Metamorphosed sedimentary rocks .......... 20 Occurrence 23 Northeastern San Francisco .............. 23 Central highlands of San Francisco.. 24 Cliff House to Bakers Beach ...... 7 ........ 24 Fort Point to Potrero Hill .................. 25 Marin Peninsula ................................ 25 Angel and Alcatraz Islands and ........ 26 Tiburon Peninsula .......................... 26 Origin .......................................................... 26 Environment of deposition . 26 Source area of detritus ................ 27 Age ..................................................... .. 27 Fossils ................................................. 27 Significance of potassium feldspar content .............................................. 28 Greenstone 29 Megascopic features .................................. 29 Mineralogy-microscopic features ............ 32 Weathering and hydrothermal alteration 33 Occurrence .................................................. 33 Marin Peninsula 33 San Francisco ...................................... 34 Angel Island and vicinity ........................ 35 Origin ......................................... .. 36 Radiolarian chert and shale .............................. 36 Megascopic features .................................... 37 Chemical composition 41 Microscopic features .................................. 43 Chert .................................................... 43 45 Hydrothermal alteration .......................... 45 Weathering .................................................. 46 Origin _________________________ 47 Origin of silica .................................... 47 Origin of chert ...................................... 47 Origin of shale ...................................... 48 Page Stratigraphy—Continued Bedrock—Continued Franciscan Formation—Continued Radiolarian chert and shale—Continued Origin—Continued Origin of color ....................................... 48 Age ......................................................... 48 Occurrence ........................................... 49 Marin Peninsula .............. 49 Angel Island and vicinity 49 San Francisco ..................... 49 Metamorphic rocks ............................................. 50 Metamorphic rocks on serpentine borders .................................................... 51 Parent rock types ................................. 51 Metamorphic rocks within serpentine bodies ........................................................ 53 Metamorphic rocks distance from known serpentine bodies ..................................... 53 Tectonic setting of metamorphic rocks... 55 Hydrothermal alteration and weathering 55 Sheared rocks ............................................................... 55 Occurrence . ...................................... 56 Serpentine .................................................................. 56 Megascopic features ............... 56 Mineralogy and petrography ............................. 58 Bastite ........................................................... 58 Relict olivine and orthopyroxene. 58 Blue color of serpentine ............................... 59 Antigoritic serpentine ................................. 59 Veins in serpentine ........ 60 Sheared serpentine ........................ . ............ 60 Chemical composition ......................................... 61 Weathering and hydrothermal alteration ...... 62 Shape of serpentine bodies ................................ 63 Origin .................................................................... 64 Age ............ 65 Pyroxenite ..................................................... 65 Gabbro .................................................................. 65 Surficial deposits ................................................................ 66 Colma Formation ........................................................ 66 Megascopic features ............................ 67 Composition and physical properties..... 69 Weathering ............................................ 69 Occurrence ................... 69 Origin ......................................................... 71 Correlation with nearby deposits ....................... 73 Age relation to the ancient Sacramento River 73 Related deposits .................................................. 74 Older beach deposits 74 Age ......................................................................... 75 Modern beach deposits ............................................... 76 Source of sand on Ocean Beach ......................... 77 Dune sand ..................................................................... 78 Origin and age ....................................................... 80 V VI CONTENTS Page Page Stratigraphy—Continued Stratigraphy—Continued Surficial deposits—Continued Surficial deposits—Continued Slope debris and ravine fill ........................................ 80 Landslide deposits ...................................................... 86‘ San Francisco Bay sediments... 81 Characteristics of landslides .............................. 87 Transverse sand bars ....... 82 Artificial fill ................................................................ 88 Bay mud and clay ................................................ 83 Structure ....................... 90 Engineering properties of bay mud Seismicity 93 and clay .................................................... 84 Engineering geology ..................................................... 94 Stratigraphic correlation, fossils and age 84 References ..................... 100 Alluvium ...................................................................... 85 Index ...................................................... . 107 ILLUSTRATIONS [Plates are in pocket] FRONTISPIECE. Aerial view of San Francisco Bay area PLATE 1. Geologic map of the San Francisco North quadrangle. 2. Composition and grain size of surficial deposits, San Francisco North quadrangle. 3. Bedrock-surface and landslide locality map of the San Francisco North quadrangle. Page FIGURE 1. Index maps of part of central California and the San Francisco Bay area 4 2—5. Photographs of: 2. Angel Island, Raccoon Strait, and Tiburon Peninsula 5 3. Marin Peninsula and part of Sausalito ..... 5 4. Belvedere Island and Tiburon Peninsula .............. 5 5. Mount Caroline Livermore and southwestern slope of Angel Island 5 6. Map showing distribution of Franciscan Formation, related rocks, principal structural features of western California ........... ________ 10 7—12. Photographs of: 7. Horseshoe Bay and vicinity, southern Marin Peninsula ................................................................................. 11 8. Shale and thin-bedded sandstone of the Franciscan Formation showing graded and distorted bedding and laminae. Russian Hill, San Francisco ................................................................................................. 12 9. Jointed thick-bedded sandstone and interbedded 3-foot-thick section of shale and thin-bedded sandstone, Franciscan Formation, Telegraph Hill, San Francisco 12 10. Sheared and shattered sandstone and shale of the Franciscan Formation. North of Golden Gate Bridge, Marin Peninsula .............................................................................................................................. 13 11. Sheared shale of the Franciscan Formation containing large sandstone masses. Near Sausalito, Marin Peninsula ................................................................................................ 13 12. Shale fragments in coarse-grained graywacke sandstone of the Franciscan Formation. Laguna Honda Reservoir area, San Francisco ............................................................................................................ 14 13. Triangular diagram showing classification of sandstone samples of the Franciscan Formation __________________________ 14 14. Map showing locations of sandstone samples of the Franciscan Formation and potassium feldspar content... 16 15—61. Photographs of: 15. Semischistose coarse-grained graywacke sandstone of the Franciscan Formation. Campbell Point, Angel Island ................................................................................................................................................... 20 16. Massive sandstone of the Franciscan Formation. Telegraph Hill, San Francisco .................................... 23 17. Thick-bedded sandstone, laminated sandstone, siltstone, and shale of the Franciscan Formation, near Bakers Beach, San Francisco ........................................................................................................... 24 18. Thick- bedded sandstone interbedded with shale and thin- bedded sandstone of the Franciscan Formation. North of Golden Gate Bridge, Marin Peninsula 25 19. High cliff 1n greenstone 1n the Franciscan Formation. Lime Point, Marin Peninsula .............................. 29 20. Greenstone and radiolarian chert of the Franciscan Formation Lime Point, Marin Peninsula ............ 29 21. Close random fracturing in greenstone of the Franciscan Formation. Near Sausalito, Marin Peninsula .............................................................................................................................................. 30 22. Basalt pillows. North of Golden Gate Bridge, Marin Peninsula ...... ,. 31 23. Pillow basalt of the Franciscan Formation. Fort Baker, Marin Peninsula .................................................. 32 24. Radiolarian chert lying on basaltic greenstone, Franciscan Formation. Near Sausalito, Marin Peninsula ........................................................................................................................................... 34 25. Internal structure of radiolarian chert bed of the Franciscan Formation. Roadcut on Sausalito lateral, Marin County. ..................................................................................... 38 26. Radiolarian chert and shale of the Franciscan Formation, Twin Peaks, San Francisco .......................... 39 FIGURES 27—61. TABLE 62. 63. 64. HH ["PSDWHP‘P‘PWNT‘ CONTENTS VII Photographs of——Continued Page 27. Pinching and swelling in radiolarian chert beds of the Franciscan Formation. Roadcut on Sausalito lateral, Marin Peninsula ............................................................................................................................... 39 28. Chevron folds in radiolarian chert of the Franciscan Formation. Near Golden Gate Bridge, Marin Peninsula ................................................................................................................................................ 39 29. Radiolarian chert containing possible drag folding at Lime Point, Marin Peninsula ................................. 40 30. Massive chert in the Franciscan Formation, Sunset Heights, San Francisco ............................................ 42 31. Quartz veins in radiolarian chert lying on altered greenstone, Twin Peaks, San Francisco .................... 42 32. Irregular band of hydrothermally altered radiolarian chert of the Franciscan Formation, Mount Sutro, San Francisco ........................................................................................................................ 46 33. Thin soil developed on radiolarian chert and shale. North of the Golden Gate Bridge, Marin Peninsula ........................................................................................................................................... 46 34. Franciscan Formation exposed on the north shore of Golden Gate, Marin Peninsula .............................. 49 35. Central highlands area of San Francisco showing peaks of radiolarian chert on the Franciscan Formation ................................................................................. 50 36. Radiolarian chert and sandstone 1n the Franciscan Formation, 17th Street, Mount Olympus area, San Francisco ................................................................................................................................................. 50 37. Radiolarian chert and shale, Sunset Heights, San Francisco ..... 50 38. Serpentine body and metamorphic rocks, Angel Island ...................................................... 52 39. Shoreline along Bakers Beach, the Presidio, Fort Point, and vicinity, San Francisco. 56 40. Serpentine in a landslide. Fort Point, San Francisco ............................................................... 57 41. Serpentine nodule in sheared serpentine. Presidio, San Francisco .............. 57 42. Clinochrysotile veinlets in basaltic serpentine. Fort Point, San Francisco .................................................. 61 43. Sheared serpentine containing large augens of altered and sheared gabbro-diabase. Potrero Hill, San Francisco ................................................................................................................................................ 63 44. Serpentine and sandstone Potrero Hill, San Francisco 63 45. Colma Formation overlying greenstone of the Franciscan Formation. Sutro Reservoir, San Francisco 68 46. Rubble deposits and crossbedding 1n the Colma Formation. Twin Peaks, San Francisco ........................ 68 47. Colma Formation. Angel Island ............................................................................................................................ 70 48. Colma Formation filling channel in greenstone. Twin Peaks, San Francisco... 71 49. Raised older beach deposits covered by sand dune. Presidio, San Francisco .............................................. 75 50. Ocean Beach .......................................................................... .. ...... 76 51. Berm development on Ocean Beach .................................................................................... 76 52. Dune sand exposed in excavation for Brooks Hall. Civic Center, San Francisco ...................................... 78 53. Sand dunes. Sunset District, San Francisco ..................................................................................................... 79 54. Unstable embankment of dune sand on lee side of Sunset Heights, San Francisco .................................... 79 55. Sand dune, Aquatic Park area, San Francisco ............... 79 56. Slope debris on weathered greenstone of the Franciscan Formation ............................................................ 81 57. Bay mud, showing desiccation cracks, lying under artificial fill, San Francisco ........................................ 84 58. Alluvium lying on Colma Formation. Twin Peaks, San Francisco .. ........ 85 59. Steeply cut slopes in serpentine, Potrero Hill, San Francisco .. ...... 86 60. Landslide. South Bay of Golden Gate, San Francisco ...................... . .. .. 87 61. Landslide. One mile northwest of Lime Point, Marin Peninsula .................................................................. 88 Map showing shoreline of San Francisco in 1853, present shoreline, and areas formerly covered by water that are now artificially filled present shoreline .................................................................................................... 89 Generalized geologic map of parts of the San Francisco North and South quadrangles showing shear zones 91 Photographs of differential settlement of buildings in former Mission Swamp. Near Sixth and Folsom Streets, San Francisco ....... 95 TABLES Page Modal composition of sandstone samples of the Franciscan Formation ................................................................ 14 Potassium feldspar in sandstone of the Franciscan Formation ........................ 17 Frequency of heavy minerals in sandstone of the Franciscan Formation ......................................................... 18 Analyses of sandstone, shale, and a phosphate nodule from the Franciscan Formation ............................... 19 Chemical composition of sandstone samples of the Franciscan Formation from Angel Island... ..... 22 Analyses of greenstone samples of the Franciscan Formation ................................................................................ 35 Analyses of radiolarian chert and shale samples of the Franciscan Formation ...................................................... 43 Partial mineral composition and bulk density of serpentine ...... 59 Optical properties of olivine and orthopyroxene and suggested composition .......................................................... 59 Approximate semiquantitative spectrochemical analyses of serpentine and nontronite veins in serpentine. 62 Generalized description of engineering properties of map units ......... 96 . Wxxmz-gs» “my. GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE, CALIFORNIA By JULIUS SCHLOCKER ABSTRACT The San Francisco North quadrangle, about 58 square miles in area, includes the north half of the city of San Francisco, the south tip of Marin and Tiburon Peninsulas, Belvedere, Angel, and Alcatraz Islands, and part of Treasure Island. The topog- raphy of San Francisco proper is characterized by rolling hills, rounded slopes, and broad valleys, whereas other parts of the quadrangle are characterized by steeper and more rugged slopes, rounded ridges and spurs, and steep-sided V-shaped canyons. San Francisco Bay covers nearly half the quad- rangle. It is connected to the Pacific Ocean by the Golden Gate channel. Water depths in the tidal channels generally range from 100 to more than 360 feet below mean sea level; else- where water depths are generally less than 30 feet. Two distinct groups of rocks, bedrock and surficial deposits, differ greatly in age, lithology, and topographic expressions. The bedrock consists of the Franciscan Formation and asso- ciated intrusive serpentine and gabbro-diabase of Jurassic(?) and Cretaceous age. The Franciscan Formation is at least 16,000 feet thick and consists mostly of graywacke sandstone, shale and siltstone, and mafic volcanic rocks, and lesser amounts of radiolarian chert, conglomerate, limestone, and glaucophane schist. Exposures of the Franciscan constitute 15— 20 percent of the land area of the quadrangle, generally on hills. The surficial deposits, of Quaternary age, consist largely of unconsolidated dune sand and water-laid sand, mud, and clay. However, slopewash deposits, alluvium, landslide debris, and artificial fill are fairly extensive. Above sea level, surficial deposits mantle slopes and fill valleys to depths of a few feet to more than 100 feet. In San Francisco Bay, surficial deposits range in thickness from 100 to about 300 feet. The major rock types of the Franciscan Formation in this quadrangle which are mappable at a scale of 1:24,000 are clastic sedimentary rocks, greenstone, radiolarian chert, meta- morphic rocks, and sheared rocks. The elastic sedimentary rocks of the Franciscan Formation consist predominantly of massive graywacke sandstone beds separated by sequences of shale and thin-bedded sandstone. Conglomerate lenses and zones, as much as 10 feet thick, are common on Angel and Belvedere Islands and on Tiburon Peninsula. Pillow lava and radiolarian chert are interbedded with the clastic rocks. Metamorphic equivalents of these rock types are present on Tiburon Peninsula and Angel Island. Although the elastic rocks of the Franciscan Formation con- sist mostly of alternating massive and thin beds, two excep- tionally thick sections of predominantly thin-bedded shale and sandstone, each totalling more than 1,000 feet, crop out in the northeastern part of San Francisco. The clastic rocks of the Franciscan Formation probably were deposited by turbidity currents in deep water in an orogenically active eugeosyncline lying west of a continental mass. A source area including metamorphic rocks is indicated by the clasts in the sandstones and conglomerates. Rocks whose composition is similar to the sand-size detritus in Franciscan sandstone are exposed in the Coast Ranges, Klamath Moun— tains, and Sierra Nevada. The pebbles in some conglomerates are very similar to sandstones and shales of the Franciscan, indicating that the conglomerates may have been derived from topographically high areas created by orogeny during Fran- ciscan time. The term “greenstone” is used to include all volcanic and related intrusive rocks in the Franciscan Formation. These consist mostly of fine. to medium-grained metabasalts, in large part altered to pumpellyite-albite-chlorite rock. Most basalts show pillow structure; a few are pyroclastic. Pillow structure, the presence of marine chert and limestone between pillows, and the interbedding of flows and marine sediments through- out the formation suggest a long history of submarine vol- canism. Thin chert and shale beds and less common massive chert bodies are closely associated with the greenstones. Radiolaria are common in the chert, less common in the shale. The larger deposits of radiolarian chert are closely associated with green- stone, which is believed to be the ultimate source of the silica from which the chert was formed. Metamorphic rocks include slate, phyllonite, phyllite, fine- to coarse-grained schist, and granofels, commonly having relict textures. Many masses of metamorphic rocks occur along the borders of serpentine bodies or as tectonic inclusions from diapiric movement in the serpentine. They are believed to be metamorphic equivalents of Franciscan rocks and may have been formed under high shearing stresses or by hydrothermal alteration. The map unit termed “sheared rocks” consists of a soft intensely sheared matrix of shale or serpentine that has moved plastically. It encloses hard blocks, as much as hundreds of feet in diameter, of various rock types found elsewhere in the Franciscan Formation. The blocks are rounded and slicken- sided externally, but most are relatively unsheared internally. Blocks greater than 20 feet in diameter are generally shattered internally. The map unit designated serpentine also includes a few small bodies of calc-silicates (rodingites), pyroxenite, and gabbro. Serpentine intrudes the Franciscan Formation and is widely associated with it. The serpentine was derived from peridotite, and its present position is probably due to tectonic movement at relatively high pressures and low temperatures. Shearing is widespread within the serpentine bodies. Unconsolidated late Pleistocene and Holocene surficial deposits cover about 80—85 percent of the land area of the quadrangle. These deposits are mostly sand, but they include rubbly slope debris and ravine fill, bay mud and clay, land- slide deposits, and artificial fill. The Colma Formation is the oldest of the surficial deposits. It includes a group of marine and dune sands that have grossly 1 2 GEOLOGY OF THE SAN FRANCISCO-NORTH QUADRANGLE similar physical properties and that occupy approximately the same stratigraphic position; however, the formation probably includes several marine deposits related to different sea levels. It also contains alluvial and colluvial deposits. The Colma Formation generally is horizontally bedded. The formation lies stratigraphically below latest Pleistocene and Holocene deposits. A raised beach sand, probably deposited during a warm interglacial epoch when sea level was higher than it is now, has been exposed on the Presidio Military Reservation. This deposit was mapped separately from the Colma Formation because it is less cemented than the Colma Formation and its grains appear to be less weathered. Surficial deposits mapped as alluvium are very restricted in extent and distribution, largely because of low surface runoff due to low rainfall and a cover of highly permeable material over large areas. Most of the alluvium in the quadrangle is related to old drainage systems and is now moderately dis~ sected. The alluvium is composed of medium sand mixed with silt and clay. Beach sand and coarser beach deposits occur along all shores except along the east edge of San Francisco and at Sausalito. Beach deposits vary seasonally—even hourly—in thickness and extent, depending on the nature of the waves and the supply of sediments. Along the Pacific shore of San Francisco, the probable sources of the beach sand and the related onshore dunes are the poorly consolidated Pliocene and Pleistocene Merced Formation, the younger formations along the shore to the south, and the sands of the continental shelf. The sands of the continental shelf probably were deposited by the ancestral Sacramento—San Joaquin River, during the Wisconsin Glaciation, when sea level was lower. Dune sand, swept by prevailing westerly winds, underlies more than half the city of San Francisco and extends as far east as the area between Telegraph and Rincon Hills. It has been deposited more than 600 feet above sea level and attains a thickness of about 150 feet. The chief source of dune sand was the Ocean Beach area. Slope debris derived largely from weathered rock mantles most slopes and fills the adjoining ravines. The slope debris as mapped locally includes soils developed on bedrock and minor amounts of alluvial, eolian, and landslide materials. The debris thickens downhill and locally is as much as 60 feet thick. The ravine-fill deposits, locally, are interbedded with and grade into alluvium. Bay mud, the youngest deposit in San Francisco Bay, con- sists of soft unconsolidated sediment generally containing more than 90 percent of clay- and silt-size detritus. A similar deposit, bay clay, is nearly the same in texture but is more consolidated and contains less water. Bay mud and clay are found on the eastern shores of San Francisco and Sausalito, under surficial deposits and artificial fill adjoining the Bay, and under San Francisco Bay. Landslides are among the chief agents of erosion in the quadrangle. Sheared rocks and serpentine are the most sus- ceptible materials. The numerous landslides have resulted 'from combinations of hilly terrain, unconsolidated surficial deposits or sheared bedrock, abundant highly plastic and swelling clay, occasional periods of prolonged rainfall, fre- quent earthquakes, and disturbance and alteration of the terrain by man. Artificial fill is extensive along the margins of San Francisco Bay, more than 3 square miles of land having been created by dumping artificial fill on the gently shelving tidal flats. The average thickness of the fill in San Francisco, north of China Basin, is about 10 feet; south of China Basin it is about 60 feet. The fill consists of dune sand, alluvium, debris from the bay, and manmade debris. Serious engineering problems in areas of artificial fill have been caused by differential sub- sidence. The San Francisco North quadrangle lies in the western part of the Coast Ranges near the west edge of the North American crustal plate. The oldest exposed rocks in this part of the plate are those of the Franciscan Formation and its age correlative, rocks of the Great Valley sequence. The plate is bounded on the west by the San Andreas fault zone and far- ther south on San Francisco Peninsula by the Pilarcitos fault (both faults outside the map area), which separate the Fran- ciscan Formation from granitic and associated metamorphic rocks. The Hayward fault zone is nearly parallel to and about 20 miles east of the San Andreas fault zone in this area. East of the Hayward fault, the Coast Ranges consist mostly of Cenozoic marine and nonmarine sedimentary rocks and marine sedimentary rocks of the Mesozoic Great Valley sequence. Movement along the San Andreas and Hayward faults is right lateral and has been responsible for the great earthquakes in this area. The major faults and shear zones within the quadrangle are the northwest-trending Fort Point—Hunters Point and the City College shear zones. The major folds are a syncline that plunges northwest between Russian and Telegraph Hills, an anticline on Marin Peninsula whose axis lies in Richardson Bay, and a broad northwest-plunging syncline on Angel Island and Tiburon Peninsula. ‘ The engineering properties of the natural foundation mate- rials in the quadrangle vary greatly over short distances; so, some building sites are in part on soft bay mud and in part on hard bedrock. Almost every site on the Franciscan Formation encompasses shear zones less than 1 inch to several feet wide. Unconsolidated surficial deposits, which generally contrast greatly in engineering properties from bedrock, cover more than half the land area of the quadrangle. The location of the bedrock surface is therefore important in engineering work. As each construction site presents a unique combinaton of natural conditions, such as slope, hydrology, and foundation materials, detailed geologic investigations and slope stability analyses are needed in the design of specific structures. The San Francisco North quadrangle is in a region of high seismic activity. Structures have been seriously demaged by three major earthquakes since 1865. INTRODUCTION PURPOSE San Francisco is a heavily built up cit-y in a region of active faults and is subject to earthquakes and land- slides. In such an area careful selection of building sites and foundations is imperative. This geologic investigation of the San Francisco North quadrangle was carried out to obtain basic geologic data to aid in construction and other civil engineering problems. The study also contributes to a better understanding of the geology of this part of the Coast Ranges in general and of the Franciscan Formation and surficial deposits in particular. INTRODUCTION 3 On the geologic map (pl. 1), consolidated rocks are delineated by their composition, age, and origin. Such a classification also gives an approximate indication of the engineering properties of each formation. To make the map more useful to the civil engineer, un- consolidated deposits are shown where thick enough to be depicted. This report and the accompanying map should be a useful guide for planning foundation studies necessary in engineering design. LOCATION AND TOPOGRAPHY The San Francisco North quadrangle includes the north half of the city of San Francisco, part of San Francisco Bay, the south tip of Marin and Tiburon Peninsulas, Angel Island, Belvedere Island, Alcatraz Island, and part of Treasure Island (fig. 1). San Francisco Bay covers approximately half of the quadrangle. The bay is connected to the Pacific Ocean by the Golden Gate channel, a deep narrow waterway between the San Francisco Peninsula on the south and Marin Peninsula on the north. The Golden Gate chan- nel and much of the bay are drowned parts of the val- leys of the Pleistocene Sacramento—San Joaquin River and its tributaries, and the Santa Clara Valley. Water depths in the quadrangle vary from shallows of less than 60 feet located south of Rincon Point, north of Treasure Island, southwest of Angel Island, in Rich- ardson Bay, and along a narrow shoal between Alcatraz Island and Fort Point to deeps of more than 300 feet in the Golden Gate channel. The deepest point, approxi- mately 360 feet below sea level, is in the bottom of a hole near the middle of the Golden Gate Bridge (Carl- son and others, 1970). The present topography of the San Francisco North quadrangle has resulted principally from erosion of a lithologically complex terrain and from deposition of sand dunes. Marine and estuarine processes and late Quaternary tectonism also have affected the topo- graphic evolution of the quadrangle. In general, the city area consists of bedrock hills surrounded by 'broad valleys underlain by dune sand and other unconsoli- dated deposits. Hills along the north (Golden Gate) border of the city reach heights of approximately 375 feet above sea level above Point Lobos, in the Presidio, and at Lafayette Square park; heights of approxi- mately 325 feet are attained on Russian Hill, and heights of 275 feet on Telegraph Hill at the northeast edge of the city. Hills on the north border vary in relief from about 150 feet in the Richmond District on the west to 200-225 feet in Russian and Telegraph Hills on the east. Potrero Hill, in the southeast, has a relief of about 225 feet. The central and western parts of the city rise to the south border of the quadrangle to a small group of hills including Sunset Heights, more than 750 feet in altitude, Mount Sutro, 908 feet in altitude, and South Twin Peak, 922 feet in altitude. The central hills of Mt. Sutro and Twin Peaks have a relief of about 850 feet on the east and about 500 feet on the west. Approximately one-third of the city area, the central and western parts, is more than 200 feet above sea level. ' The topography varies from gently rounded hills with broad basins between to the sheer sea cliffs that border the Golden Gate. The rolling terrain of Golden Gate Park and the moderate slopes in the central part of the city in part are modifications by the deposition of tremendous quantities of‘dune sand. The strong northwesterly trend of ridges and valleys that characterizes most of the Coast Ranges is ob- scured in the city itself, although it is suggested by such minor features as Russian and Telegraph Hills and the valley between them, by Hayes Valley between the hills at Lafayette and Alamo Square parks, and by Potrero Hill. The gradient of abandoned channels filled with old alluvium in the Twin Peaks area appears to be much more gentle than that of present-day surface runoff channels. Evidently, Twin Peaks formed by dissection of a gently rolling surface. The shoreline of San Francisco Peninsula has a varied topographic form. The west shore along the open sea is alternately low and high, marked by sandy beaches, such as Ocean and Bakers Beaches, and irreg- ular bedrock cliffs such as those of the Point Lobos— Lands End area and the Fort Point area. Except for Black Point, the north and east shores are low and mostly made land. Prior to filling, high cliffs were present along the shore at Telegraph Hill and at Rin- con Hill. The topography of Marin Peninsula, Belvedere Island, Tiburon Peninsula, and Angel Island is gen- erally steeper and more rugged than that of San Fran- cisco, and the shorelines are generally high bedrock cliffs (fig. 2). Rounded ridges and spurs and steep- sided V-shaped canyons characterize these areas, although a few of the larger canyons widen near their mouths into flat-floored debris-filled valleys. Marin Peninsula has an average relief of 800-900 feet. The highest point in the quadrangle — 1,125 feet above mean sea level — is on a ridge half a mile west of Sausa- lito (fig. 3). The typical northwesterly structural and topo- graphic trend of the northern California Coast Ranges is well shown by Belvedere Island and Tiburon Penin- sula (fig. 4) and by Alcatraz Island, Richardson Bay, and Corinthian Island. This trend can also be found in GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE 123° 122° Point Reyes Farallon Islands (S_AN FRANCISCO COD Pillar Point Man Mao my 6 /\ Angel Island \ A“ .0 QB \ / Sausalito NCISC 7:» FW‘ 0 AN \ ( S Alcatraz .7 §lsland : E o _ . 5 1- V ,- . 37 f .N w :52": *- /-\ ( SAN FRANCISCO 7 L/ij \/,0m 7 E f / 7 /1 J \"3 (/ExV l6". 5Q: f\ l O 5 10 15 20 25 MILES FIGURE 1.——Part of central California and the San Francisco Bay area showing the San Francisco North quadrangle and major structural and physiographic features. INTRODUCTION 5 FIGURE 2,—Angel Island, Raccoon Strait, and Tiburon Penin- sula, viewed northwest. Prominent point in middle distance is Stuart Point, Angel Island. White building at right bottom is the lighthouse at Knox Point. Both points are greenstone of the Franciscan Formation. FIGURE 3.——Marin Peninsula and part of Sausalito. Highest point in the quadrangle, with an altitude of 1,125 feet, is the broad peak right of center. View northwest. FIGURE 4.—Belvedere Island and Tiburon Peninsula, view north. The island is elongated northwestward along the regional structural trend. Cliffs above the shore of Belvedere Island expose Franciscan sandstone to right of trees at the shore, Franciscan greenstone to left. the ridges of Marin Peninsula, although somewhat modified by the erosional features associated with a large transverse stream valley whose mouth is drowned by Rodeo Lagoon. Angel Island is the upper part of a drowned moun- tain cut off from Tiburon Peninsula by Raccoon Strait, a former channel of the Sacramento River. The rounded summit, 781-foot Mount Caroline Livermore, and the broad radial spurs are modified by sharp ridge crests and by steep shorelines (fig. 5). FIGURE 5.—Mount Caroline Livermore and southwestern slope of Angel Island. Viewed toward the northeast. Serpentine and metamorphic rocks are exposed in cliffs above the shore. CLIMATE The influence of the Pacific Ocean keeps the temp- erature moderate and the air breezy, especially in the vicinity of the Golden Gate channel and in those parts of the quadrangle not separated from the ocean by high hills. The following climatic data were obtained from the US. Weather Bureau in July 1968: The average annual rainfall (1931—60) in San Francisco (Civic Center) is 20.78 inches, falling mostly from November through March. About 10 days per month in this period are rainy. Maximum precipitation meas- ured at the Civic Center in 1 hour is 1.07 inches (for the period 1889—1950) ; maximum for 24 hours is 3.65 inches. Drought conditions exist during the summer. Snow is exceedingly rare. The mean annual number of days of frost in the quadrangle is zero. Mean relative humidity for 4 a.m., noon, and 4 pm. is 85, 67, and 71 percent respectively. Annual mean temperature (1931— 60) is 56.8OF. Highest temperature recorded is 101°F; lowest is 27°F. Highest monthly mean temperatures are 59.40F, 62°F, and 61.4°F for August, September, and October, respectively. Highest wind velocity re- corded at the San Francisco Civic Center is 51 miles per hour, for Mount Tamalpais, 120 mph, and for the San Francisco International Airport, 62 mph. (from 6 GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE the southwest). During the rainy season, highest winds are southeast through southwest at 53—62 m.p.h. with gusts to 74 m.p.h. The average storm lasts 3 days. The summer climate is greatly influenced by the constant presence of an anticyclonic high-pressure air- mass west of the quadrangle, associated with the cold California Current and a zone of even colder upwelling water in the Pacific Ocean. Summer air temperatures along the ocean shore are among the coldest within the United States (excluding Alaska), though maximum temperatures 60 miles inland are among the hottest in the United States outside the Sonora—Mojave Desert (Patton, 1956, p. 113). In the summer high relative humidity, precipitation, and thunderstorms are exceed- ingly rare. Summer fog, from late afternoon through the night to late forenoon, is common in many parts of the quadrangle, especially in western San Francisco, Sausalito, and places reached by marine air in the vicinity of the Golden Gate channel. Winter fog of an inland origin is common also; at times it covers much of San Francisco Bay and connected bays and rivers into the Sacramento—San Joaquin Valley. EARLY SETTLEMENT When first discovered by Europeans in 1769, the area was inhabited by two Indian tribes, the Costanoan on San Francisco Peninsula and the Coast Miwok on Marin and Tiburon Peninsulas (Heizer, 1951, p. 40). The Indians used a variety of natural materials, such as seashells, rocks, and plants, for making tools and ornaments. They subsisted on plants and especially on marine life, as is shown by the contents of their refuse mounds (“kitchen middens”) which are numerous on the shores of the bay and somewhat less common on the ocean shore (Nelson, 1909, map 1; Kroeber, 1911, p. 27—28). Some of the mounds are thought to have been started 3,500 years ago. The first Europeans to visit the site of San Fran- cisco were a small party of men led by Sergeant J osé Ortega. They were trying to reach Point Reyes from a base camp of the Portola expedition (1769) in San Pedro Valley, near the ocean 11 miles south of the quadrangle. The main expedition, led by Gaspar de Portola, started in Baja California and was the first attempt by Spain to colonize Alta California. Portola planned to settle at Monterey Bay but failed to recog- nize it and continued northward. On November 1, 1769, the Ortega scouting party left the sand dunes of Ocean Beach behind and climbed Point Lobos only to find their way to Point Reyes, which they had sighted earlier, blocked by the Golden Gate. They then turned eastward across the ridge and from a high hill saw San Francisco Bay. Point Lobos was visited again in 1773 or 1774 by Commandante Rivera with the Father Palou party. In August 1775 a small boat from the ship San Carlos, under the command of Juan Manuel de Ayala, was the first vessel to enter the Golden Gate. The San Carlos sailed into the Golden Gate at night, searching for the boat, and dropped anchor 01f Sausalito. Using Hospital Cove on Angel Island as its main anchorage, the expedition made the first survey of San Francisco Bay and gave names to many of the landmarks in the quadrangle, such as Angel Island and Alcatraz Island. In 1776, Captain Juan Bautista de Anza brought a group of colonists to the area. Captain Anza first left the settlers to rest in Monterey while he and a small party went on to San Francisco. On March 27, 1776, they camped at Mountain Lake and the next day erected a cross at Fort Point to mark Anza’s choice for the site of the military post, or presidio. Fort Point at that time was bounded by high unstable cliffs of ser- pentine. In 1853 these cliffs were cut back in connec- tion with the construction of Fort Scott by the United States, and the surface on which the cross had been erected was lowered. Anza selected a mission site about 3 miles southeast of Fort Point, near a small creek flowing from the slopes of Twin Peaks into a lake. The position of this lake is indicated on the geologic map (pl. 1) by the area of artificial fill that extends westward from about 17th and Mission Streets. The Presidio was established on September 17, 1776, about half a mile southeast of Fort Point. Neglected by Spain and later by Mexico, it never became an impor- tant military post. The mission, however, became a prosperous agricultural and industrial community. Mexico won its independence in 1821 and acquired California as a province in September 1822. In 1834 the mission in San Francisco was secularized. Captain William A. Richardson selected the Yerba Buena Cove area, now artificial fill between Telegraph Hill and Rincon Hill, as a port site in 1835. The port settlement growth soon outstripped that of the Presidio and mission. In 1846 the United States flag was raised in Yerba Buena. In 1847, when the name San Francisco was formally adopted, the population was approximately 400. In May 1848 the town was abandoned by most of its inhabitants, who became gold seekers, but world- wide immigration brought its population to 20,000 by 1849; the dune sand west of the cove started to acquire" its cover of manmade structures. In 1850 more than 500 ships were in the bay, many of them abandoned by their crews. Some of them in the cove became stores and hotels and eventually became part of the artificial filling of the cove, which by 1851 was largely reclaimed. INTRODUCTION ’ 7 Scott (1959) gave additional information on the city’s expansion. The population of San Francisco in 1960 was 750,000; its land area is 46.5 square miles. Sausalito was long a port for Whalers and other mariners where fresh water was obtained. In 1960 its population was 5,500. PREVIOUS WORK The first known geologic observations in the quad- rangle were made in the early 1800’s by voyagers on foreign exploring vessels visiting this remote Spanish port. In 1816 the Russian ship, Rurik, under the com- mand of Otto von Kotzebue, brought Adelbert von Chammisso, a botanist who collected and described serpentine, sandstone, flinty slate (chert?) , and quick- sand (dune sand?) in San Francisco (VanderHoof, 1951, p. 110). The first geologic map of any part of California was one of this quadrangle and vicinity on a scale of 1:217,000 made in 1827 by Edward Belcher and Alex Collie of H .M .S. Blossom, sailing under the command of Captain F. W. Beechey. The map, repro- duced in the popular “Geologic Guidebook of the San Francisco Bay Counties” (VanderHoof, 1951, p. 110), also shows a sketch of the Needles islets, located near Lime Point, and the general distribution of “serpentine, jasper, clayslate, sandstone, and alluvial.” It was accompanied by a two-page text (Buckland, 1839, p. 174—176), which states that Angel Island is “of very confused formation” and identifies the hill west of Mission Dolores as serpentine. The base for the geologic map was a hydrographic and topographic map prepared by the Beechey expedi- tion and published in 1833 by the Hydrographic Office of The British Admiralty. It was used with minor revi- sions until the 1850’s, when the US. Coast and Geo- detic Survey maps were issued (Lincoln, 1969, p. 6—8). In connection with a survey for railroad routes to the west Blake (1857, p. 145—162) in 1853 prepared a geo- logic map on a scale of 1:211,000 of the area from San Francisco to Richmond, including most of the San Francisco North quadrangle. He recognized and gave fair descriptions of the principal rock types of the Franciscan Formation (except greenstone) in chapter 12. His geologic cross section from Point Lobos to Yerba Buena Island can be used today with only minor changes. He interpreted the serpentine in the Fort Point—Hunters Point shear zone as an eastward- ‘ dipping sill in the sandstone. He also described num- erous artesian wells in the surficial deposits on the flanks of the hills in the northeastern part of present- day San Francisco and in the vicinity of Mission Dolores. Unfortunately, he gave no information on rate of flow or on declining water levels, except for the following vague statement (p. 162): “and when the borings first commenced, an overflow was generally obtained.” The report of the 1860—64 Geological Survey of Cali- fornia, led by J. D. Whitney (1865, p. 76-79) , includes observations probably made by Brewer (1930, p. 365— 373), Whitney’s principal assistant, on the geologic features in the San Francisco area. Unfortunately, this survey did not carry out its stated intention to begin “a map of the city and its vicinity, on a large scale, *** on which all the varieties of strata and of the superficial covering of soil and sands will be laid down.” (Whitney, 1865, p. 78). In 1891 Professor A. C. Lawson of the University of California and his students began geologic mapping that led to the publication of the monumental San Francisco Folio (Lawson, 1914) with a geologic map scale of 1 : 62,500. Other papers that resulted from their work include those of Ransome on Angel Island (1894) and Point Bonita (1893) and Palache (1894) on the serpentine of Potrero Hill. An earlier, more detailed report, “Sketch of the geology of the San Francisco Peninsula” (Lawson, 1895), included a geologic map of the peninsula from San Francisco almost to Half Moon Bay on a 1:113,000-scale shaded-relief base made by photographing a relief model. The field note- books and field maps for the San Francisco Folio were not available to me for examination as a part of my study. A report by Crandall (1907), “The Geology of the San Francisco Peninsula,” which is accompanied by a map on a scale of 1:126,000, is of interest because it includes a few geologic details not mentioned by Lawson. The 1906 earthquake was a great stimulus for geo- logic study within the quadrangle because of the great havoc it wrought. Many geologists contributed to the voluminous report of the California State Earthquake Investigations Commission, edited by Lawson (1908). Map 17 of the atlas is a fine geologic map of the City of San Francisco and Yerba Buena Island on a scale of 1:40,000. It differs from the geologic map in the San Francisco Folio in having a more legible base, in show- ing more detail in bedrock areas, and in showing areas of land reclaimed from San Francisco Bay. The “Report on the Underground Water Supply of San Francisco County” (Bartell, 1913), popularly known as the “O’Shaughnessy Report,” is a valuable source of borehole logs and well data. It contains a geologic map of the City of San Francisco by Lawson at a scale of 137,500, the base for which, apparently made by the City Engineer’s office, shows streets and topography by contours. The geology depicted differs slightly from that on other geologic maps of the city. 8 GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE “Hydraulic-Mining Debris in the Sierra Nevada” (Gilbert, 1917 ) contains much information on sedi- mentation in San Francisco Bay. A report on “Subsi- dence and the Foundation Problem in San Francisco” (American Society Civil Engineers, 1932) is a rich source of data on engineering, geology, and the histori- cal development of the city. G. D. Louderback was greatly interested in the geol- ogy of San Francisco Bay as well as in the Quaternary geology of the bay area. Early in his career, Louderback participated in a survey of the bay sediments (Sumner and others, 1914). Later he published a short paper on bay sediments (Louderback, 1939) and a geologic history of the bay (Louderback, 1951). Studies in connection with constructing bridges across the bay have given valuable data on the geology of the east edge of the quadrangle. An early effort was reported by the Hoover-Young San Francisco Bay Bridge Commission ( 1930). Trask and Rolston (1951) presented data based on logs of holes bored to investi- gate the route of a bridge parallel to, but 300 feet north of, the present San Francisco—Oakland Bay Bridge and on holes bored to investigate a site for another bridge between the island of Alameda and the south edge of the quadrangle. ACKNOWLEDGMENTS The author is indebted to a great number of people of private and public engineering organizations and other organizations and to local residents who fur- nished logs of boreholes and other data. Ralph Wads- worth, San Francisco City Engineer, furnished, through the services of Sherman Duckell, Reuben Owens, and William Daly of his staff, many useful maps and re- ports and gave access to valuable data in the files of the Department of Public Works. Thomas Mullaney, Engineer, Department of Public Works, was a frequent guide for M. G. Bonilla and the author in tunnel exca- vation projects and was generous with his time and excavation data. E. I. Muheim, Superintendent of Sewers, furnished geologic cross sections of tunnels; the late C. E. Lee, consulting foundation and hydrologic engineer, furnished many logs of boreholes and other engineering geology data. William Moore, consulting engineer of the firm of Dames and Moore, Soil Mechanics Engineers, gave access to a large number of logs of boreholes made by his firm. J. G. Gratton and J. N. Pitcher, well drillers, furnished logs of boreholes and permitted observation of drilling operations and sampling. The California State Division of Highways and Division of Bay Toll Crossings furnished copies of logs of their boreholes. Several Federal agencies assisted by furnishing boat transportation: US. Public Health Service; Corps of Engineers, US. Army; US. Coast Guard; Federal Bureau of Prisons. Colleagues in the Geological Survey helped in many phases of the work. In particular, E. B. Eckel and Ernest Dobrovolny conceived the project and con- tributed much sound advice and support. The sections on the Franciscan Formation benefited from many dis- cussions with E. H. Bailey, W. P. Irwin, and D. L. Jones. The review of part or all of the manuscript by the following Geological Survey colleagues was benefi- cial and greatly appreciated: C. A. Kaye, W. P. Irwin, Richard J anda, Helen Beikman, W. R. Hansen, J. T. McGill, and Katherine Reed. Most of all I am grateful to M. G. Bonilla for his invaluable contributions. FIELDWORK Fieldwork was started in August 1947 by C. A. Kaye, who mapped approximately 2 square miles of the south- ern Marin Peninsula. In November, Kaye was joined by M. G. Bonilla, who began systematically combing streets and alleys in San Francisco for rock exposures. The project was recessed from May 1948 to November 1948, when the author and Bonilla resumed geologic mapping and started a program of collecting, from pri- vate and public agencies, logs of boreholes and engi- neering and geologic reports of specific construction sites in the San Francisco Bay area. D. H. Radbruch started work on the project in June 1949. W. I. Kon- koff assisted in geologic mapping during part of 1950 and 1951. Fieldwork was largely completed by the close of 1952. The geologic map (pl. 1), however, contains data obtained at a few excavation sites as recently as 1964. Results of this study are published reports on the geology of the quadrangle (Schlocker and others, 1958), on a Cretaceous ammonite found in the Fran- ciscan Formation in San Francisco (Schlocker and others, 1954), on the engineering geology of Islais Creek Basin, located at the southeast corner of the San Francisco North quadrangle (Radbruch and Schlocker, 1958), and on rodingite on Angel Island (Schlocker, 1960). METHODS OF MAPPING Because the land surface of much of the quadrangle is hidden by streets, buildings, and other manmade structures, geologic observations were confined largely to undeveloped lots and to excavations opened for utility lines and building foundations. These observa- tions were supplemented by data from boreholes and earlier foundation construction. Several thousand logs of boreholes drilled by private firms and Federal, State, county, and municipal agencies provided data that were invaluable in preparing the geologic map. FRANCISCAN Aerial photos taken in 1946 were used extensively in relatively open areas such as the Presidio, the Mount Sutro—Twin Peaks area, Marin Peninsula, Angel Island, Belvedere Island, and Tiburon Peninsula. The most commonly used scales were 1:12,000 and 126,000. Excellent aerial photographs, at a scale of 1:2,400 — made available by the San Francisco City Engineer’s office — were used for part of the Twin Peaks—Mount Olympus—Corona Heights area. The topographic quad- rangle map enlarged to 1 : 6,000 scale was used for parts of San Francisco covered with a dense street network. STRATIGRAPHY The geologic formations of the San Francisco North quadrangle fall into two distinct groups, bed- rock and surficial deposits, that differ greatly in age, lithology, and topographic expression (pl. 1). The older group, the bedrock, comprises the sedimentary, igneous, and metamorphic rocks of the Franciscan For- mation of J urassic(?) and Cretaceous age, and the ser- pentine and gabbro associated with them. Areas of exposed Franciscan Formation are generally hilly. The younger group, unconsolidated surficial deposits of Pleistocene and Holocene age, are predominantly dune sand and water-laid sand, mud, and clay, but they include some fairly extensive deposits of slope wash, alluvium, landslide debris, and artificial fill. The sur- ficial deposits above sea level mantle and extensively modify the lower slopes and fill the valleys between the bedrock hills; their thickness varies from a few feet to more than 100 feet. In the bay, borings show that the pre-Tertiary bedrock is overlain by deposits of sand, clay, and mud ranging in thickness from 100 to 300 feet. In some of the channels cut in the bay floor, unconsolidated material is locally absent. BEDROCK FRANCISCAN FORMATION w The Franciscan Formation is a complex assemblage of various rock types, predominantly sedimentary, but also volcanic and metamorphic, exposed in southwest- ern Oregon and along much of western California from the Oregon border at least to Santa Catalina Island (fig. 6). Its exposures in California are generally bounded on the east by the Klamath Mountains and the Coast Range thrust fault (Bailey, 1970), which lies near the west border of the Great Valley, and on the west by the Pacific Ocean. It is exposed over roughly 15,000 square miles of the Coast Ranges, and its total extent on land and offshore may be as much as 75,000 square miles. The Franciscan Formation and its possible correlatives in Baja California, California, Oregon, Washington, Canada, and Alaska are typical 534—039 0 - 74 - 2 FORMATION 9 of rocks deposited in orogenically active borders be- tween oceanic and continental crustal plates. The Franciscan Formation ranges in age from Late Jurassic to Late Cretaceous; however, rocks deposited during the entire age span generally are not present in any one area. The rocks were folded, shattered, and sheared many times after they were depbsited. Be- cause of the lack of key marker beds, the scarcity of fossils, and structural complexities, a standard section has not been established, and the total thickness is not known. Thickness measurements of part of the Fran- ciscan at several places show that it is probably more than 50,000 feet thick. A comprehensive description of the Franciscan For- mation throughout its entire distribution and a discus- sion of its significance to the geology of the Coast Ranges is contained in a report by Bailey, Irwin, and Jones (1964). A detailed discussion of the history of the study of the formation is also contained in this 1964 report and in a report by Taliaferro (1943, p. 112— 122). In the San Francisco North quadrangle, the Fran- ciscan Formation may be as much as 10,000 feet thick, and it consists of about 80 percent graywacke sand- stone, 10 percent shale and siltstone, 6 percent mafic volcanic rocks, 3 percent radiolarian chert, and less than 1 percent conglomerate, limestone, and glauco- phane schist. All these rocks have been intruded by ultramafic rocks, mostly serpentine. Fossils found in the Franciscan Formation in the San Francisco North quadrangle are Cretaceous in age. The Franciscan Formation is exposed in the north- eastern part of San Francisco where it is 3,650 feet thick and in the sea cliffs from the Cliff House to Bakers Beach where it is about 2,350 feet thick. In these places, the formation consists mostly of graywacke sandstone and shale. The formation also occurs as tectonic inclusions in a shear zone that extends from Hunters Point to Fort Point. An unknown thickness of graywacke sandstone, greenstone, and radiolarian chert forms the central highlands area of the city, which includes Sunset Heights, Mount Sutro, Twin Peaks, and the adjacent hills to the east. The Franciscan Formation is estimated to be about 17,000 feet thick in a southwest-dipping section between Richardson Bay and Point Bonita (southeast tip of Marin Peninsula; not in quadrangle). Here the formation includes 8,000 feet of greenstone, more than 6,000 feet of sandstone and shale, and about 3,000 feet of radiolarian chert, which is more abundant here than in most Franciscan terranes (fig. 7). On Angel Island and Tiburon Peninsula, the Fran- ciscan Formation consists of sandstone and shale, with a minimum thickness of 1,850 feet, interbedded green- 10 GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE 42“ _ _C_)_REGON l EXPLANATION Franciscan Formation and related rocks KJf, rocks of predominantly Late Jurassic and Early Cre- taceous age Kf, rocks of predominantly 40° _. late .[z‘arly and early Late Cretaccou 3 age f, rocks ()fundctermincd age \/|// \ ’Dfl / f \ l \— / \ Crystalline basement ‘1 / HAYWARD ‘ / LS“ in /> Fault o \l‘ l \ / 38 _ + FAU LT 5.9/1? ’1 r\—/\\>\ Dashed where concealed . V CALAVERAS L, C>jll ~//\‘/\/\ (”W-erred Location of «3.1 FAULT /: //\, \I/ L \ plate1 -p,\l,‘,$’x':‘/‘/|;§ \\ \ /§L/\7\ \\/r //\\//\I, —/[/\/ \ / /\ I T7,\/>/\ \ x’ A? \\/\\\ \ 91m no 4’ I l\/:/ \l /| “’ (Mr—2w, . / > ’l/ll/|\\\I/’\//\\: o O \ 36 - + w + m /~ 0 / \ //, \x /\/ TJ><6€®IOOP ‘\/'/\/’\/\l—\/\’l\\/\ O \\/\ ,;\7\1/\\\\ //\/I\l\\l\:l<\l:\\/\:ll\ / \ \ \ \ / _ , / (Cs /\ 7\\/l\l/\l/l/—/\\//\ \/:,://7 '7 \ l<\\’ I, 34" — / _ \p + + «L c <3? /‘/\/I/ \ \l/ \/\ 4 I\/ o 100 200 MILES ’\ ,S/(I—flx‘ | l I -,\\)7\ // I l l | I 124° 122° 120° 118” 116° FIGURE 6,—Distribution of Franciscan Formation, related rocks, and principal structural features of western California (modified from Bailey and others, 1964, fig, 29). stone 350 feet thick, and minor radiolarian chert. Most of the Franciscan here has been metamorphosed. Lawson (1895, p. 415) named the Franciscan Series and later (1914, p. 4) employed the name Franciscan Group. He designated San Francisco as the type area, stating that “the Franciscan group was named from San Francisco, where it occurs in extensive exposures ***.” He believed that two radiolarian cherts inter- bedded with the sandstone constituted “well-defined and easily recognizable stratigraphic horizons,” and he used these cherts to divide the Franciscan into five formations. He named these formations, in ascending stratigraphic order, the Cahil Sandstone, the Sausalito Chert, Marin Sandstone, Ingleside Chert, and Bonita Sandstone. These were later reduced to members with- in the Franciscan Formation. All but the Bonita were: recognized by Lawson in this quadrangle.The present writer found numerous sections of lithologically indis- tinguishable radiolarian chert interbedded with sand- stone and for the most part was unable to use them as stratigraphic marker beds as suggested by Lawson. These five names are therefore considered obsolete and are abandoned. Because the Franciscan Formation in this quad- rangle consists of a very complex assemblage of rocks, with no marker beds, the formation has been divided CLASTIC ' SEDIMENTARY ROCKS FIGURE 7 .—Horseshoe Bay and vicinity, southern Marin Penin- sula. Franciscan Formation dips about 50° westward (to left). In the photograph, greenstone terranes appear nearly white; radiolarian chert, sandstone, and shale appear medium to dark gray. The large out along US. Highway 101 exposes sandstone and shale on left (west) and radiolarian chert on right. Parking area at level of freeway was (1949) site of 75- foot-high hill of radiolarian chert and shale. (See figs. 10, 18) . into its major rock types for purposes of discussion: clastic sedimentary rocks, greenstone, radiolarian chert, metamorphic rocks, and sheared rocks. CLASTIC SEDIMENTARY ROCKS Clastic sedimentary rocks of the Franciscan Forma- tion consist of massive sandstone beds, commonly more than 10 feet thick and as much as 35 feet thick, sections of alternating thin-bedded shale and sandstone, and rare thin conglomerate beds. Sandstone of the Franciscan Formation generally fits the description of graywacke (in Williams and others, 1954, p. 297) as “aggregates of sharply angular fragments of every size between sand or fine gravel and impalpable particles.” The grains are mostly angular and range in size from coarse sand to clay. The most common sandstone consists predominantly of fine to medium grains. The sandstone has other attributes characteristic of graywacke, such as poor sorting, 10 percent or more clayey matrix, low porosity, and dark color. On the geologic map (pl. 1), units consisting of mas- sive sandstone and minor amounts of interbedded shale are shown as sandstone. In the northeastern part of San Francisco, two exceptionally thick sequences of shale and thin-bedded sandstone are separated from the massive sandstone. The shale and thin-bedded sandstone sequences vary considerably. In some places they are predominantly shale with a few 1—3-inch-thick beds of sandstone. In other places they are predomi- nantly 2—3-foot-thick beds of sandstone separated by 3—4-inch-thick beds of shale. Where exposures in the city are too poor to determine which combination of lithology and bedding predominates, the elastic rocks are shown as sandstone and shale, undifferentiated. 11 Structural and stratigraphic relations between dis- continuous exposures of elastic rocks are generally obscure, owing to poor exposures, scarcity of fossils, lack of bedding in sandstone, and absence of key marker beds. Some separated exposures evidently are stratigraphically continuous, but only in a few places could sandstones be recognized in distinguishable stratigraphic positions. Soils developed on the elastic rocks vary from slightly sandy clay to clayey sand, but most are sandy silty clay of low permeability. The soils swell greatly and are highly plastic when wet and shrink greatly when dried. Distinct soil horizons are poorly devel- oped. Thickness of the soil (A and B horizons) above weathered rock (C horizon) varies from 3 to 20 feet. Total thickness of soil and moderately weathered rock is between 5 and 30 feet in most places but as much as '70 feet in some places. Most slopes underlain by elastic rocks of the Fran- ciscan Formation are rounded or only moderately steep because of the thick cover of clayey soil. Landslides of surface debris are common on these slopes. Where ero- sion is rapid, such as along shorelines, the sandstone maintains natural cliffs by landsliding. SANDSTON E Bedding within the massive sandstone beds generally is absent or obscure even in good exposures. In some sandstone, bedding is indicated by the orientation of detrital mica, shale flakes, or carbonaceous matter. Some of the sandstone beds, as much as 25 feet thick, appear to be lenses that fill channels in the shale and thin-bedded sandstone sequences. Beds 1/2—4 inches thick show grading from fine- to medium-grained sand- stone at the base to shale at the top (fig. 8). The top few inches of massive sandstone beds also grade up- ward to shale and show channeling, ripple-type cross- bedding, and, in places, distortion of beds that suggests preconsolidation slumping. In some localities stria- tions, grooves, and spatulate depressions are preserved as casts on the underside of bedding planes. Most of the structural attitudes shown on the geo- logic map (pl. 1) were observed on shale and thin- bedded sandstone sequences that separate the massive beds. The color of fresh sandstone, according to the rock color chart (Goddard and others, 1948) is dark gray (N3) to medium gray (N5), with rare tinges of blue (53 5/1). Weathered sandstone ranges from dark greenish gray (5GY 4/1) and olive gray (5Y 6/2) when slightly altered to grayish orange (IOYR 7/6) when greatly altered. Hydrothermally altered sand- stone is white (N9) with streaks of yellowish orange 12 GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE FIGURE 8.—Shale and thin-bedded sandstone of the Franciscan Formation showing graded and distorted bedding and laminae. Broadway Tunnel excavation, Russian Hill, San Francisco. FIGURE 9.—Jointed thick-bedded sandstone and interbedded 3-foot-thick section of shale and thin-bedded sandstone, Franciscan Formation. Northeast side of Telegraph Hill, San Francisco, (10YR 7/7). Chloritized sandstone is light olive (10Y 5/2) to dark greenish gray (5GY 4/1). Closely spaced joints are common in the sandstone beds (fig. 9). Most joints are randomly oriented and have plane or curved surfaces, but many joints are in poorly to moderately well defined parallel sets. Some joint systems consist of one set parallel to the bedding and two or more perpendicular to the bedding. Closely spaced joints are especially prominent in ir- regular shear and mylonite zones, which consist of thoroughly mashed shale and crushed sandstone (fig. 10). Such zones are commonly found in exposures that also contain less fractured sandstone in which the joints are from 4 to 48 inches apart. Joints in unsheared sand- stone are most commonly 2—3 inches apart. Some massive sandstone beds are adjacent to large bodies of dark-gray soft thoroughly sheared shale, in which rounded sandstone fragments, 1 inch to 25 feet in diameter, are embedded (fig. 11). The contact be- tween the sheared shale bodies and the enclosing sand- stone is usually sharp. Tabular shale bodies in the sand- stone (1—6 in. thick) also have intruded fractures at all angles to the bedding; these condition indicate that shale apparently becomes very plastic under shear stresses. The mineralogy of some of these shale segre- gations is given in a later section. Grain size of sandstone in the Franciscan Formation in this area ranges from coarse sand to clay; however, fine to medium sand sizes predominate. The predomi- nant grain-size range in any one bed is fairly narrow and is a parameter that can readily be assigned in the field. Nearly all the sand-size grains are angular. In a few sandstone beds, as much as 3 percent of the grains are subangular to round; in some tuffaceous sandstone, as much as 25 percent of the volcanic rock grains and about 3 percent of the quartz and feldspar grains are subround to round. In one unusual sandstone found in the excavation for the foundation of the Masonic Temple at California and Taylor Streets, about 25 per- cent of the grains were subround to round. The Douvil- leiceras-bearing sandstone west of James D. Phelan Beach contains 1—2 percent well-rounded grains and about 5 percent of grains that are well rounded on one side and angular on the other side. Evidently, these grains were once well rounded and subsequently were broken. I In 50 percent of the sandstone specimens that were examined for grain sphericity, about 90 percent of the grains were equidimensional and 10 percent were elon- gate. In the other specimens, 50—65 percent of the grains were equidimensional and the remaining grains were elongate. The long axes of elongate grains gener- ally lie in the plane of the bedding. Dark-gray to black angular gravel-size fragments of shale, commonly oriented in the plane of the bedding, occur in some sandstone, especially coarse-grained sandstone (fig. 12). The largest fragments, as much as 4 by 41/; inches in size, were found in the Douville- iceras-bearing sandstone. In a few places, large abun- dant paper-thin flakes of coaly material, 1 or 2 square inches in area, parallel the bedding. ' MATRIX The amount of matrix (detritus smaller than 0.02 mm) in sandstone beds of the Franciscan Formation CLASTIC SEDIMENTARY ROCKS 13 FIGURE 10.——Sheared and shattered sandstone and shale of the Franciscan Formation. Continuity of beds, seen near bottom of cut, is uncommon in the Franciscan Formation. Sandstone and shale at right edge of cut are especially strongly sheared. Old deposit of slope debris and ravine fill at top of cut. West side of US. Highway 101, three-quarters of a mile north of Golden Gate Bridge. varies irregularly over short distances but generally constitutes 15—20 percent of the rock. Dikelike segrega- tions of matrix-rich sandstone, generally less than 1 inch thick, are common. The amount of matrix in such FIGURE 11.——Sheared shale of the Franciscan Formation con- taining large hard sandstone masses. Exposure is near top of blanket (melange) of sheared Franciscan rocks, approxi- mately 1,000 feet thick, that covers many squares miles northwest of the San Francisco North quadrangle. Note berms and pipes draining water from nearly horizontal holes in slope used to prevent and control landslides along Cali- fornia highways. West side of US. Highway 101, 1 mile west of Sausalito Point, Marin Peninsula. 14 GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE O 2cm FIGURE 12.—Shale fragments (black) in coarse-grained gray- wacke sandstone of the Franciscan Formation, Section parallel to bedding. Laguna Honda Reservoir area, San Francisco. segregations in sandstone beds on Marin Peninsula is so great that the rock is more properly called sandy mudstone. Microscopic examination shows the matrix to be ag- gregates of silt- and clay-size crystals with equigran- ular, acicular, and micaceous habits. In some places, the matrix consists of shale clasts squeezed around harder grains. Constituents that can be easily identi- fied are quartz, plagioclase, muscovite, biotite, and lig- nitic material. Exceedingly fine-grained micaceous minerals with moderate to high birefringence are very abundant. To investigate the clay minerals in the matrix, the rock was crushed; the <2-micron fraction of the mate- rial passing a 65-mesh (0.210 mm) screen was assumed to be largely matrix and was used to prepare oriented aggregates for X-ray diffractometer analyses. This material from unweathered sandstone collected from widely separated localities consists mostly of mica and chlorite which are fairly well ordered, though some contain a low percentage of expandable layers. Mica is both the dioctahedral (muscovite) and trioctahedral (biotite, phlogopite) type; kaolinite is absent. The matrix of the sandstone from Alcatraz and Yerba Buena Islands, Quarry Point on Angel Island, Telegraph and Russian Hills in San Francisco, and from north of Lime Point contains fibrous aggregates of acicular pumpellyite crystals. Some of the adjoining volcanic, feldspar, and quartz clasts also contain pum- pellyite. The pumpellyite in the matrix and adjoining clasts appears to be metamorphic. Some isolated clasts of pumpellyite in radial fibrous sheaves with or without quartz may be detrital. DETRITAL GRAINS The lithologic composition of the sandstone beds varies from arkose to volcanic graywacke, but most beds are arkosic and lithic graywacke (fig. 13). The plotted positions of sandstone samples shown in figure 13 are estimates based on examination of thin sections and cut surfaces. Results of point-count analyses of thin sections of some of the sandstone samples are given in table 1. With one exception, data are insufficient to suggest a correlation between sandstone composition and strati- graphic position or geographic location. The mineral and chemical composition of sandstone beds west of Corona Heights and east of Twin Peaks and their megascopic appearance suggest that this belt of sand- stone is largely volcanic graywacke. In several other places throughout the quadrangle, greenstone grades into volcanic graywacke. STABLE GRAINS (Quartz, chart, and quartzite) QUARTZ GRAYWACKE FELDSPATHIC GRAYWACKE 8 73 1039. Q, 52. 94.6429 39. LLTHICGEAY‘N‘)‘ D GRAYWACKE .' 34 33 90 35 29 88 CHIEFLY VOLCANIC GRAYWACKE 67:89 .80 V V V V V V UNSTABLE GRAINS 830 V V V V FELDSPAHS ROCK FRAGMENTS FIGURE 13.—-Classification of sandstone samples of the Fran- ciscan Formation. Location of samples is shown in figure 14. TABLE 1.—Modal composition of sandstone samples of the Francrscan Formation Approximate volume percent determined by point- count analyses of thin sections s A t ‘6' d F: m m h 5' g 3:” . ‘53 "E u ‘8‘ é’ .6 a, V o x. “l ’5‘, n 0.23 g f, . 5 g £5 .9 a as I—c N a {-5 1: fi «7 w 4,: 11% ’52 5 5s 3 5 ‘3 s 1: 5’ fi 2 2 E 9 .55 5 S m z a In > E71 0 a (.7 S 91 ................ 1649 32.5 14.9 12.5 14.9 3.6 10.5 7.6 0.2 3.3 45... . 1228 55.7 18.4 17.6 1.6 1.5 1.6 2.0 — 1.5 40... . 1899 24.4 35.0 16.2 10.4 4.1 4.3 4.4 — 1.2 46... . 1934 17.7 13.7 19.2 19.5 16.4 2.7 9.4 .8 .6 39... ..... 1979 50.3 20.2 15.1 3.0 6.8 3.6 — — .5 lGrains smaller than about 0.03 mm. but matrix also includes all sizes of mica crystals and carbonaceous material. CLASTIC SEDIMENTARY ROCKS 15 Mineral grains are most", quartz and feldspar. About one-half to three-fourths of the quartz grains found in the sandstone of the Franciscan Formation show slight strain effects at extinction under crossed nicols. A small number of quartz grains show strong strain effects and brecciation. Most of these grains were classed as quart- zites in the grain composition studies. The slight strain effects shown by most of the quartz grains are much less intense than those shown by quartz in semischis- tose sandstone of Angel Island but are of slightly greater intensity than those shown by quartz crystals in veinlets cutting the sandstone. Most of the feldspar grains are plagioclase. They range in composition from albite (Am) to sodic labra- dorite (Ann). Andesine, An33—An39, is the plagioclase in two-thirds of the sandstone. In some sandstone, andesine is accompanied by oligoclase, Ann—Ange. Even in sandstone containing large amounts of basic volcanic rock fragments, the predominating feldspar detritus is andesine. Less than 1 percent of the plagioclase shows zoning. A remarkable aspect of the sandstone is the scarcity of potassium feldspar grains (fig. 14; table 2). An ex- ception is the sandstone of Point Lobos, which contains 5—10 percent potassium feldspar. Sandstone from a few other localities also contains small but persistent amounts of potassium feldspar; these are unsheared, unmetamorphosed sandstone beds of Quarry Point on Angel Island, the Douvilleiceras-bearing sandstone and adjoining sandstone in the cliffs of South Bay and Bakers Beach, and the sandstone of Alcatraz Island. Heavy mineral determinations of nine sandstone samples were made, using procedures outlined by Hut- ton (1950, p. 639). The approximate proportions of heavy minerals in the sandstone (table 3) are weighted- average frequencies (Hutton, 1950, p. 650) from esti- mates made by counting grains in the 20—74- and 7 4— 149-micron-size ranges identified under the petro- graphic microscope. The weighted-average frequencies also include less accurate estimates for coarser sizes, for which only representative grains were subjected to petrographic and X-ray diffraction analysis. Biotite, chlorite, vermiculite, and, to a lesser extent, nontronite predominate in the heavy fraction of almost all sandstone. X-ray diffraction analysis shows that these minerals are present together in composite grains. Individual flakes of chlorite and muscovite are fairly abundant. The epidote group minerals, epidote, clino- zoisite, and zoisite, the related mineral pumpellyite, and sphene and garnet are present in conspicuous pro- portions. The heavy-mineral analyses show some slight, incon- sistent differences between sandstones containing po- tassium feldspars and those that do not. Because potas- sium feldspar in a sandstone suggests that granitic rocks were components of the source area, minerals characteristic of a granitic source, such as zircon and tourmaline, would also be expected to be more abun- dant in the potassium feldspar-bearing sandstone, but the results do not show this. Sandstone sample 1931 from the Golden Gate Bridge area in Marin County contained rare grains of the chromian spinel, picotite, as did a sample of semichistose sandstone near the west shore of Angel Island 1,000 feet south of Point Ione, suggesting that ultramafic rocks were in the source area. All the sandstones contain sedimentary, volcanic, granitic, and metamorphic rock grains. Shale is con- spicuous in many sandstone samples, and silty, clayey, and carbonaceous varieties can be seen in a single thin section. Chert grains, in which no Radiolaria have been found, are common in all sandstones. Graywacke sandstone grains are sparse; quartz sandstone or ortho- quartzite grains are somewhat more common. Detrital carbonaceous matter, probably representing plant re- mains, is also common. Altered volcanic rocks are found in almost all sand- stone. Many sandstones contain grains of chlorite or vermiculite-chlorite-nontronite aggregates that appear to be altered glassy basalts. Some of these grains con- tain microlites of albite-oligoclase and rarely labrador- ite. Grains of greenstone with spherulitic, hyaloophitic, intersertal, and pilotaxitic textures are common, as are what appear to be grains of acidic felsites. Acidic to intermediate volcanic grains are abundant in coarse- grained sandstone at California and Taylor Streets. Small amounts of lithic grains with fine-grained granitic textures are found in most sandstones. Most of these grains are quartz-plagioclase in composition, but grains of quartz-plagioclase-orthoclase, quartz- perthite, and quartz-microcline rocks are present in small amounts in a few sandstone beds. For example, the Douvilleiceras-bearing sandstone on the cliffs of the South Bay part of Golden Gate channel is distinguished by the presence of such lithic granitic grains as well as by a more-than-average amount of mica quartzite and graphite quartzite. Thin sections of sandstone from the Presidio, Nob Hill, and southern Marin Peninsula con- tain a small percentage of serpentine grains. Grains of fine-textured metamorphic rocks, such as quartzite or what appears to be metachert, are com- mon in almost all sandstones. These consist largely of quartz, generally with suture texture, with one or more of the following minerals: Epidote-group minerals, particularly epidote and clinozoisite, rarely zoisite; amphiboles, generally hornblende pleochroic in brown hues, pumpellyite in radial-fibrous sheaves; micas, com- monly muscovite, rarely brown biotite; and graphite. GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE 16 37° 1 22° 30' 52' 30" 4%? “7v? . O Sausallto «PO : ¢ <9 ”CL R A N c 1 s c 0 LimePoint Alcatraz Island oATE Fort Point PRESIDIO°MILITARY V 57': RESERVATION [Q SOUTH BA y \ g D Q . 91 ,. ______ 93 89 ____________ LINCOLN 929° PARK Point G Lobos Q ‘II D N __________ it. _______________ " GOLDEN D ---------------------------- ‘8 V: s I °~ s 2 a E g a )> E 51 37° 45' o 1 2 MILE s .1 v 35 ‘ 91 I 100 None Trace 1/2 to 3 More than ‘ percent 3 percent Sample localities Bedrock Quaternary outcrops deposits Showing potassium feldspar center” of sandslone. Map num» bers refer to table 2 w v 122° 22'30“ Treasure Island Potrero Point FIGURE 14.—Locations of sandstone samples of the Franciscan Formation and annotations as to potassium feldspar content. CLASTIC SEDIMENTARY ROCKS 17 TABLE 2.—Potassium feldspar in sandstone of the Franctscan Formation [Grain size: VF=very fine, F=fine, M=medium, C=coarse, and VC=very coarse. Samples are fresh except as noted. Massive sandstone samples are from beds more than 3 ft thick, and thin-bedded samples are from beds less than 4 in. thick. Bedding indications not noted where exposures are poor. Tr.=trace] . 0 O _ H z 2° 5 w ‘3: "" 5 Descri t‘ k '2 2 x a m p we remar s as 7; hi 9. 2.2 e w it: E 908‘; a 8 c. o! 9‘ x. o A o m < E a m Angel Island 1 S5 F—M Shear-ed by faulting .............................................................. 2 $64 M—C Semischistose 3 S39 C o 4 528 M Semischistose. slightly to moderately weathered. 5 S27 M—C Semischistose 6 S33 C Semischistose, slightly to moderately weathered. 7 S23 M do 8 B48 C Semischistose 9 S74 F—M 4 ...... S75 F Thin bedded, graded bedding, laminated 10 B40 M Massive ....... 11 B34 M~C Semischistos 12 1879 F—M o ...... 1879 VF Laminated 13 1667 14 B55 VF Semischistose 15 S47 F do 16 B56 M—C do 17 813 C do 18 $52 VF do 19 S71 0 ., , #10. .. ,, , Tiburon Peninsula and Belvedere Island 20 704 V0 Much detrital shale, semischistose.... 21 709 F 22 710 M Moderately weathered . 23 692 C Moderately weathered, 24 735 F—M Semischistose 25 739 C Marin Peninsula N O3 0‘ {D s- ”I" E 27 598 M Moderately weathered .......................................... Tr. 28 602 M 29 2026 M Massive ...... 2027 M do 2028 M do 2029 M do ...... 2031 F Thin-bedded, laminated 30 612 V0 ...... 621 M 32 627 M—C 33 2000 Massive 34 2041 F—M Graded bedding, thin bedded, small-scale channeling 2039 M Massive 2035 M Massive 2036 F—M Thin bedd . 2037 M—C Massive ...... 2038 M do 37 1931 C do ...... 2032 C—M do 1/2 2033 M—C Thin bedded, small—scale channeling... 1/4 ...... 2034 o 38 1980 M-C Massive, moderately weathered... 39 1979 M—C Massive ............................................. Alcatraz Island 40 1899 M Massive 1 Pyritic orthoquartzite, mica schist, amphibolite, and epidote-albite rocks are also present in some sand- stones. Chemical composition—Franciscan sandstone con- tains less SiO2 and more A1203, FeO, MgO, and NaZO than Clarke’s (1924, p. 30) average sandstone (table 4, column 6a). Taliaferro (1943, p. 136) showed the TABLE 2.—Patassium feldspar in sandstone of the Francrscan F ormation—Continued . U o . +7 z :2 g q, ‘1: "" 5 Descri t' k 3 2 >< : m p we remar s Ev *-‘ hi) ‘1 8;! a: a 8:: E pug-5 2 o g d 9‘ h o A o to <1 E be 9. San Francrsco 41 2042 M Massive 42 1250 M do 43 2083 F—M 44 1135 F Thin bedded, small-scale channeling ................................ ...... 1149 M Massive ...... 1363 F Thin bedded, small-scale channeling._...___...__.,,__....,......... 45 1195 M Massive ...... 1228 M do ...... 1230 M do 46 1934 VG ...... 1935 M 4‘7 149 M 48 38 M Laminated 49 834 M Massive 50 2082 F Laminated 51 148 M Moderately weathered .......................................................... 52 358 M 53 2005 F Very coarse grained muscovite flakes .............................. 54 VC Semischistose, slightly to moderately weathered. 55 1272 F—M 56 1263 M 57 1345 Massive 58 117 VC Semischistose ...... 118 C do 59 1276 F—M 60 150 M Massive 61 341 C 62 110 C vMassive .......................................................... > Tr. 63 131 M 64 173 F 65 923 F . Tr(‘l) . 66 1275 M Massive ______ 1275 F Thin bedded 67 112 M Massive .................................................................... 1/2 68 147 M 69 407 VF 70 348 M Moderately weathered ____________ 71 36 M 72 213 M Moderately weathered .. 73 109P VF ...... 1773 C 74 108P F 75 107P F—M .1 a: H O ilk ’11 < ’11 77 185 M Moderately weathered, thin bedded .................................... ...... 2018 F do 78 1556 C 79 1624 F Moderately weathered, laminated ...................................... 80 1442 F Slightly altered ...... 1442 C do 81 1666 F—M Tr(‘.7). 82 1461 MAC Slightly altered 83 1466 M do Tr('1). 84 510 M 85 340 F 86 95 M Moderately altered, laminated, tufiaceous ...................... 87 96 M do Tr. 88 374 M 89 1882 M Carbonaceous partings 90 373 M—C Massive ............................ 2 91 1566 M—C 2 ...... 1566 F—M 2 ...... 1649 M—C Massive, Douvilleiceras sp.—bearing 2 92 405 F 1 93 1906 M Massive 94 769 F4M do 95 2155—2 M Massive, hydrothermally altered ...................... 15 96 2155—1 M Massive, hydrothermally altered; much potassium feldspar veining 5 97 2155—7 M Massive .......................................... 7 98 2155—6 MeC Massive, hydrothermally altered 5 99 2155~5 M—C Massive ................................................. 7 100 2155—8 do 7 chemical composition of Franciscan sandstone is close to that of granodiorite, but microscopic observation shows that in this area sandstones contain substantial amounts of other rock types. The composition of a representative graywacke from the sandstone belt west of Corona Heights and east of Twin Peaks (table 4, 18 GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE TABLE 3.—Frequency of heavy minerals in sandstone of the Franciscan Formation [Modified from Evans and others (1934, p. 39 5.)] Frequency Approximate percent Frequency Approximate Map No. (fig. 14) .................................................................. 40 Sample No. 621 1931 1195 1363 374 1649 1899 1938A 180 Andalusite 1— 1— .............. 1— .......................................... 1— .............. Apatite 1— 1— l 1— 2 1 1 1—‘ 2 Biotite—chlorite—nontronite—vermiculite composite ........ 8 8 5 8 4 7 6— 6 7+ Brookite 1— Cassiterite 1— ............................ Chlorite 3 4 4 6+ 1 .............. 5 Clinopyroxene Clinozoisite 1 4 3 5 2 .............. Calcite 5 Epidote 3 2 1 3 5 5 2 Garnet 3 1 1 5 1 2 2 Hornblende 1— 1— 1-— 1-— 1— .............. Hypersthene—enstatite 1 — 1 Kyanite 1— Magnetite—ilmenite .............................................................. l 2 1 .............. 2 2 2 1 1 Monazite 4 .............. 1 1— 2 1 3 1— 3 Muscovite 3 2 5 5 1 1 4 3 4 Picotite 1— Pumpellyite 1— 1 2 1 5 5 Pyrite 3 .......................................... l— Rutile 1— 1— Saussuritic rock 7+ .............. 6— 7 6+ Sphene 4 3 4 4 5 1-— 3 Spinel (see picotite for chromian spinel) 1— .............. Tourmaline 1— 1— 1—— .............. 1—- 1 1— 1— 1—~ Tremolite-actinolite 3 2 .............. l 1 1— 2 Zircon 1— 1 1 2 2 3 1—- 1— Zoisite 4 l— . 3 4 4 4 .............. Percent of heavy minerals (by weight) ........................ 4.6 , 1.5 1.2 1.5 1.2 1.6 10.3 1.5 3.1 1East shore Yerba Buena Island, 0.8 mile east of San Francisco North quadrangle. column 5) substantiates its identity as a volcanic gray- wacke. Specific gravity.—The specific gravity of fresh sand- stone ranges from 2.62 to 2.69; most samples are 2.68. Fresh sandstone from the Laguna Honda area, with the lowest specific gravity, contains a more-than-usual amount of shale detritus. Specific gravity of semischis- tose sandstone on Angel Island, Belvedere Island, and Tiburon Peninsula ranges from 2.68 to 2.83. Sand- stones containing jadeite generally have a specific grav- ity greater than 2.72. Specific gravity of fresh laminated siltstone and sandstone from the excavation for the Broadway Tunnel in Russian Hill is 2.69—2.70, though the value for fresh laminated siltstone and sandstone 1 mile north of the Golden Gate Bridge on Marin Pen- insula is 2.64. Weathering and alteration.—Weathering reduces the sandstone to a sandy silty clay containing detrital quartz and the clay minerals montmorillonite, vermi- culite, and mixed layered minerals of expansive clay minerals with mica and chlorite. The high vermiculite and montmorillonite content causes the clayey soil to swell and become highly plastic when wet and shrink and crack when dry. Hydrothermal alteration locally produces kaolinite- group clay minerals and iron oxides by replacement of quartz, feldspar, and other silica-bearing minerals. Joints in altered sandstone are commonly coated or filled with a brown waxy clay mineral that varies in thickness from a film to nearly one-fourth inch and composed randomly mixed layered mica, chlorite, ver- miculite, and montmorillonite. Within or near the edge of serpentine bodies, some sandstone blocks retain the textural and structural appearance of sandstone, but their grains have been completely replaced, mostly by chlorite and randomly mixed layered mica-chlorite, and minor talc, calcite, and pyrite. SHALE Shale beds are massive, banded or laminated, silty and clayey layers. Fissility is poor even in the lami- nated variety, and the shale breaks along irregular slickensided shear surfaces that are commonly parallel to the bedding. The color of fresh shale is dark gray (N2), and the colors of altered shale are similar to those of altered sandstone described elsewhere. J oint- ing is generally more pronounced and more closely CLASTIC SEDIMENTARY ROCKS 19 TABLE 4.—Analyses of sandstone, shale, and a phosphate nodule from the Franciscan Formation [Chemicwal analyses (rapid rock methods) by Paul Elmore, Samuel Botts, I. H. Barlow, and Gillison Chloe. Semiquantitative spectrochemical analyses by WWorthing. Looked for but not found: As, Au, Bi, Cd, Cs, Dy, Er, Eu, Gd, Ge, Hf, Hg, Ho, In, Ir, Lu, Nb, Os, Pd, Pr. Pt, Rb, Re, Rh, Ru. Sb, Ta, Tb, Te, Th Tl, Tm, U, W,Z Phosphate Sandstone Shale nodule 1 2 3 4 5 6 Ga 7 7a 8 Chemical analyses (weight percent) 0.8 68.9 67.0 67.1 60.9 67.3 78.7 63.2 58.4 43.4 4.0 12.7 14.1 14.9 16.4 15.5 4.8 16.1 15.5 9.2 .6 1.5 .9 1.0 1.4 .4 1.1 .7 4 .7 2.5 2.8 4.0 2.9 4.4 3.8 .3 4.9 2.5 11.2 1.7 2.5 2.8 1.6 3.1 1.9 1.2 3.1 2.5 2.6 1.5 1.9 1.3 2 3.9 .61 5.5 1.1 3.1 15.3 3.7 2.7 3.4 3.1 4.2 4.2 .45 2.4 1.3 .52 2.2 2.1 1.6 2.3 .58 3.2 1.3 2.5 3.3 .54 2.2 2.8 2.8 2.6 3.7 1.8 1.3 3.7 3.7 4.6 .22 .68 .59 .78 .50 .18 .31 .49 1.3 .52 42 68 .60 51 .65 .60 .25 .68 .65 32 10 13 .12 12 .15 .15 .08 .20 .17 9 8 05 .06 .08 06 .10 .08 Trace .09 Trace 12 ._ 07 .28 .14 17 .10 <.05 5.04 (.05 2.64 34 .. 100 100 99 99 100 100 100 99 99 99 S (aqua. regia soluble) .. 0.06 0. 03 0. 22 0.08 0.02 0.01 0.22 0.18 Specific gravity (powder) 2.70 2. 68 2. 66 2. 69 2. 73 2. 66 2.70 2.83 Semiquantitative spectrochemical analyses (weight percent) 0.000015 0.000015 0.000015 0 0 0.00003 .003 .007 .007 .007 .007 003 .003 .03 .03 .03 .03 .007 15 .07 .00015 .00015 .00015 .00015 0 0 0 0 0 0 0 0 .015 .0015 .003 .003 .0015 .003 .003 0003 .007 .007 .007 .003 .003 .003 .0007 .0015 .0015 .003 .0015 .0015 .0007 .015 0007 .0007 .0007 .0007 .0007 .0007 .0007 0 0 0 0 .007 0 0 0 .015 .003 .007 .007 0.03 .007 .007 .003 .00015 .0003 .0007 0.003 .0001 0 .015 .0007 .0007 .0007 .0007 .0007 .0007 .0007 0 0 0 .003 .015 .007 .007 .015 .007 .015 .015 0 0 0 0 0 0 .015 .003 .007 .007 .003 .007 .003 .0015 .0015 .0015 .0015 .0015 .0015 .0015 .015 .00015 .00015 .00015 .00015 .00015 .00015 .0015 .007 .007 .007 .007 .007 .007 .003 1. Graywacke, northeast base of Telegraph Hill, 200 ft west of Kearny St., 50 ft south of Francisco St., San Francisco (sample No. SF—2148). 2. Lithic graywacke, west of James D. Phelan Beach State Park, fresh rock rsiear sea level below Douvilleicerus-bearing graywacke (sample No. F~373). 3. Graywacke, Marin Peninsula, 5,050 ft northwest of Lime Point light- house, north end of roadcut west side of US. Highway 101 (sample No. SF—2114). 4. Graywacke, south base of Mount Sutro, east end of Laguna Honda (sample No. SF—2141). 5. Volcanic graywacke, 1,000 ft south of top of Buena Vista Park, San Francisco (sample No. SF—2140) . spaced in shales of the Franciscan Formation than in sandstones. The shales are made up mostly of micaceous min- erals in the silt sizes (2—62 microns). However, the shales range from those containing no material greater than silt size to those containing substantial amounts. Most contain more than 15 percent silt-size quartz and feldspar. Carbonaceous material is in sand- to clay- size plates and fibers that are nearly opaque, dark brown in transmitted light, and dark gray to black in reflected light. Pyrite is often found in tiny crystals disseminated in large carbonaceous particles and layers. The mineral composition of shale is similar to that of the sandstone matrix. Predominating constituents are mica, clay minerals, quartz, feldspar, and carbonaceous material. About half the shale consists of quartz and 6. Arkosic graywacke, San Francisco South quadrangle, 4,100 ft east of peak 51%;.) San Bruno Mountain, Macco-P.C.A. quarry (sample No. SF— 6a. Average sandstone. Clarke (1924, p. 30). 7. Shale near east end of Broadway Tunnel, Russian Hill, San Francisco (sample No. SF—1140) . 7a. Average shale, Clarke (1924, p. 30). 8. Phosphatic nodule in shale, southeast side of Laguna. Honda, San Fran- cisco (sample No. SF—lOl) feldspar. The micaceous cleavage surfaces and the car- bonaceous plates and fibers are roughly parallel. The mica crystals are generally not alined in a and b crystal- lographic directions. In a slaty shale obtained from the excavation for the Broadway Tunnel in Russian Hill, San Francisco, the mica plates appear to be thicker than those in nonslaty shale. Microscopic examination shows that the micaceous minerals in the shale contain abundant unidentified inclusions of colorless generally equidimensional crys- tals, about 1&4 micron in diameter, of moderate refractive index and birefringence. The mica may be partly or wholly authigenic. X-ray diffraction analyses of the smaller-than-Z-micron fraction of pulverized and deflocculated fresh unsheared shale show that mica predominates, though chlorite is abundant. Randomly 20 mixed layers of mica, chlorite, vermiculite, or mont- morillonite are present in some shales. Kaolinite is gen- erally absent or minor. Soft, sheared, and plastically deformed shale within shear zones consists predominantly of clay minerals and smaller amounts of pyrite-bearing carbonaceous streaks, nonclay minerals, and sheared rock pieces. Clay minerals are mostly chlorite or randomly mixed layered chlorite-montmorillonite and randomly mixed layered mica-montmorillonite. Some sheared shales are mostly montmorillonite and randomly mixed layered montmorillonite-mica-chlorite. The chemical composition of shale obtained from the excavation of the Broadway Tunnel is given in table 4. As compared with the average shale, most of the differ- ences are similar to the differences between Franciscan graywacke-type sandstone and Clarke’s (1924, p. 30) average sandstone. Weathered shale consists 'of mica, montmorillonite and (or) vermiculite, and randomly mixed layered mica-montmorillonite. Hydrothermally altered shale contains these minerals and abundant kaolinite-group clay minerals. CONGLOMERATE Conglomerate lenses and pebbly zones, 1 or 2 feet thick and interbedded with sandstone, are rare in San Francisco and on Marin Peninsula but are common on Angel Island, Belvedere Island, and Tiburon Penin- sula, where some reach a thickness of about 10 feet. At the north end of Simpton Point on Angel Island, con- glomerate beds in sandstone are well exposed. The thickest is a 12-foot bed composed by volume of about one-third well-rounded large clasts and two-thirds matrix of coarse-grained graywacke. The average diam- eter of the large clasts is approximately 2 1/2 inches and the maximum is 12 inches. Sandstone beds above and below the conglomerate bed contain isolated well- rounded clasts 2—3 inches in diameter. Conglomerate at approximately the same strati- graphic position is also exposed about 2,400 feet east of Stuart Point. Here it is about 10 feet thick and con- sists of about 20 percent well-rounded large clasts and 80 percent coarse-grained graywacke matrix. The aver- age diameter of the large clasts is 11/2 inches, and the maximum is 4 inches. Along the beach on Angel Island, halfway between the north end of the serpentine exposure and the point west of Hospital Cove, a pebbly sandstone which is probably at the same stratigraphic position as the con- glomerates was described by Ransome (1894, p. 197) as follows: “The pebbles are usually disposed as bands in the sandstone, and not as separate sharply defined beds. They are generally rather flattened and of all GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE sizes up to 10 inches (25 centimeters) in diameter, and 4 inches (10 centimeters) thick.” Pebbles in the conglomerate at Simpton Point, Angel Island, consist of graywacke, 30 percent; shaly sandstone and shale, 20 percent; red, brown, and black chert, 15 percent; black orthoquartzite, 15 percent; and felsite, porphyry, and intermediate volcanic rocks, 20 percent. The pebbles in the conglomerate exposed east of Stuart Point are largely black or dark-gray chert and orthoquartzite and a small amount of gray- wacke. Phosphate nodules were found in a shear zone in Franciscan clastic rocks near the east end of Laguna Honda. Nodules are 1—12 inches in diameter. The small nodules are massive. The largest one found consisted of concentric layers, 2-5 mm thick, of brown phos- phatic material around an angular chert fragment (2 by 5 by 5 in.) and angular pieces of gray limestone. Both phosphatic material and limestone contained abundant Radiolaria. X-ray diffraction analysis showed the phosphatic layers to consist mostly of carbonate fluorapatite and smaller amounts of chlorite, calcite, quartz, and mica. The chemical composition of a small massive nodule is given in table 4, column 8. METAMORPHOSED SEDIMENTARY ROCKS Although their clastic, sedimentary nature is unmis- takable, most of the sandstone, shale, and conglom- erate of Angel and Belvedere Islands and Tiburon Peninsula are semischists of low metamorphic grade with a foliation more or less parallel to the bedding (fig. 15). They have been metamorphosed under con- ditions of the zeolite to glaucophane schist facies. The semischists are associated with sedimentary rocks that are not sheared and that otherwise appear to be unmetamorphosed; however, the boundaries between the two types are vague, and to show them separately FIGURE 15.~—Semichistose coarse-grained graywacke sandstone of the Franciscan Formation. Campbell Point, Angel, Island. CLASTIC SEDIMENTARY ROCKS 21 on the geologic map (pl. 1) was impractical. Sedimen- tary rocks in the vinicity of Quarry Point on Angel Island and on Corinthian Island are unfoliated and relatively unmetamorphosed, though weak metamor- phism has formed pumpellyite in part of the matrix and in some of the quartz, feldspar, and volcanic clasts. Metamorphic pumpellyite is also found in the Fran- ciscan sandstone of Alcatraz and Yerba Buena Islands and scattered localities in San Francisco and Marin Peninsula. The semischists show several degrees of metamor- phism. Those most strongly affected that retain sedi- mentary textures are tough and have a bluish or greenish cast. Foliation in the sandstone is shown by light-colored hard detrital grains of quartz and feld- spar, which are commonly elongated and lens and spindle shaped. Foliation is also shown by darker mate- rial, representing easily deformed and chemically unstable detritus, and by carbonaceous and micaceous matrix, which appears as crenulated wisps wrapped around the larger harder grains. The shales are gen- erally slaty. Semischistose conglomerates show charac- teristics of the semischistose sandstones with sheared lens-shaped pebbles. Variations in degree of metamorphism are best shown in thin section. In the least affected sandstones, some large detrital grains of quartz and feldspar may have undulose extinction, internal cracks, or a fine-grained mosaic texture and a sutured lens-shaped outline. All the matrix and unstable detritus such as shale, green- stone, and micaceous minerals show marked effects. The unstable detritus is flattened and squeezed into streamlined clots, thin layers, veinlets, and complex plications around and between quartz and feldspar grains. Most of this material appears to be recrystal- lized largely to stilpnomelane, muscovite, and chlorite. In the least metamorphosed rocks it may be difficult to distinguish recrystallized micaceous material from mechanically distorted detrital micaceous material. Sparse biotite is believed to be distorted detrital mate- rial, but the microscopic evidence is inconclusive. Mag- netite and other iron oxides, as well as ilmenite, leu- coxene, and sphene, are commonly found in the mica- ceous layers. In interlaminated shale and fine-grained sandstone, the laminations are crenulated into tiny folds in which the shale is squeezed into the sandstone at the fold axes. Although much relict quartz and feld- spar is present in the sandstone laminae, the shale is recrystallized largely to stilpnomelane, muscovite, quartz, and albite (‘2) , and the rock has a phyllitic sheen. In some weakly metamorphosed sandstones, some of the matrix and unstable detrital grains are converted to fibrous bundles of exceedingly small amphibole crystals of the riebeckite-glaucophane series, partly replaced by chlorite and containing laths of lawsonite. In places, fibrous amphibole has been recrystallized to large euhedral crystals. Bluish-gray slaty shale consists mostly of fine-grained amphibole needles of the glauco- phane-riebeckite series, small amounts of platy stilp- nomelane, and albite veinlets. In sandstones that have been subjected to more severe metamorphism, most of the detrital quartz grains are recrystallized and elongated into a mosaic of small crystals. The dark micaceous aggregates, rep- resenting the matrix and unstable detrital grains, are replaced by varied mineral assemblages. Lawsonite- jadeite aggregates are common, and lawsonite-quartz- chlorite aggregates are also present. Stilpnomelane commonly replaces biotite and chlorite. Clots of stilp- nomelane-muscovite studded with lawsonite laths are common in sandstones that contain much unmetmor- phosed plagioclase. Leucoxene-magnetite aggregates are commonly associated with stilpnomelane. One semi- schistose sandstone shows jadeite replacing stilpnome- lane in aggregates of stilpnomelane and lawsonite. Some metasiltstones near Knox Point consist of law- sonite, stilpnomelane, albite, and quartz. In other sand- stones that are closer to serpentine borders where mag- nesium metasomatism was affective, tremolite instead of lawsonite is present. A conglomerate exposed at Blunt Point on Angel Island contains bluish sheared pebbles enclosed in a pale-blue-green matrix with a micaceous sheen. Iso- lated bluish pebbles are also found in semischistose sandstone along the main road on the west-facing slope southwest of Hospital Cove. The conglomerate con- tains a large variety of low-grade metamorphic mineral assemblages that vary with the composition and texture of the pebbles and their matrix. Lawsonite, in isolated laths and in clots of randomly oriented laths, is widely distributed. Bundles of fine-grained amphibole of the glaucophane-riebeckite series and muscovite plates are common in the matrix. Chlorite, jadeite, hornblende with glaucophane rims, and intergrowths of muscovite- stilpnomelane, zoisite-lawsonite, lawsonite-stilpnome- lane are found in some of the pebbles, whereas meta- quartzite pebbles have little or none of these minerals. The pebbles and the matrix appear to have been meta- morphosed after the conglomerate was deposited, for both appear to be of the same metamorphic grade and metamorphic mineral crystals traverse the boundary between pebbles and matrix. J adeite occurs in sandstone on Angel Island as porphyroblasts and as tiny stumpy prisms intergrown with lawsonite. The sandstone also contains quartz, muscovite, biotite, chlorite and, locally, glaucophane and veins and clots of aragonite (Coleman and Lee, 1962, p. 581). Detrital plagioclase is generally absent. 22 GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE J adeite-bearing rock is abundant at or near intrusive greenstone. Point-count analyses of thin sections of sandstones between Campbell Point and Simpton Point and west of Quarry Point show 34—56 percent jadeite. Jadeite is also found in some of the semischistose sandstone near Knox Point. Bloxam (1956, p. 489) reported that “Franciscan graywackes are locally jade- itized***” on the shore midway between the point west of Hospital Cove and a greenstone exposure to the south. Chemical analyses for three sandstones, including a jadeite-bearing semischist, a jadeite-free semischist, and a jadeite-free unmetamorphosed sandstone, are given in table 5. The analyses show that the three sand- stones are remarkably similar in composition and indi- TABLE 5.-—Chemical composition of sandstone samples of the Franciscan Formation from Angel Island [Rapifii-rock analyses by Paul Elmore, I. H. Barlow, S. D. Botts, S. Mack, and Gillison Chloe. N. d. =not determined] Sample locality .............................. 1 2 3 SiOZ ...................................... 68.2 69 72.9 13.0 11.7 11.3 1.6 1.0 1.1 2.7 4.2 2.8 2.4 3.8 2.7 2.0 1.3 .60 2.2 2.0 3.8 2.2 2.3 .90 3.2 3.3 2.2 .50 .30 .39 .62 .70 .56 .10 .17 .12 .08 .09 .20 1.0 .05 .15 Total ........................ 100 100 100 S (aqua regia soluble) ...... N.d. 0.02 N. 6. Specific gravity (powder) 2.75 2.72 2.62 1. Semischistose; estimated 20 percent jadeite by volume. Collected at shore- line 950 ft southeast of Campbell Point at the boundary line of state park (sample No. 80— RGC 58). 2. Semischistose, jadeite free; contains detrital quartz and plagioclase. authigenic muscovite, stilpnomelane, chlorite, lawsonite, and pumpel— llyrite.S (ignited at shoreline 750 ft southeast of Campbell Point (sample 3. Masgive, unsheared, unmetamorphosed, jadeite free. Collected 4,000 ft N. 25° W. of Blunt Point lighthouse (sample No. 80—RG0758-1). cate that jadeitization and metamorphism took place under isochemical conditions and that sodium meta- somatism was not involved. J adeite porphyroblasts consist of single anhedral to subhedral equant crystals as much as 1.5 mm in diam- eter or of slightly larger clots of several crystals. The average diameter of most of the porphyroblasts is larger than what is believed to have been the diameter of the largest detrital grains. J adeite porphyroblasts are dis- tributed fairly evenly and randomly in the sandstone. Several stages in the development of jadeite por- phyroblasts may be seen in thin section. In an early stage, the jadeite appears in straight-sided masses (suggesting relict plagioclase) of a fine-grained aggre- gate with considerable amounts of finely divided ' quartz. In a later stage, it is less clouded with quartz inclusions but occurs in single crystals or in clots of a small number of crystals showing prominent cleavage fractures. At a still later stage, it is mostly transparent and free of inclusions, except for large lawsonite laths in some, and appears as large crystals or clots with irregular borders. Detrital plagioclase is sparse or completely lacking in semischistose sandstone rich in jadeite, whereas jadeite porphyroblasts are sparse or absent in semi- schistose sandstone having approximately the usual complement of detrital plagioclase. This relationship strongly indicates that plagioclase is involved in the formation of jadeite porphyroblasts; however, the change is evidently not a direct, simple replacement of the albite component of plagioclase by jadeite and quartz, as suggested by de Roever (1955, p. 289) for jadeitized sandstones from the Celebes. Coleman (1965, p. C27) showed that the Angel Island jadeite contains acmite, diopside, and hedenbergite compo- nents. Plagioclase can furnish only part of the elements needed to form the jadeite of Angel Island; thus other detrital minerals must have been involved in its formation. The chemical composition, physical properties, and paragenesis of the jadeite in the metagraywacke sand- stones of Angel Island were given by Coleman (1965, p. 25—34). Coleman found that the jadeite contains 86 percent jadeite molecule (NaAlSiZOG) and that its unit cell volume is larger than pure jadeite. He suggested that this is evidence that the formation of impure jadeite in the metagraywackes of Angel Island requires less pressure than is required to form pure jadeite. Additional evidence of less pressure is the preservation of clastic textures and other sedimentary features in the jadeitized sandstone (Bloxam, 1956, p. 495). The metamorphic grade under which the semischists formed—though it varied in intensity from place to place—was that of the greenschist facies, comparable with Hutton’s (1940, p. 28, 60—61) subzones Chlorite 1 and 2 of the Chlorite zone of the Otago area of New Zealand. Blake, Irwin, and Coleman (1967, p. C3—CS) described metagraywackes in northern California and southwestern Oregon similar to those of the San Fran- cisco North quadrangle and placed them in their tex- tural zones 1 and 2. Some of the rocks shown on the geologic map (pl. 1) as metamorphic rocks are also low-grade schists and in addition may have been gray- wacke-type sediments. These metamorphic rocks, how- ever, differ from the semischists in having been sub- jected to more intense metamorphism, though nowhere more intense than that of the greenschist or albite- epidote-amphibolite facies. These schists are generally coarser in texture than the semischists, are wholly crystalloblastic, and are without obvious sedimentary CLASTIC SEDIMENTARY ROCKS 23 features, such as detrital grains. They contain well- developed amphibole crystals of glaucophane-riebeck- ite and tremolite-actinolite, in addition to lawsonite, muscovite, albite, quartz, and other minerals. They correspond to metagraywackes of textural zone 3 of Blake, Irwin, and Coleman (1967). These schists are unevenly disposed through the semischists, generally at or near intrusive greenstone or serpentine, but many of them are too small to be shown on the geologic map (pl. 1). Such an erratic distribution of metamorphic inten- sities is to be expected in an area where rocks that are heated to temperatures in the lower part of the meta- morphic range by burial during orogeny are subject to local shearing or to reaction with chemically active waters (Turner, in Williams and others, 1954, p. 209). Evidently, the presence of the greenstone body influenced the formation of jadeite for the sandstone richest in jadeite lies near or at the border of the green- stone. Semischistose sandstones elsewhere generally lack jadeite. Conceivably, the border zone sustained high pressure during orogeny and also provided access for fluids needed to jadeitize the sandstone. The sodium content (table 5) of jadeite-bearing and jadeite-free sandstones from Angel Island indicates that sodium metasomatism was not significant in the formation of jadeite. Bloxam (1956, p. 493—494) found similar evi- dence at Valley Ford, Calif. Inasmuch as the semischists have a greater bulk density than unmetamorphosed rocks, pressure prob- ably also was a factor in jadeitization and metamor- phism. High pressure is suggested also by veins and clots of aragonite in the jadeitized sandstone, for cal- cite is converted to aragonite under pressures greater than 4,000 bars (Coleman and Lee, 1962). Nonschis- tose sandstone contains calcite rather than aragonite. Blake, Irwin, and Coleman (1969) suggested that the lawsonite- and jadeite-bearing metagraywackes formed at high-pressure low-temperature conditions below a regional low-angle thrust fault. The presence on Angel Island and Tiburon Peninsula of metagray- wackes similar to those of textural zones 2 and 3 of Blake, Irwin, and Coleman (1967) suggests that the metagraywackes were not far below the postulated thrust fault. Hard dark-gray siliceous beds are abundant in sand- stone near some greenstone bodies. The siliceous beds are silicified shale beds and quartz veins following bedding planes. They are 343—114 inches thick and are separated by three-fourth of an inch of sandstone in some places and 3—5 feet of sandstone in other places. Where they are abundant and closely spaced, they are cut by abundant white quartz veinlets as much as one-half inch thick. Weathering of metamorphosed sedimentary rocks produces sandy silty soils as much as 20 feet thick and weathered rock as deep as 70 feet. The chief clay min- eral of the soil and weathered rock is vermiculite, which causes the soil to become plastic and lose considerable shearing strength when wet. OCCURRENCE NORTHEASTERN SAN FRANCISCO Clastic rocks of the Franciscan Formation are the predominating bedrock in northeastern San Francisco. Russian, Nob, and Telegraph Hills consist of two sequences of massive sandstone and of two thick sequences of shale and thin-bedded sandstone. The observed thickness of the four units is 3,650 feet. The basal 400 feet of the youngest sequence, a massive sandstone, is exposed on the north side of Russian Hill. It overlies a shale and thin-bedded sandstone sequence, 1,350 feet thick, exposed on the south side of Russian Hill and on the West side of Telegraph Hill. The shale and sandstone sequence in turn overlies a lower mas- sive sandstone 700 feet thick, exposed on the north side of Nob Hill and over most of Telegraph Hill (fig. 16). The oldest sequence is shale and thin-bedded sand- stone of which the upper 1,200 feet is exposed. It con- stitutes most of Nob Hill. Bedrock on hills such as Rincon Hill and the hills west of Van Ness Avenue is so poorly exposed, primar- ily because of intense urban development, that recog- nition of the predominant lithology was impossible. Consequently, the bedrock in those areas has been mapped as undifferentiated sandstone and shale of the Franciscan Formation. FIGURE 16.—Massive sandstone of the Franciscan Formation. Old quarry face on the northeast side of Telegraph Hill, San Francisco. 24 GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE CENTRAL HIGHLANDS OF SAN FRANCISCO The sandstones exposed in the central highlands of San Francisco also appear to be in two separate strati- graphic positions. The older sandstone is found at the south base of Mount Sutro and south of Laguna Honda. Radiolarian chert, which forms Mount Sutro, separates it from younger sandstone exposed in the hills south of Mission Dolores. The younger sandstone sequence may be the same sandstone as that exposed to the northwest between Buena Vista Park and Mount Olympus. In other parts of this area, thin sandstone lenses are found interbedded with radiolarian chert and greenstone. The sandstone at Laguna Honda evidently lies nearly horizontal, judging from the attitude of its con- tact with overlying radiolarian chert; if so, no more than 200 feet of sandstone is exposed here. The sand- stone in the hills south of Mission Dolores dips north- eastward and is at least 1,000 feet thick. In the area between Buena Vista Park and Mount Olympus, the sandstone is only about 400 feet thick. CLIFF HOUSE T0 BAKERS BEACH Clastic rocks of the Franciscan Formation form the 100- to 200-foot-high cliffs along the shore from Cliff House to Bakers Beach and also occur at the north border of Bakers Beach (fig. 17). In the cliffs from Cliff House to the shear zone west of Lands End, massive sandstone is interbedded with 100-foot-thick sequences of shale and thin-bedded sandstone. The estimated total thickness is 750 feet. Because of its high potassium feldspar content, the sandstone here is believed to be correlative with sand- stone exposed at San Bruno Mountain, 6 miles to the southeast. Mapping by Bonilla (1961) in the San Francisco South quadrangle disclosed the northwest- trending City College fault, which separates the potassium feldspar-bearing sandstone at San Bruno Mountain from the potassium feldspar-free sandstone northeast of San Bruno Mountain. The shear zone at Lands End is thought to be an extension of the City College fault and similarly is the east boundary of potassium feldspar-bearing sandstone. East of the FIGURE 17.—Thick-bedded sandstone, laminated sandstone, siltstone, and shale of the Franciscan Formation. Note a small offset in closeup view (right). Near the north end of Bakers Beach, San Francisco. CLASTIC SEDIMENTARY ROCKS I 25 shear zone, the sandstone exposed from the east side of the landslide east of Lands End to Bakers Beach is approximately 2,400 feet thick, if the section is a con- tinuous homocline, and dips 30°. This thickness excludes the serpentine at Phelan Beach. In a few places near Bakers Beach, sedimentary features, such as small-scale channels and graded bedding, suggest that part or all of this sequence is overturned. FORT POINT TO POTRERO HILL Sedimentary rocks —— mostly sandstone —— and vol- canic and metamorphic rocks occur as tectonic inclu- sions in a northwest-trending zone of sheared rocks between Fort Point and Potrero Hill. The largest ex- posed sandstone body is the northeast spur of Potrero Hill. Sandstone predominates in the sheared-rock zone between the US. Mint and the Presidio, although most of this area is shown by Lawson (1914) as a continuous body of serpentine. MARIN PENINSULA On Marin Peninsula sandstone occurs in at least three separate sections. The oldest sequence, Lawson’s (1914) Cahil Sandstone, now abandoned, is about 1,700 feet thick and is exposed along the northeast shore in a narrow belt that widens on the ridges to the northwest of Sausalito. It is overlain by greenstone and radiolarian chert beds 1,000—3,000 feet thick, which in turn are overlain by younger sandstone, about 1,000 feet thick, exposed from near Lime Point nearly con- tinuously for more than 2 miles along its strike. The roadcut on the west side of US. Highway 101, about three-fourths of a mile north of the Golden Gate Bridge, contains an excellent exposure of the lower part of this sandstone (fig. 18). The youngest sand- stone sequence crops out about one-third mile west of the second sandstone. It is largely concealed by slope debris and ravine fill, but it is nearly continuous for about one-third mile from the large exposure shown on FIGURE 18.-—Thick-bedded sandstone interbedded with shale and thin-bedded sandstone of the Franciscan Formation, West side of US. Highway 101, three-fourths of a mile north of Golden Gate Bridge, Marin Peninsula. 534-039 0 - 74 - 3 26 GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE the geologic map (pl. 1) northwest toward the build- ings in Fort Baker. Thin belts of sandstone are also exposed to the west and southwest. ANGEL AND ALCATRAZ ISLANDS AND TIBURON PENINSULA Sandstone is exposed over about two-thirds of Angel Island and makes up the crest of Mount Caroline Livermore (fig. 5). Almost all the elastic rocks here, however, are semischistose. (See page 20.) The south- westward-dipping section from Campbell Point to Hos- pital Cove is approximately 1,300 feet thick. The northward-dipping section from the south shore, west of Blunt Point, to the top of Mount Caroline Liver- more is estimated to be at least 1,850 feet thick. An intervening greenstone body, 350 feet thick, was excluded from this total. Ransome (1894, p. 198) believed that 2,000 feet is a conservative estimate for the exposed thickness of the sandstone on Angel Island. In the 1860’s, sandstone of the Franciscan Forma- tion was quarried on Angel Island and used extensively as a building material. The rock is described as being “used chiefly for purposes where beauty and strength are not absolutely required, for this rock possesses neither qualification in a high degree, although it answers very well for ordinary uses in the mild and equable climate of San Francisco.” (Whitney, 1865, p. 77). Conglomerate is interbedded with sandstone at the north end of Simpton Point on Angel Island. It also is exposed about 2,400 feet east of Stuart Point. A pebbly sandstone crops out along the west shore, north of the serpentine body. All these occurrences probably have the same stratigraphic position. (See p. 20.) Alcatraz Island, except for a few patches of surficial material, is entirely sandstone and a few minor shale beds. The exposed thickness of rocks is approximately 500 feet. On Tiburon Peninsula exposed sequences of sand- stone and shale are 5—30 feet thick. Much of the south- eastern part of Belvedere Island, as well as a narrow belt along the southwest shore, is sandstone that con- tains small amounts of shale. The exposed thickness of elastic rocks on Belvedere Island is estimated to be 1,000 feet. Most sandstone and shale on Tiburon Penin- sula and Belvedere and Corinthian Islands are semi- schistose. ORIGIN ENVIRONMENT OF DEPOSITION A marine environment of deposition for the Fran- ciscan Formation is indicated by fossils and the inter- bedded greenstones. Though sparse, marine Mollusca are found in clastic rocks, and Radiolaria, today exclu- sively marine, are found in chert and in some limestone. Radiolarian chert is also found interbedded with green- stone flows which generally show pillow structure, a characteristic of subaqueous eruption. Formations equivalent in age to the Franciscan, such as the Knoxville Formation of Late Jurassic age and the Lower and Upper Cretaceous beds of the Great Valley sequence, represent slope and shelf sedimenta- tion in the less orogenically active parts of the conti- nental crustal plate that existed in Franciscan time (Irwin, 1957, p. 2292). The juxtaposition of the eugeo- synclinal facies and the shelf and slope facies in the modern Coast Ranges is a result of movements of the oceanic and continental crustal plates. Turbidity currents are thought to be capable of car- rying sand to deep-sea basins (Kuenen, 1950, p. 360, 367; Shepard, 1951, p. 53—65). Kopstein (1954, p. 63) suggested also that the high velocities measured on submarine turbidity currents render them capable of carrying “large pieces of gravel.” The materials for the sparse conglomerate beds and for the massive non- graded or poorly graded graywacke sandstone beds and the interbedded shale and thin-bedded sandstone sequences were probably transported by turbidity cur- rents. A mass of unconsolidated detritus of various sizes from clay to sand or possibly gravel, lying on a nearshore submarine slope, could be set in motion by an earthquake or by storm waves (Heezen and Ewing, 1955, p. 2505—2514). The ensuing slump or mudflow could become a turbidity current of sufficient energy to transport most of the material en masse (Kuenen, 1951, p. 31). Thin graded beds of sand and mud would be deposited by weaker turbidity currents. Many of the lithologic features of the Franciscan clastic sedimen- tary rocks, such as graded bedding, small-scale chan- neling, ripple marks, and cross-bedding, can also be produced by turbidity currents. Thus preponderant evidence suggests a deepwater turbidity current origin for the Franciscan Formation. As compared with Kopstein’s (1954, p. 92—93) tabu- lation of evidence for deepwater accumulation of gray- wacke in Harlech Dome, Wales, further evidence for deepwater accumulation in the Franciscan includes the following: (1) Shale layers between graded beds; (2) extremely rare fossils; (3) abundant small-scale current bedding; (4) current-ripple marks; and (5) graded bedding, load casts, flow markings, slump struc- tures, and associated features in some sections. Fur— thermore, Sanders and Swinchatt (1957, p. 1791) believed that the deepwater hypothesis is correct for the origin of the radiolarian cherts of the Franciscan Formation. Local large masses of greenstone in the Franciscan Formation have been compared with seamounts (Bailey and others, 1964, p. 43), and limestone asso- CLASTIC SEDIMENTARY ROCKS 27 ciated with greenstone may represent corals and other shallow-water calcareous organisms that grew on the tops of seamounts (E. H. Bailey, oral commun., 1966). Gradation of greenstone agglomerate and tuff into tufiaceous sandstone and nontuffaceous sandstone could have been brought about by reworking and mix- ing of detritus from land and from pyroclastic rocks erupted during Franciscan time. SOURCE AREA OF DETRITUS Clasts in the conglomerates and sandstones provide a few clues to the kinds of rocks present in their source area. The coarse clasts consist of (1) sandstone, shale, and volcanic rocks similar to those of the Franciscan Formation and (2) cherts, quartzites, volcanic, or dike rocks unlike those of the Franciscan Formation. The sandstones of the quadrangle also contain clasts of granitic rocks, metamorphic rocks, and serpentine as well as clasts of the same rock types as the conglom- erates. Heavy mineral grains in the sandstone are pre- dominantly mica, chlorite, vermiculite, epidote, clino- zoisite, pumpellyite, sphene, and garnet, which are common in low-grade metamorphic rocks and altered volcanic rocks. The heavy mineral fraction of the sand- stones contains only small amounts of zircon, rutile, and tourmaline, which are minerals characteristic of granitic and related metamorphic rocks. Chromian spinel found in sandstone near the Golden Gate Bridge and on Angel Island was probably derived from ultra- mafic rocks or from an older sedimentary rock contain- ing ultramafic-rock detritus. Abundant clasts of Franciscan-type sandstone, shale, and volcanic rocks in the sandstones and conglom- erates suggest “self digestion” of topographically high areas above or below sea level created by local orogeny during the time of accumulation of the Franciscan For- mation. The clasts may also have been derived, how- ever, from pre-Franciscan sedimentary and volcanic rocks. In addition, sandstone and conglomerate clasts suggest a source area containing metamorphic and non- Franciscan volcanic rocks. Rocks similar to the clasts are exposed in the Coast Ranges, Klamath Mountains, and Sierra Nevada (fig. 6). Bailey, Irwin, and Jones (1964, p. 39—41) are in agreement with these conclu- sions. Taliaferro (1943, p. 136—139, 141) suggested that granodiorite was a prominent rock in the landmass from which the Franciscan Formation was derived. Granodiorite or other granitic rock types, however, were not seen in the Angel Island conglomerates and are rare in other conglomerates in the quadrangle. If detrital grains had been derived from a granodiorite ter- rane, their composition would be similar to the con- stituents of the granodiorite and would be present in similar proportions. The freshness of the plagioclase grains and their angularity indicate that mechanical weathering predominated over chemical weathering. The composition of detrital grains is more similar to metamorphic and volcanic rocks. The average granodiorite, according to J ohannsen (1932, p. 321), contains 22 percent potassium feldspar. The granodiorite or quartz diorite of Montara Moun- tain (14 miles south of the mapped area), which would be typical of a possible local source rock, contains as much as 10 percent potassium feldspar. The complete absence of detrital potassium feldspar in most of the Franciscan sandstones in the map area precludes granodiorite as a prominent rock type in the source area. And as potassium feldspar is generally more stable than plagioclase feldspar, which is present in the sandstone, the absence of potassium feldspar cannot be attributed to differential alteration of the sand grains. The mapped area is too small and the data are too sparse to determine the direction of the source area from textural coarsening and thickening of conglom- erate beds. A few linear structures suggest a westerly current flow, such as casts of striations and grooves on the bottom of sandstone beds in the section nearest the Golden Gate Bridge in Marin County. AGE FossrLs Fossils have been found at only five localities in elastic rocks of the Franciscan Formation in this quad- rangle. The only ones useful for age determination are Cretaceous in age. At an unnamed point along the shore of the South Bay of the Golden Gate, 800 feet west of James D. Phelan Beach State Park (pl. 1), the ammonite Dou- villeiceras sp. was found in beds called Marin Sand- stone by Lawson. This ammonite is thought to be of late Early Cretaceous (Albian) age (Schlocker and others, 1954, p. 2373). At this same locality, the writer found a gastropod resembling the holotype of Palad- mete perforata and an undetermined pelecypod,.but neither was useful for age designation. On the beach north of the Needles on Marin Penin- sula near the Golden Gate Bridge, Mantelliceras sp. was found in float probably derived from nearby sand- stone beds. This ammonite is thought to be of Early or early Late Cretaceous age “a little later than that given for the specimen of Douvilleiceras** *” (Hertlein, 1956, p. 1987.). Pelecypod casts from an unrecorded locality on Alca- traz Island were described by Gabb (1869, p. 193), by Stewart (1930, p. 106), and by Anderson (1938, p. 28 GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE 121). According to Stewart, I noceramus ellioti Gabb is probably Cretaceous in age. Anderson assigned an Early Cretaceous(?) age to Lucina alcatrazis. Bailey, Irwin, and Jones (1964, p. 115—123) dis- cussed fossils in the Franciscan Formation in localities outside the quadrangle. Gray limestone lenses found in sheared sandstone and shale north of Sausalito contain Radiolaria pre- served in calcite and pyrite. Similar Radiolaria are abundant in phosphate nodules found in sheared sand- stone and shale near the southeast end of Laguna Honda Reservoir. The Radiolaria are similar in appear- ance to some of those in the chert and shale described by Hinde (1894, p. 235—240) and by Riedel and Schlocker (1956, p. 357—360). The Radiolaria indicate a Jurassic or Cretaceous age. In a discussion of the phosphate nodules just men- tioned, Dickert (1966, p. 292) attributes to W. R. Evitt the identification of M arthasterites tribrachiatus [recte Discoaster tribrachiatus Bramlette and Sullivan], a calcareous nannofossil known from the California Eo- cene. Evitt reexamined the thin section studied by Dickert and reported (W. R. Evitt, oral commun., 1966) that the objects referred to are not M arthaster- ites or triradiate discoasters but are probably radio- larian spines with conspicuously triradiate cross sections. Similar fossils are common in Franciscan cherts north of the Golden Gate in association with abundant Radiolaria. According to Evitt, no age sig- nificance should be attached to these triradiate objects; certainly they should not be the basis for suggesting that Tertiary fossils have been identified from Fran- ciscan sedimentary rocks. Lawson (1914) thought that the oldest formation of his Franciscan Group was the Cahil Sandstone; he thought this sandstone was separated from the over- lying Marin Sandstone by the Sausalito Chert. Law- son’s Cahil Sandstone contains the foraminiferal Calera Limestone Member, whose type locality is along the Pacific Ocean shoreline 9% miles south of the map area. Foraminifera from the Calera Limestone are con- sidered to be early Late Cretaceous in age (Kupper, 1956, p. 41). Inasmuch as Lawson referred to the Dou- Uilleiceras-bearing sandstone of late Early Cretaceous age as his Marin Sandstone, paleontologic data suggest that his Cahil Sandstone is younger than his Marin Sandstone. SIGNIFICANCE OF POTASSIUM FELDSPAR CONTENT Bailey and Irwin (1959) and Bailey, Irwin, and Jones (1964, p. 138—141) determined the potassium feldspar content of the thick shelf and slope facies (miogeosynclinal) in the Coast Ranges (Great Valley sequence) along the west border of the Sacramento Valley. This facies represents almost continual depo- sition from Late Jurassic to Late Cretaceous time. The same workers determined the potassium feldspar con- tent of Mesozoic miogeosynclinal sandstone in other parts of the Coast Ranges. They found that the potas- sium feldspar content increased systematically with decreasing age of the rocks. The median potassium feldspar content of Upper Jurassic rocks was found to be less than 0.5 percent, Lower Cretaceous rocks 1.1 percent, and Upper Cretaceous rocks 13 percent. Bailey, Irwin, and Jones (1964) suggested that all the sediments were derived from the same source—the ancestral Sierra Nevada and the Klamath Mountains of northwestern California, which were created during the Nevadan orogeny. As the granitic rocks emplaced during the orogeny became increasingly exposed and eroded, the potassium feldspar content of the sedi- ments derived from them increased. Thus the potas- sium feldspar content should roughly indicate the age of the sediments. Contradictions unfortunately arise, however, when this dating technique is applied to the Franciscan For- mation in the mapped area, with the assumption that the rocks have all been derived from the same source area. Thus, the complete absence of potassium feldspar in most sandstones (fig. 14, table 2) would indicate a Jurassic age; yet, the fossils are of late Early or early Late Cretaceous age. Bailey, Irwin, and Jones (1964, p. 141) suggested, therefore, that the assumption of a similar source for all the rocks is invalid and that the absence of potassium feldspar is not necessarily diag- nostic of age. A notable exception to the general absence of potas- sium feldspar in Franciscan sandstone is the sandstone in the Point Lobos area, which averages 5—10 percent. On the basis of its high potassium feldspar content and its location southwest of a shear zone thought to be part of the City College fault (see section “Cliff House to Bakers Beach”), this sandstone may be part of the Great Valley sequence sandstone of San Bruno Moun- tain (Bailey and others, 1964, pl. 1). The high (about 20 percent) potassium feldspar con- tent of the sandstone of San Bruno Mountain together with such features as well-developed bedding, abundant flow casts and load casts, and general absence of shear- ing and interbedding greenstone and chert suggest that this sandstone is part of the miogeosynclinal (Great Valley sequence) facies. Lawson (1914, p. 17) also noted that this sandstone “differs from the usual sand- stone of the Franciscan.” , Thus, until more is known of the structure of the Franciscan Formation and better indications of its age are found, the age of the Franciscan in the quadrangle GREENSTONE must be considered to range from probable Jurassic to Early and Late Cretaceous. The probable Jurassic age is suggested by the Jurassic or Cretaceous age of the Radiolaria and by the presence of a great thickness of Franciscan Formation lying structurally below the sandstones containing the Cretaceous fossils. GREENSTONE Greenstone is a term often used for dark igneous rocks of indeterminate composition and origin having green alteration products. The greenstone of the San Francisco area is mostly fine- and medium-grained basalt that has been subjected to little or no alteration and basalt that has been subjected to a variety of local and regional conditions of alteration and metamorph- ism. The most common types are unmetamorphosed moderately fresh basalt and rocks of basaltic chemical composition that consist mostly of pumpellyite, primary pyroxene, albite, and chlorite. Some of these are rich in albite and could be called spilite. The metamor- phosed greenstone of Angel Island, Belvedere Island, and Tiburon Peninsula also contains epidote, lawsonite, glaucophane, hydrogarnet, vesuvianite, and pyroxene. The field identity of many of these rocks is obscured by their dense, aphanitic texture, by close fracturing, and by a variety of chemical and mineralogical alterations. The bulk of the greenstones are probably flows and commonly show pillow structure. Pyroclastic rocks are present in smaller volume. Except for thin intrusive rocks in radiolarian chert, intrusive greenstone could not be distinguished from flows. A long history of volcanism during Franciscan time is indicated by the large masses of greenstone that occur throughout thick sections of clastic rocks and radiolarian chert. Like other rocks of the Franciscan Formation, green- stone forms the hilly parts of the quadrangle. The wide- spread alteration of greenstone to rock containing substantial proportions of clay minerals and the almost universal randomly oriented close fracturing of green- stone are important influences on its natural slopes. With a few notable exceptions, greenstone slopes are smooth and subdued. Small landslides of weathered greenstone debris are common. On Marin Peninsula several large valleys, such as the one draining into Horseshoe Bay and the one draining into the Golden Gate channel west of Lime Point, are bordered by greenstone and presumably have been cut into it (fig. 7). Fresh greenstone is very resistant to erosion in such places and forms the steep sea clifis and stacks at Stuart, Knox, and Blunt Points on Angel Island, Point Diablo along the Golden Gate channel, Marin Peninsula, and Lands End (figs. 19, 20). 29 FIGURE 19.—High cliff of greenstone in the Franciscan Forma- tion. Lime Point, Marin Peninsula, viewed north. The cliff rises 400 feet above the water in the Golden Gate. In the middle distance, greenstone underlies white to light-gray slopes; radiolarian chert or graywacke sandstone and shale underlie dark slopes. West (left) side of Golden Gate Bridge tower rests on greenstone; east side rests on radiolarian chert and shale. FIGURE 20.—Greenstone and radiolarian chert of the Francis- can Formation. Point Diablo, north shore of Golden Gate, Marin Peninsula, on the west edge of the quadrangle. The top of the cliff is 625 feet above sea level. Conspicuously jointed rock at the extremity of the point is greenstone (KJg). It is separated by an east-west shear zone from radio- larian chert (KJc) that makes up the ridge and the light- colored base of the cliff. Greenstone also makes up the central part of the cliff. MEGASCOPIC FEATURES Fresh greenstones is dark gray or greenish gray. Moderately altered greenstone is grayish green and grayish olive. The greenstone most commonly exposed is weathered or hydrothermally altered rock of mod- erate brown, dark yellowish-brown, or moderate red- dish-brown colors. Hydrothermally altered greenstone 3O GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE is also grayish orange and light yellow green at and near radiolarian chert contacts. Iron and manganese oxidation and hydration prod- ucts are commonly found as thin films of bluish black, brownish black, yellow, orange, or brown on joints in altered greenstone. Brown and yellow-green waxy scaly deposits of nontronite, 1/2—4 mm thick, are also commonly found in joints. Moderately fresh green- stone may have the deceptive appearance of being highly altered, owing to the presence of thin earthy coatings of nontronite and iron oxides on joints. Most of the freshest pillow basalt near Horseshoe Bay has a unique greenish-gray vitreous or pitchy luster that is caused by a thin film of slightly sheared chlorite on slickensides, joints, or small faults. Fresh fractures in this rock usually reveal a medium- to fine-grained holo- crystalline rock. On Angel and Belvedere Islands and on Tiburon Peninsula, bright blue joint fillings of crocidolite, as much as one-eighth inch thick, are conspicuous in exposures of reddish-brown altered metamorphosed greenstone. Closely spaced fractures are the most conspicuous structural feature of greenstone (fig. 21). Most of the greenstone is so thoroughly fractured that it shatters easily into pieces 14—1/1 inch across. Only rare expo- sures yield coherent pieces more than 2 inches across. Examples of the latter rock are the aphanitic and porphyritic greenstone on Anza Street on the north- west side of Lone Mountain in San Francisco and the fresh aphanitic pillow greenstone exposed in the west roadcut on US. Highway 101, 400 feet northwest of the Waldo Tunnel in Marin County. Fracturing may have resulted from violent steam explosions during FIGURE 21.—Close random fracturing in greenstone of the Franciscan Formation. South side of road to Sausalito, 0.1 mile east of US. Highway 101, 0.8 mile north of Lime Point, Marin Peninsula. The wood stake lying on the cut is 1 foot long. submarine eruptions, rapid cooling-shrinkage of the hot lava and pyroclastic deposits, or severe and repeated stresses during several orogenic episodes. Tectonic breccias occur in 5—40-foot-wide zones bordering some faults. In these zones the rock has been fractured, sheared, and crushed and now consists of rock pieces, 14—1 inch in diameter, in a matrix of crushed rock of sand and smaller sizes. It is difficult to distinguish such breccias from agglomerates and vol- canic breccias. Because of small-scale random fractur- ing, massive and pillow greenstone at many exposures shows rough surfaces that consist of irregularly shaped 1434/2 -inch projections, which give the rock the appear- ance of a tectonic or volcanic breccia. In some exposures of massive greenstones, in con- trast with randomly oriented fracturing, fractures at intervals of 1/2—1 inch are open, parallel, and straight sided. Subparallel, irregular, and discontinuous vesicles also occur in such exposures. Both the fractures and vesicles have drusy linings of calcite, doubly terminated quartz, zeolites, and hematite. These fractures and vesicles apparently formed during and shortly after solidification of the lava. Pillow structures occur in basalts, in altered rocks of basaltic composition, and in spilites. Pillow structures are rounded bodies 6 inches to 10 feet across in great- est dimension. About half the observed pillows are 8—10 inches thick, 12—18 inches wide, and 18—24 inches long, although some exposures consist predominantly of pillows several times as large. Exposures consisting predominantly of large pillows also contain some smaller more spherical pillows, 6—10 inches in diameter. Small pillows tend to be spheroidal rather than ellipsoi- dal, although spheroidal pillows 4 feet in diameter have been seen. The surfaces of adjacent pillows are closely conform- able in shape to one another, like mold and cast, but are separated by gaps of clayey material 1/3—12 inches thick. Pillows are generally shaped like a stretched-out bun and usually have a convex top and a flat or concave bottom (fig. 22). The bottom of some pillows projects downward between adjacent underlying pillows. The middle of the upper surface of some pillows is depres- sed; such pillows commonly are overlain by pillows with convex bottoms that conform in shape to the pil- lows below. Commonly, a pillow may consist of a thick central part and a thinner appendage. In the Fort Baker area of Marin County, the bottom surfaces, top surfaces, and longest axes of the pillows are roughly parallel to the bedding of the enclosing rocks (fig. 23). The outer surface of a fresh pillow commonly con- sists of a rind of slickensided chloritic material about one-fourth inch thick, evidently derived from glass GREENSTONE \ 3 1 FIGURE 22.—Basalt pillows. The basalt here consists of pumpellyite, pyroxene, albite, and chlorite. West side of US. Highway 101, 11/2 miles north of Golden Gate Bridge, Marin Peninsula. formed by rapid cooling. Shallow chlorite-lined joints about 1/2—2 inches apart thoroughly dissect the sur- face and probably also formed during cooling. Other pillows have a rough, knobby surface. Some pillows show a slight increase in grain size from rim to center, although many fresh pillows are wholly aphanitic or consist wholly or partly of fine- to medium-grained varioles. Many pillows are cut by strong radial joints and veins. The radial joint pattern is considerably modified in many pillows by a swarm of randomly oriented onionskinlike fractures that cut the pillow into pieces not more than one-half inch thick (fig. 24). Although most pillows are not vesicular, the outer 3—4 inches of a few are spotted with sparse vesicles or amygdules as large as three-eighths inch in diameter. Still others have vesicular cores and solid rims. Pillow structure is readily discerned in fresh and moderately altered greenstone, but it can also be recognized even in highly altered greenstone by strong curving joints that follow the surface of the pillows. Intense shearing and brecciation, however, destroy pillow structure. ' Most pillows are separated from each other by soft sheared chloritic or nontronitic clay, which may be easily dug out with the fingers. Rarely, the material between the pillows is massive red or green chert or green, gray, or brown limestone. The separation of the pillows averages 1 inch or less; a greater spacing is not uncommon, but a spacing as great as 1 foot is rare. Massive greenstone is second to pillow greenstone in abundance. Although its altered and fractured con- dition does not permit positive identification of flow surfaces or flow jointing or banding, much of this greenstone is believed to have originated as submarine lava flows, for it is closely associated with pillow greenstone, radiolarian chert, and marine sandstone. The better exposed contacts show that the association of massive greenstone with these rocks is probably depositional, although many contacts are abrupt faults perpendicular to the bedding. Tabular bodies of mas- sive and pillow greenstone, 2—5 feet thick, intrude radiolarian chert and shale. At some places they con- nect with irregular intrusive bodies 10—15 feet in diameter. In a few localities massive greenstone is moderately vesicular or amygdaloidal. Pyroclastic greenstones are minor in bulk, but they occur in small volume in nearly every greenstone terrane. One of the largest exposures of pyroclastic greenstone is on Marin Peninsula facing Golden Gate channel at the west end of a small elbow beach, 3,500 feet west of Lime Point (fig. 34). The rocks are bedded tuffs, lapilli tuff, agglomerate, and pillow lava, all dip- 32 GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE FIGURE 23.—Pillow basalt of the Franciscan Formation, Fort Baker, Marin Peninsula. Dip down to the right (southwest) is suggested by the orientation of the pillows. ping 35°—45° southwest. The lowermost bed, about 130 feet thick, is mostly grayish-orange laminated tufi containing a few lenses and nodules of red and green chert and a few agglomeratic beds made up of 1—4- inch ejecta. The overlying bed, about 40 feet thick, is predominantly agglomerate, but it contains a few thin flows of pillow lava and a few beds of lapilli tuff. The topmost part of the section, about 80 feet thick, is mostly pillow lava with a minor amount of lapilli tuff. Reddish-brown tulf, lapilli tuff, and agglomerate beds, closely associated with pillow lava, can also be seen at Stuart Point, Knox Point, and northwest of Simpton Point on Angel Island, on Sunset Heights (near 16th Avenue and Moraga Street) in San Francisco, and on Twin Peaks at the south edge of the quadrangle. In the volcanic breccias and agglomerates, pieces as much as 1 inch in diameter are most abundant; pieces 1—4 inches in diameter are common, but larger pieces are uncommon. Greenstones at the shore between Campbell and Simpton Points on Angel Island con- tain rounded blocks, as large as 1 foot in diameter, that appear to be ejecta that may be transitional between volcanic breccia and pillow lava. Tufi beds are sparse. They are of fine to medium grain size. Some of them may have been mapped inad- vertently with sandstone and shale of the Franciscan Formation, which they resemble closely and into which many of them grade. Pipelike bodies of greenstone consisting of unsorted breccia pieces as much as 8 inches in diameter may be volcanic vent fillings. MINERALOGY-MICROSCOPIC FEATURES The greenstones of San Francisco are aphanitic to fine-grained altered basalts or altered rocks of basaltic chemical composition. The most common textures in basalt are intergranular, intersertal, or subophitic. Ophitic and hyaloophitic textures are less common. One thin section may show several textures. Glomero- porphyritic clots of pyroxene ophitically enclosing plagioclase laths are common in some pillow basalts. Other pillow basalts contain intergrowths of plagio- clase and pyroxene laths that are arranged radially as varioles in the outer half of the pillow. Plagioclase laths in basalt reach lengths of 2 mm and vary from andesine to labradorite (An35—An62). Pyroxene is mostly augite and pigeonitic augite; pigeonite and titaniferous augite are less common. One rock con- tains aegerine microphenocrysts, whereas the ground- mass pyroxene is plumose augite. Magnetite and ilmenite are abundant in basalt. Olivine, now replaced by chlorite, is found in a few basalts. Pyrite is com- mon in small amounts in the freshest basalts; apatite is rare. A medium-grained basalt northwest of Horse- shoe Bay contains about 5 percent quartz; its major constituents are labradorite and augite. The basalts are all altered to varying degrees. Non- tronite, chlorite, and leucoxene are common altera- tion products. Plagioclase cores in some basalts are altered to epidote and micaceous minerals. Veinlets of secondary minerals are abundant and include non- tronite, chlorite, calcite, aragonite, quartz, stilpno- melane, hematite, leucoxene, zeolites, pumpellyite, and albite. The occurrence of veinlets of pumpellyite, albite, and laumontite indicates a gradational metamorphism to the zeolite facies (Turner and Verhoogen, 1960, p. 532). Tiny amygdules, about 1 mm in diameter, are found sparsely disseminated in most basalts. Most of them are chlorite and (or) nontronite. Larger amyg- dules are abundant in some rocks and consist of prehnite, calcite, and mordenite, enclosed and veined by chlorite and nontronite. The altered rocks of basaltic composition (zeolite facies) contain relict basalt textures preserved mostly GREENSTONE 33 by primary pyroxene and by albite (Ano—Ane), chlorite, and pumpellyite. Some rocks contain abun- dant altered olivine phenocrysts; others may be spil- ites (Turner and Verhoogen, 1960, p. 258). Pyroxene is generally pigeonitic augite or titaniferous augite, though it is absent in some rocks. Nontronite, stilpno- melane, quartz, calcite, epidote, sphene, amphiboles, zeolites, iron oxides, and leucoxene are also present. Calcic plagioclase is generally sparse. Greenstones on Rincon Hill, Russian Hill, Lone Mountain, and a few other locations are porphyritic with plagioclase (albite, Ang, to andesine, An43) phenocrysts as much as 5 mm long set in a dense, cryptocrystalline groundmass. Albite is also in the groundmass as microlites or as clear rims on cloudy plagioclase microlites. Another common type of greenstone is typified by the dense pillow lavas exposed in cuts along US. Highway 101, 1,000 feet north of the twin vehicular tunnels in Marin County; this rock consists largely of varioles, a few microphenocrysts, and abundant veinlets. Most of the varioles are radially arranged laths of pumpellyite, diopsidic augite, and albite, small amounts of opaque ores, and sphene. Pyroxene and olivine microphenocrysts are replaced by chlorite and pumpellyite; calcite, quartz, albite, and pumpellyite veins cut the rock. Veinlets of quartz with borders of pumpellyite needles are very common. WEATHERING AND HYDROTHERMAL ALTERATION Most exposed greenstone is weathered or hydro- thermally altered to brown and orange rock. Fresh gray and green greenstone is rarely seen in natural exposures except at steep shores that are being actively washed by waves. At such places on Marin Peninsula and on Angel Island, the weathered zone is a band 10—40 feet thick. The base of the weathered band is a 1—2-foot-thick transitional zone that is parallel to the configuration of the ground surface at the top of the exposure. Hydrothermally altered greenstone masses generally do not conform to the configuration and may be much thicker than the zone of weathered greenstone. In many places weathered greenstone is hard to distinguish from hydrothermally altered greenstone. The rocks in many natural expo- sures, moreover, have been affected by both processes. Hydrothermal alteration in greenstone is most intense in and adjacent to faults and at contacts with radiolarian chert. The zone of hydrothermal alteration along faults may be as much as 50 feet wide. Intense hydrothermal alteration along many contacts with radiolarian chert has converted the greenstone, in an irregular zone 5-20 feet wide, to a soft clayey and sandy material of a distinctive grayish-orange color (fig. 24). Because of its light color and softness, this material resembles an acid tuff; however, hand-lens examination reveals a relict intergranular texture. Wide, gradational, and irregular contacts with fresher massive or pillow lavas also suggest altered greenstone. The mineralogy of the altered greenstones is vari- able. In weathering, the chief alteration product is generally nontronite, an iron-bearing member of the swelling-clay group montmorillonite. An early effect of weathering is the conversion of chlorite to vermicu- lite and nontronite. At a further stage of weathering, feldspars, pyroxene, and other minerals are converted to nontronite. Hydrothermally altered greenstone contains one or more of the following minerals in the clay-size frac- tion (smaller than 2 microns): Vermiculite, halloy- site, hydrous mica, chlorite, and nontronite. Random mixed-layering of mica, chlorite, and vermiculite is common. In many localities halloysite or hydrous mica is the predominating mineral in the greenstone at con- tacts with radiolarian chert. The formation of halloy- site, a common mineral in hydrothermally altered greenstone, is favored by an acidic environment during alteration (Keller, 1956, p. 2701). Radiolarian chert at contacts with greenstone is often replaced almost entirely by oxidized manganese minerals in a zone a few inches thick. Joints in the adjacent chert and in highly altered greenstone are lined with fihns of con- spicuous black manganese oxide minerals. The chemical composition of greenstones is given in table 6. The composition of the basalt (analysis 1) is similar to that of the pumpellyite-pyroxene-albite- chlorite rocks (analyses 2, 3, 4), and compositions of both rock types generally fall within the composition range for basalts as given by Kuno, Yamasaki, Iida, and Nagashima (1957, p. 213) , Turner and Verhoogen (1960, p. 208, 220), Poldervaart (1955, p. 134), and MacDonald and Katsura (1961, p. 362). OCCURRENCE MARIN PENINSULA On Marin Peninsula, the greatest thickness of greenstone in the quadrangle, totaling 4,800 feet, is exposed at several stratigraphic positions in the thick southwestward-dipping section of the Franciscan Formation. Here the greenstone consists predomi- nantly of pillow lavas but also includes massive greenstone and pyroclastic rocks. The lowermost mass exposed along the shore of San Francisco Bay west of Yellow Bluff is at least 1,700 feet thick; it may be as much as 2,700 feet thick if it underlies the Quater- nary deposits in the valley north of Horseshoe Bay. 34 GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE FIGURE 24.—Radiolarian chert, KJc, overlying basaltic greenstone, KJg, all within the Franciscan Formation. The contact is offset by a small fault. Light-colored greenstone at the contact is almost completely altered to clay minerals. North roadcut on lateral road to Sausalito, 400 feet east of US. Highway 101, 1.2 miles north of Lime Point, Marin Peninsula. Radiolarian chert north of Horseshoe Bay is inter- bedded with this greenstone. The chert increases in thickness northward, and at a point 11/2 miles north- west of Yellow Bluff, the greenstone lenses out (pl. 1). Such large variations in thickness over short dis- tances and close association with radiolarian chert are characteristic of greenstone. Unaltered basalt containing augite and labradorite is especially abun- dant, though erratically distributed, near Lime Point and west of Horseshoe Bay. SAN FRANCISCO The largest mass of greenstone in San Francisco, approximately 2,400 feet thick, is exposed from Twin Peaks southward into the adjoining quadrangle. Two greenstone bodies, 700—1,000 feet thick, are found in the hills between Portola Drive and Valencia Street. Unaltered basalt containing augite and labradorite is also erratically distributed in this area. A large mass of greenstone may underlie the cover of dune sand at Buena Vista Park hill. Small exposures of greenstone are found widely scattered in other parts of San Francisco. A small body of porphyritic dacite breccia near Lands End and one of medium-grained hornblende quartz keratophyre along Market Street east of Twin Peaks are included with the greenstones, but nothing is known of their field relations with other rock types. In contrast with the great volume of greenstone interbedded with the sedimentary rocks elsewhere, greenstone is almost completely absent from the elastic rocks exposed between Rincon Hill and the Presidio. Volcanism, therefore, was largely inactive when these rocks were deposited; these rocks prob- ably are not the same age as the Franciscan Forma- tion a short distance to the north on Marin Peninsula and a short distance to the south at Twin Peaks and vicinity, unless—as seems unlikely—volcanic centers were so localized that erupted material did not reach this area. ’ GREENSTONE 35 TABLE 6.—Analyses of greenstone samples of the Franciscan Formation [Chemical analyses (rapid rock methods) by Paul Elmore, Samuel Botts, I. H. Barlow, and Gillison Chloe. Semiquantitative spectrochemical analyses by H. W. Worthing. Looked for but not found: As, Au, Be, Bi, Cd, Ce, Cs, Dy, Er, Eu, Gd, Ge, Hf, Hg, Ho, In, Ir, La, Lu, Nb, Nd, Os, Pd, Pr, Pt, Rb, Re, Rh, Ru, Sb, Sn, Sm, Ta, Tb, Te, Th, Tl, Tm, U, W, Zn] 1 2 3 3a 4 4a 5 Chemical analysis (weight percent) 47.8 49.3 42.6 45.6 44.9 46.2 43.3 12.5 17.3 15.2 16.3 15.9 16.4 14.4 4.8 2.6 5.7 6.1 3.9 4.0 4.0 11.7 5.5 4.4 4.7 6.6 6.8 5.6 4.8 3.7 4.8 5.1 7.9 8.1 10.9 8.3 10.8 15.8 13.3 8.5 7.2 5.2 3.0 3.8 3.1 3.3 2.7 2.8 .14 .30 .25 .02 .02 .76 .78 2.0 2.3 3.8 3.8 4.1 5.0 5.2 7.4 . .22 .4 .4 1.2 1.2 1.7 1.6 .76 .81 .95 .98 .62 .22 .08 .09 ' .08 .08 .04 .14 .18 19 .22 23 .20 <.05 2.7 .................. 1.2 .................. 1.2 99 100 100 100 100 100 S (aqua regia soluble).. 0.22 0.01 0.03 .................. 0.01 .................. 0.02 Specrfic gravity (powder) .................... 3.04 2.97 3.02 .................. 2.86 .................. 2.76 Semiquantitative spectrochemical analyses (weight percent) 0 0.000015 0 0 .003 .007 0.003 .003 .003 .003 .15 .015 .007 .007 .003 .003 03 .03 .007 .015 .003 .003 .007 .007 0007 .0007 .0007 .0007 0 0 0 0 .0003 0 0 015 .03 .015 .015 .00015 0 .0015 0 .0015 .0015 .0007 .0015 .003 03 .003 .015 .015 .015 .003 .007 .0015 0007 .0003 .0007 .00015 0 .................. 0 .................. 0 .007 .003 .................. .0015 .................. .0015 1. Basalt, 0.4 mile north of Lime Point, Marin Peninsula (sample No. 3a. Analysis No. 3 recast by subtracting CaCOa corresponding to CO: 60—804). 2. Pumpellyite—pyroxene-albite-chlorite-quartz rock. Spilite(?) slope of Lone Mountain, San Francisco (sample No. SF—97). 3. Pumpellyite-pyroxene-albite-chlorite rock. Core of pillow 1.55 miles north of Lime Point, Marin Peninsula; west out along U.S. Highway 101 (sample No. 60—805) . northwest ANGEL ISLAND AND VICINITY The exposed thickness of greenstone on Belvedere Island is approximately 1,800 feet. Greenstone at Stuart and Knox Points on Angel Island may be a southeastern extension of this body, which may have been separated from the greenstone on Marin Penin- sula by folding and erosion. The arcuate exposure of greenstone in the central part of Angel Island ranges in thickness from 100 feet at its western exposure to a maximum of 600 feet southwest of Mount Caroline Livermore and about 400 feet near Simpton Point, its eastern exposure. This eastern exposure may be a northeastern extension of the greenstone of Stuart and Knox Points, it may be stratigraphically higher, or it may be partly intrusive. content. 4. Rim of same pillow as rock of analysis No. 3 (sample No. 60—806). 4a. Analysis No. 4 recast by subtracting CaCOa corresponding to 002 content. 5. Matrixosbftween pillows of locality of analyses No. 3 and 4 (sample No. Ransome (1894, p. 201) believed that the central Angel Island greenstone body intrudes sandstone of the Franciscan Formation, but the presence of pyro- clastic and pillow-form phases at several widely scattered places suggests that it is partly or wholly volcanic or a shallow plutonic body intrusive into soft sediments on the sea floor. Unfortunately, the only well-exposed contact between this body and sandstone is near Campbell Point, where faulting obscures their relations. Intrusion, admittedly, is suggested by the swarms of 1/2—11/2-inch-thick calcite-quartz veins in the sandstone and the increase in size and abundance of these veins as the greenstone body is approached. Ransome (1894, p. 201) also suggested that “the irregular shape of the*** (greenstone body) and the accompanying contact metamorphism, prove it to be 36 GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE a true sill and not an interbedded flow.” He evidently believed it was a sill rather than a dike because of the general conformity of the greenstone-sandstone contact with the sandstone bedding. The rocks ascribed by Ransome (1894, p. 211) to contact meta- morphism are glaucophane-bearing metagreenstones that are found sporadically in small isolated masses along greenstone borders, but also within the green- stone body. J adeitization of Franciscan sandstone also seemsto be strongest at and near the greenstone border. The author believes that similar metamorphic processes affected the entire greenstone body and other rock masses far from its borders after emplace- ment. The extensive calcite-quartz veining may also be related to a later action. Pillow lavas and volcanic breccia are seen at‘sev- eral places in the sill-like body. These features are more likely to develop under extrusive conditions than under intrusive conditions. East of the large valley between Campbell and Simpton Points, the green- stone is a breccia containing reworked, rounded pieces as much as 10 inches in diameter. The largest part of the greenstone body, however, does not show pillow or breccia development and may indeed be intrusive, but evidence for its intrusive nature is inconclusive. The central Angel Island greenstone body is mostly metagreenstone, but it also contains dense variolitic greenstone. Chemical composition of the central Angel Island greenstone body is given by Ransome (1894, p. 231) and Bloxam (1960, p. 564). They showed that greenstones and metagreenstones of this body are generally similar in composition and that they are similar in composition range. (See table 6.) This similarity suggests that the formation of metagreen- stones from greenstones was mainly isochemical. The greenstone in central Angel Island, and most other greenstones in these areas, are either similar in texture and mineralogy to the dense variolitic green- stone (see greenstone discussion on “Mineralogy- Microscopic Features”) or they are slightly more metamorphosed or hydrothermally altered. The metagreenstones generally consist of pheno- crysts or large feathery plates of pyroxene as much as 3 mm in diameter in a matrix of small feathery pyroxene plates, lawsonite, blue amphibole of the glaucophane-riebeckite series, stilpnomelane, opaque ores, and sphene. Pyroxene is augite, titanaugite, diopsidic augite, and diopside-jadeite. In some rocks it is partly or almost completely replaced by blue amphibole of the glaucophane-riebeckite series, chlor- ite, lawsonite, pumpellyite, iron oxides, stilpnomelane, nontronite, clinozoisite, zoisite, muscovite, albite, cal- cite, and quartz. These secondary minerals are also found in the matrix. Skeletal pyrite is commonly inter- grown with pyroxene. ORIGIN Greenstone of the Franciscan Formation is a prod- uct of volcanism. It was emplaced as lava flows, tuffs, agglomerates, and associated dikes, sills, and plugs. Volcanism was active repeatedly during Franciscan time. Most if not all the volcanic rocks were erupted on the sea floor, for pillow structure is widespread, marine chert and limestone are found in the space between some pillows, and the greenstone flows and pyroclastics are interbedded with marine sediments. Pillow structure is also developed by shallow intru- sion of flows into wet unconsolidated sediments; small sills and dikes, 2—10 feet thick and showing pillow structure, intrude radiolarian chert at several local- ities. Such intrusives may have been parts of flows that sank into the soft siliceous gel that later hard- ened into chert. Tuffaceous greenstone grades into graywacke at sev- eral localities where ash was deposited concomitantly with graywacke. Volcanic graywackes are largely re- worked ash and lapilli. Coarse lapilli and ejecta greater than about 4 inches in diameter are evidence of local vents. Indeed, some masses of brecciated greenstone may be plugs or necks of Franciscan volcanoes. Metamorphosed and altered basalt is present in almost every greenstone body shown on the geologic map (pl. 1). Its erratic distribution, however, indi- cates great local variations in conditions of alteration and metamorphism, including (1) deuteric or hydro- thermal alteration from the action of volcanic emana- tions (Williams and others, 1955, p. 59), (2) the reaction of hot volcanic rocks with hot sea water, and (3) low-grade metamorphism of the zeolite, green- schist, and glaucophane schist facies (Turner and Verhoogen, 1960, p. 531—544). The common green- stone mineral assemblage of primary pyroxene and pumpellyite, albite, and chlorite would probably be created if basalt were subjected to zeolite-facies metamorphism (Turner and Verhoogen, 1960, p. 532) . Much of the greenstone of Angel and Belvedere Islands and Tiburon Peninsula probably was meta- morphosed under the slightly greater pressures and temperatures of the greenschist and glaucophane schist facies, as indicated by lawsonite, glaucophane, epidote, and metamorphic pyroxene. RADIOLARIAN CHERT AND SHALE The radiolarian chert and shale of the Franciscan Formation consist predominantly of thin alternating RADIOLARIAN CHERT AND SHALE 37 beds of chert and siliceous shale. In places these sedi- mentary rocks make up sections more than 1,000 feet thick. Radiolarian remains are common in the chert and less common in the shale. Massive generally iso- lated shale-free bodies of chert 4—25 feet thick, as well as thin-bedded shale-free cherts, are also included in this map unit. Interbedded chert and shale predomi- nate, however, and in this report the term “chert” is occasionally used for all three types. The larger deposits of radiolarian chert are closely associated with greenstone. Like other map units in the Fran- ciscan Formation in this quadrangle, this unit is primarily lithologic and has limited stratigraphic signi- ficance, for it occurs at several stratigraphic positions in the Franciscan Formation. Moreover, rocks at various stratigraphic positions are similar in appear- ance; no distinct characteristics distinguish one radio- larian chert section from another. Davis (1918b, p. 239—408) and Bailey, Irwin, and Jones (1964, p. 55—68) include descriptions of deposits outside the San Francisco North quadrangle. The radiolarian chert and shale unit generally forms topographically high areas because the chert resists weathering and hydrothermal alteration. The contact of radiolarian chert and shale with other rock types usually is easily recognized, for this unit commonly stands 3—10 feet higher than the adjoining rock. All the peaks in the central highland area of San Fran- cisco are underlain by radiolarian chert and shale. Most of the higher ridges of Marin Peninsula, includ- ing the highest point in the quadrangle, are also under- lain by this rock. MEGASCOPIC FEATURES Color of the radiolarian chert and shale varies with the content and state of oxidation of iron and man- ganese. The most common colors of fresh thin-bedded radiolarian chert are dusky red to dark reddish brown. Grayish-green chert is found in small volume. Gen- erally, the associated shale has the same color as the chert, except where it is altered. Under hydrothermal reducing action, the red and brown chert becomes grayish green, dark greenish gray, or grayish yellow green. Hydrothermal activity and weathering may also remove iron and manganese, giving the chert and shale a white, very light gray, grayish-orange, or light-bluish-gray color. In many places hydrothermal activity and weathering were limited and variable from place to place, and only part of the rock was affected. Mottled cherts are formed in this manner; they commonly are green, gray, and white adjacent to joints and along the bed- ding surfaces. The joints themselves and the shale between the chert beds are often stained with yellow and brown iron oxides or black or bluish-black man— ganese oxides. Some single chert beds in a chert and shale sequence contain distinctly colored blebs of red, orange, yellow, and brown chert elongated parallel to the bedding with irregular but sharp and tightly bonded borders (fig. 25). Other red chert beds con- tain pale-gray and pale-green irregular color bands that are generally parallel to the bedding and probably are formed by hydrothermal alteration or weathering. Near some greenstone bodies, radiolarian chert that is enriched in manganese is colored very dusky red; the associated shale is dusky brown or almost black. Chert colors, however, are difficult to distinguish because of the masking effects of numerous fractures lined with black manganese oxides. Bright red, bright orange, bright yellow, and brown mottling of chert and shale beds may be heat effects caused by contact with molten basalt. Massive chert interbedded with thin-bedded chert is generally paler in color than the associated thin- bedded chert and shale. Gray, white, pale green, and yellowish orange are common colors of massive chert. Metacherts of Angel Island and Tiburon and Belve- dere Peninsulas are commonly medium dark gray, brownish gray, and bluish gray. Radiolarian chert consists typically of alternating thin beds of chert and shale. Chert beds are generally 1—5 inches thick. Some beds, however, are much thicker. Rare single chert beds, 1—25 feet thick, are interbedded with chert and shale of normal thickness. Shale beds range from mere films to beds about three- fourths inch thick (fig. 26); they seldom exceed 1 or 2 inches in thickness. Very rare ones exceed 1 foot. The chert is aphanitic and hard. Fresh unfractured chert is difficult to break and, on breaking, yields sharp edges and conchoidal to hackly surfaces. Fractures, however, are almost universal in chert of the Fran- ciscan Formation. In most places the fractures are cemented, and the rock as a whole is fairly tough; nevertheless, it readily breaks to coherent pieces generally 1—3 inches in diameter. In a few places adjoining faults, the rock is severely fractured and breaks down to small splinters. Individual chert beds generally lack fissility, for fractures or surfaces of weakness parallel to the bedding are uncommon. Clastic micaceous silt-size and smaller particles occur on bedding surfaces of the shale. These particles give the shale its poor to moderate fissility. The shale is fairly well indurated, although somewhat brittle; however, some of it can be scratched with the finger- 38 GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE O |____l_._._J FIGURE 25.—Internal structure of radiolarian chert bed of the Franciscan Formation, 61/2 inches thick. Isolated blebs that appear to be light gray on the photograph are really moderate yellow (5Y 7/6) in a moderate~reddish-brown (10R 4/4) to dusky-red (5R 3/4) matrix. Structural features are elongated nail and all of it with a steel blade, in contrast with the chert, which cannot be scratched with a steel blade. Most shale remains hard when soaked in water for long periods of time. Where shale between chert layers is 1 inch or more thick, it often contains thin distinct layers rich in Radiolaria; these layers are somewhat harder than the average. Most chert beds are sharply bounded by interbedded shale, although some highly ferruginous chert grades into shale. Pinching, swelling, and lensing of chert beds are very common and take many forms. An individual chert bed may have bulges two to three times its minimum thickness a few inches to 1 foot apart (fig. 27). The underlying and overlying chert beds are parallel to bedding. Structure suggests this rock formed from silica gel masses that differed in amount, particle size, and state of oxidation and hydration of iron and manganese oxides. From quarry on Sausalito lateral, one-half mile northwest of Yellow Blufl’, Marin County. generally pinched at the point of the bulge, but the thickness of the interlayered shale beds is affected only slightly. As a chert bed lenses out and terminates, adjacent chert beds thicken or bend towards each other, whereas the two shale beds enclosing a wedged- out chert bed merge into one shale bed only slightly thicker than either of the individual shale beds. Or, the lensing ends of two chert beds may overlap; so, the thickness of adjacent chert beds remains un- changed. Each shale bed is generally less than 25 feet long in horizontal exposure. Bedding irregularities are also described and illustrated by Davis (1918b, p. 248— 252), Taliaferro and Hudson (1943, p. 227—229), and Bailey, Irwin, and Jones (1964, p. 55-68) . RADIOLARIAN CHERT AND SHALE FIGURE 26.—Radiolarian chert and shale of the Franciscan For- mation. Note pinching, swelling, and wedging out of indi- vidual chert beds. Twin Peaks, San Francisco. FIGURE 27.—Pinching and swelling in radiolarian chert beds of the Franciscan Formation. Raveled shale bed (undercut zone in shadow between the man’s hands) is unusually thick, On lateral road to Sausalito from US. Highway 101, three- fourths mile north of Lime Point, Marin Peninsula. Small tight rounded folds and sharp chevron folds are common in radiolarian chert and shale, although many exposures show only nonfolded bedding or a gentle waviness of bedding (fig. 28). Bedding surfaces of chert are generally curved rather than flat because of the small—scale tight folding and because of pinch- ing and swelling. Attitude measurements on small exposures of radiolarian chert beds, therefore, must be used with caution in deciphering the broad structural features of the Franciscan Formation. Where expo- sures are extensive, as on Twin Peaks and in Fort Baker on Marin Peninsula, local contortions as well as consistent attitudes can be recognized. 39 FIGURE 28.—Chevron folds in radiolarian chert of the Fran- ciscan Formation 2 miles northwest of Golden Gate Bridge, Marin Peninsula. The most common cause of contorted bedding in chert probably is submarine slumping of beds prior to hardening. In places, disturbance of incompletely hardened beds may have been caused by volcanic activity, such as flow movement or intrusive action. In other places, crumpling of chert appears to have preceded volcanic activity. Contorted bedding also occurs in the Vicinity of large faults. Small faults that show displacements of a few inches to 1 foot are con- fined to the axis of small folds and evidently ruptured during folding. Possible drag folding related to the formation of major folds in the Franciscan Formation can be seen in some localities, particularly on Marin Peninsula (fig. 29). Massive chert generally produces large bold bare exposures. In favorable exposures it is seen to be inter- bedded with thin-bedded radiolarian chert and shale. In less favorable exposures it commonly is isolated and is surrounded by slope debris and other surficial deposits that obscure its relation to other rock units. Massive chert bodies are generally wedge shaped with blunt ends and may have very irregular borders within thin-bedded radiolarian chert and shale sections. At many exposures thin-bedded chert and shale terminate abruptly against massive, thick-bedded, or obscurely bedded chert lenses. The largest known exposure of a single chert bed in a section of thin-bedded chert and shale is on Sunset Heights in San Francisco along 14th Avenue, 600 feet east of the intersection of Noriega Street and 16th Avenue (fig. 30). It is approximately 500 feet long and 25 feet in maximum thickness. It has an irregular bottom and lies partly on thin-bedded chert and partly on pale-green shale. The shale is as much as 40 GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE FIGURE 29.—Radiolarian chert containing possible drag folding. Above, Radiolarian chert at Lime Point, as seen looking northwest along the strike of the Franciscan Formation. Greenstone lies under the left side of the Golden Gate Bridge tower and the high cliff on the left of the tower. Chert lies below the right side of the bridge tower and is exposed in the 1 foot thick where it fills the irregular space between massive and thin-bedded chert, and its bedding is parallel to that of the underlying thin-bedded chert. Some chert immediately adjacent to greenstone or chert completely enclosed by greenstone differs from typical bedded chert. The differences may be from heat and chemical effects of the erupting molten greenstone acting on the typical thin-bedded chert or on siliceous gel prior to consolidation. The effect on some chert is brecciation and quartz veining, the absence of the shale partings, and no change in color (fig. 31). The chert layer next to greenstone is gen- erally 1 foot or more thick; adjoining chert beds are less than 4 inches. Stronger effects, usually seen on chert enclosed by greenstone, are extensive breccia- tion, quartz veining, a color change to pale or bright red, orange, yellow, green, and brown, and a drusy coating of tiny quartz crystals on joints. Microscopic examination of thin sections of this rock show that new minerals have formed. cliff to the right of the tower. A light-colored zone in the cliff across the road from the buildings is chert containing jarosite derived from the alteration of pyrite. Opposite, Possible drag folding in the chert exposed in the cliff to the right of the bridge tower. Chert beds are 1—4 inches thick. Dark part at upper right-hand corner is bottom of bridge. Fractures are numerous in thin-bedded chert and shale, especially in the chert. They are commonly nearly perpendicular to the bedding. Fractures inter- sect each other at various angles up to 90°, but most of them are more or less alined in a limited number of directions. Parallel fractures are generally 1444/; inch apart. Some fractures are slightly open fissures and com- monly are lined with manganese and (or) iron oxides. The rock splits readily along such fractures. Other fractures are tightly cemented, generally with quartz, or less commonly with calcite, gypsum, zeolites, and other minerals. About half the fractures in a typical chert bed are filled with white quartz veins that are generally less than one-eighth in thick. Thicker quartz veins penetrate the interbedded shale about an eighth inch; thus bedding surfaces of many chert beds from which the shale has been removed have a prominent reticulate pattern of raised quartz veins. Chert rarely breaks along uartz-cemented fractures; the brittle- RADIOLARIAN CHERT AND SHALE 41 FIGURE 29.—Continued. ness of chert stems from its tendency to break along incipient uncemented fractures. At faults, chert may be so minutely fractured that it can easily be shat- tered by the fingers into small splinters generally elongated perpendicular to the bedding. Fractures parallel to the bedding in chert are un- common. They do appear in some chert beds as one or two conspicuous uncemented joints. Many cherts contain numerous tiny discontinuous quartz veins or possible fossil spicules, either in a plane parallel to the bedding or very irregular in shape but generally parallel to the bedding. They are generally less than 0.5 mm wide and less than 10 mm long, but they give the chert a laminated appearance. Most massive chert bodies are not as brittle as thin- bedded chert. On close examination massive chert bodies are generally found to be highly brecciated and thoroughly recemented with one or more generations of quartz and chalcedony. Massive chert at most con- 534—039 0 —- 74 - 4 tacts with greenstone contains numerous quartz veins as much as 1—2 inches thick. These veins are more irregular than quartz veins in thin-bedded chert. CHEMICAL COMPOSITION The chemical composition of chert and related shale is given in table 7. Silica is the predominant constitu- ent. In chert, silica is ahnost entirely chalcedony and quartz. In shale, the content of silica, chalcedony, and quartz varies with the Radiolaria content. In shale containing no Radiolaria, only about half the silica is represented by these minerals and the rest is com- bined in silicates. In the cherts, SiO2 content is well above 90 percent; a massive chert at Grand View Park is nearly pure SiO2 (table 7, analysis 6). Fe203 and A1203 are generally next in order of abun- dance, except in green chert which evidently was subjected to iron leaching and reducing conditions (analysis 4). The three constituents, Si02, A1203, and 42 GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE FIGURE 30,—Massive chert in the Franciscan Formation. The bottom of a massive chert bed, 25 feet thick, interbedded with thin-bedded chert and shale. Material between massive chert and thin-bedded chert is green hydrothermally altered shale FIGURE 31.—Quartz veins in radiolarian chert lying on altered greenstone. Twin Peaks, San Francisco. containing quartz, mica, and kaolinite. The chert bed below the shale in the middle of the photograph is about 2 inches thick. West side of Sunset Heights, San Francisco; 14th Avenue near Ortega Street. Fe203, make up more than 97 percent of the cherts. Besides Si02, the shales contain substantial quantities of A1203, Fe203, and K20. As in the red cherts, iron is mostly in the ferric state of oxidation, though FeO is nearly 1 percent in a green shale which appears to have been a red shale reduced by hydrothermal fluids. Shale differs from chert in having less Si02 and more Fe203, FeO, MgO, K20, H20, P205, and MnO. Minor- element content as measured by spectrochemical analysis also differs. Shale contains more Cr, Sr, V, Y, and Zr and contains the following elements which are either absent from chert or are below the limits of detection: Be, Ce, Ga, La, M0, Nb, Nd, Pb, Sc, and Yb. Minor-element content of chert generally sub- stantiates Krauskopf’s (1955, p. 427) statement that the few analyses available to him indicate negligible RADIOLARIAN CHERT AND SHALE 43 TABLE 7 .—Analyses of radiolarian chert and shale samples of the Franciscan Formation [Chemical analyses (rapid rock methods) by Paul Elmore, Samuel Botts. I. H. Barlow, and Gillison Chloe. Semiquantitative spectrochemical analyses by H. W. Worthing. Looked for but not found: As, An. Bi, Cd, Cs, Dy, Er, Eu, Gd, Ge, Hr, Hg, Ho, In, Ir, Li. Lu, Os, Pd, Pr. Pt, Rb, Re, Rh, Ru, Sh. Sm, Ta, Tb, Te, Th. Tl, Tm, U, W. Zn] 1 * 2 3 4 5 6 7 8 9 1o 11 Chemical analyses (weight percent) 93.5 95.9 94.7 96 5 93 97.4 40.1 60.9 58.4 66.7 66.3 .96 1.1 1.1 1 5 2 .47 10.9 13.1 14.3 14.1 15.9 2.8 1.7 2.7 34 2.4 1.3 27.6 9.2 7.4 6.5 3.3 <.05 .34 .22 38 <.05 .26 <.05 1 <.05 .58 .96 11 10 .14 16 13 <.05 3.5 2.3 3.3 1.6 1.8 42 .05 .06 17 11 .05 .53 <.05 2 .05 <.05 01 .02 .01 11 11 .01 .12 .08 1.7 .10 .18 08 .26 .37 26 41 .04 4.5 4.9 3.9 3.8 4.9 72 .70 .63 65 1 .55 4.8 4.6 3.1 4 4.3 22 .11 .16 15 .26 07 2.2 2 2 1.2 1.3 04 .06 .06 08 12 .03 .76 .92 .66 .81 .76 03 .02 .03 04 05 .04 .14 14 1.3 .09 .27 1 3 .05 .05 03 .40 .02 4.8 .11 1.4 .10 .08 < 05 < 05 <.05 < 05 <.05 <.05 .08 <.05 <.05 <.05 <.05 Total .. 100 100 100 100 100 100 100 99 99 100 100 s (aqua regia soluble) . 0.01 0.00 0.01 0.01 0.01 0.02 0.00 0.02 0.03 0.01 0.02 Specific gravity (powd 2.68 2.66 2.65 2.63 2.64 2.64 2.97 2.73 2.66 2.69 2.66 Semiquantitative spectrochemical analyses (weight percent) 0.00007 0 o 0 0.000015 0 0.0007 0 0.00015 0.000015 0 0 .003 o .003 .003 .003 .007 .003 .003 .007 .015 .07 .03 .015 .15 .15 .007 .3 .03 .15 .07 .015 0 0 0 0 0 .00015 .00015 .00015 .00015 .00015 0 0 o 0 0 0 .015 .015 .03 015 .015 .0003 0 .00015 .00015 .003 .0007 .003 .0015 .03 .0015 .003 .0003 .0003 .0003 .0003 .0003 .0003 .003 .007 .0015 .003 .003 .015 .0015 .0015 .0015 .003 .003 .03 .015 .015 .007 .0015 o 0 0 o 0 0 .0007 .0007 .0007 .0007 .0007 0 0 0 0 0 0 .007 .007 .015 0 .003 0 0 0 o o 0 .0015 .0007 .0003 0 0 0 o 0 0 0 0 .0003 .0003 .0003 .0007 .0003 0 0 0 0 0 o .015 .015 .015 0 0 .007 .003 .003 .003 .007 .003 .07 .007 .03 .007 .015 0 o 0 0 .00015 0 .015 .003 .0015 .0003 0 0 o 0 0 .0007 0 .0015 .0015 .003 .0015 .0015 0 0 0 0 .0007 0 .003 .0007 .0003 0 0 .0007 .0003 .0003 .0015 0 0 .007 .0007 .15 .15 .007 .0015 .0015 .0015 .0007 .0015 .0007 .007 .15 .003 .007 .03 .0007 o 0 0 .0003 0 .003 .003 .015 .003 .0015 0 0 0 0 o ‘0 .0003 .0003 .0015 .0003 .00015 .0007 .0015 .0015 .0015 .0015 .0007 .015 .015 .015 .015 .015 1. Chert, thin-bedded; much iron and manganese oxide on joints; Twin Peaks, San Francisco (sample No. 60—800) . 2. Chert, thin-bedded; on Sausalito Lateral, 3,500 ft west of Yellow Bluff, Marin Peninsula (sample No. 60-802 . 3. Chert, thin-bedded: on Sausalito Lateral, 3,000 ft northwest of Yellow Bluff, Marin Peninsula (sample No. SF—1970) . 4. Chert, thin-bedded, green; west cut U.S. Highway 101, 8,500 ft northwest of Yellow Blufi' (sample No. SF—2043) . 5. Chert, thin-bedded; 600 ft east of 16th Ave., 1,100 ft north of Ortega St., Grand View Park, San Francisco (sample No. SF—2145) . amounts of rare metals (minor elements) in chert. However, the shales interbedded with chert of the San Francisco North quadrangle show the following maximum enrichment factors (the ratio of the minor element content to the crustal abundance of the ele- ment, see Krauskopf, 1955, p. 417—428): Ba, 12; Co, 13; Cu, 4; Mo, 15; Ni, 11; Pb, 9; rare earths including Y, 5; Sr, 5; V, 10. The shales interbedded with cherts differ from gray shales interbedded with sandstones of the - Franciscan Formation (table 7, analysis 8) mostly in having more Fe203 and K20 and less FeO and Na20. MICROSCOPIC FEATURES CHERT Thin-bedded chert consists of a matrix of chalce- dony and cryptocrystalline to microcrystalline quartz 6. Chert, massive; 600 ft east of 16th Ave., 500 ft north of Ortega St., San Francisco (sample No. SF—2143). 7. Shale bedded with chert of analysis 1 (sample No. 60—801). 8. Shale bedded with chert of analysis 2 (sample No. 60—803) . 9. Shale bed: 1 ft thick; bedded with chert; east cut U.S. Highway 101, 3,750 ft west of Yellow Bluff (sample No. SF—2111) . 10. Shale bedded with chert of analysis 5 (sample No. SF—2145A) . 11. Shale, green; 1 ft thick; below massive chert of analysis 6 and above thin- bedded chert (sample No. SF—1941). enclosing Radiolaria and a small amount of spicules. The fossils also consist of chalcedony and quartz. Veinlets that cut the matrix and the fossils consist of quartz, chalcedony, and subordinate amounts of cal- cite, gypsum, stilpnomelane, chlorite, kaolinite, rare- earth phosphates, and zeolites. In some cherts the matrix appears to be isotropic or shows only scattered specks of light under crossed nicols. This material is thought to be largely quartz of exceedingly small grain size for the following reasons: (1) The index of refraction is approximately 1.535, slightly lower than those of quartz, but far above that of opal and (2) X-ray difiraction powder analysis shows only quartz spacings and does not contain spacings of beta- cristobalite, which are obtained by X-ray diffraction analysis of most opals. Previously, Lawson (1895, p. 422), Davis (1918b, p. 255), and Taliaferro and Hud- son (1943, p. 231) reported that some cherts of the 44 GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE Franciscan Formation are composed of amorphous silica. Most grains of chalcedony and quartz in the fossils and veinlets are at least 3 and generally more than 10 times larger than grains in the matrix. The veinlets contain the largest grains of quartz, some being more than 0.5 mm in diameter. Veinlets in some chert are calcite or gypsum with a thin border of quartz. Small patches of stilpnomelane and chlorite are common. Iron and manganese oxides and rare-earth phosphates rich in cerium coat joints in some cherts. The matrix of the typical red thin-bedded chert is heavily charged with tiny brownish-red particles. The fossils and the veinlets are relatively free of these particles. The proportion of fossils to matrix varies within a single chert bed and from bed to bed. In some chert beds fossils are so crowded that they touch each other. Few if any are seen in light-colored pigment-free chert beds or in the pigment-free parts of otherwise red chert beds rich in fossils. Radiolaria generally appear in thin sections as sharply defined clear circular or conical areas, 0.04—0.5 mm in diameter, of relatively coarse chalcedony and more rarely of microcrystalline quartz. The rim of many fossils is equigranular micro- crystalline quartz enclosing a large core of one or several grains of chalcedony, each grain consisting of a'sheaf of fibers of negative optical elongation (length fast) radiating from the edge or center of the grain. The chalcedony and quartz that preserve the former chambers and pores within radiolarian tests usually contain some red pigmentation, in contrast with the clear parts of the rest of the test. Spines and other delicate surface ornamentation are generally absent. As would be expected, the color of the chert varies with the amount of pigment in the matrix: the dusky cherts contain more pigment than the light-colored ones. Parts of the matrix of dusky-red chert appear opaque under the microscope because they are so heavily charged with pigment. The pigment particles are mostly embedded in chalcedony and quartz grains and to some extent between these grains. At high magnification the pigment in red chert is seen to con- sist mostly of equidimensional and elongated rounded transparent reddish and brownish crystals that aver- age 1 micron in diameter. They have a moderate to high index of refraction. Most of them appear iso-V tropic under crossed nicols, but some, particularly the thin laths, show a moderate to high birefringence. Definitive studies of the pigment have not been re- ported, but published descriptions of these red chert beds identify the pigment as hematite or iron oxide (Davis, 1918b, p. 254; Taliaferro and Hudson, 1943, p. 150, 227, 231; Williams and others, 1954, fig. 124A, p. 363). Taliaferro and Hudson (1943, p. 232) found that manganese oxide and carbonate are also coloring agents. Davis (1918b, p. 258, 260) reported finding pyrite in green and gray chert and a glauconitelike mineral in chert from Point Richmond, 3.5 miles north of Angel Island. X-ray diffraction powder patterns of typical red thin-bedded shale-free chert show much quartz, small amounts of hematite (alpha-Fe203) , goethite, possibly lepidocrocite, and small amounts of minerals with spacings suggestive of poorly crystalline micalike and chloritelike ' structures. These minerals no doubt account for much of the A1203, MgO, K20, NaZO, and H20, and for some of theiron shown in table 7. Talia- ferro and Hudson (1943, p. 232) suggested that “the alumina, magnesia, lime, and alkalies represent the small amount of fine clayey detritus mechanically entangled when the colloidal silica was flocculated.” Silicates in the veinlets also account for part of these components. Microscopic examination of thin sections of the brown cherts of Angel Island and vicinity reveal abundant Radiolaria remains in the form of clear circular quartz-crystal aggregates that are slightly coarser in texture than the inclusion-filled quartz crystals that surround them. Also, some radiolarian structures are outlined by lepidocrocite, which may represent altered pyrite. The quartz surrounding the Radiolaria is crowded with tiny pale-yellow-green to colorless slightly pleochroic hornblende (rarely aeger- ine) needles, 2—20 microns long, and stumpy, euhedral hexagonalprisms of apatite, 3—10 microns in diameter. The apatite may be recrystallized from the chitinous organic remains found in shale interbedded with un- metamorphosed chert. Veins of magnetite, mostly altered to lepidocrocite, and of stilpnomelane are common. Pale-blue chert is interbedded, or irregularly inter- spersed, with the brown variety. It appears to be slightly more metamorphosed than the brown variety, for no radiolarian remains were found in it and much of the quartz appears to be recrystallized. It also con- tains abundant hornblende needles, some of them partly altered to stilpnomelane, and small amounts of epidote, crossite, and lepidocrocite pseudomorphing magnetite. Shale, interbedded with either pale-blue or brown chert, is dark blue in color and is completely recrystallized. It consists predominantly of various types of blue amphiboles of the glaucophane-riebeckite series and stilpnomelane and minor amounts of aeger- ine, hornblende, sphene, and lepidocrocite. The metacherts described here probably formed under conditions similar to those for Hutton’s (1940, p. 27-28) subzone Chlorite 1 of the chlorite zone. RADIOLARIAN CHERT AND SHALE 45 Microscopic examination of baked chert shows differences from unbaked chert. Most of the baked rock consists of microcrystalline quartz or large spherules of chalcedony. Radiolaria are generally obliterated but are recognized in some baked cherts. Colorless isotropic material is absent. Baking tended to recrystallize the pigment into larger particles than exist in the unbaked chert. The particles may take the form of long irregular veinlets of magnetite, 0.4 mm in maximum thickness, or of euhedral crystals of magnetite, and unidentified opaque or. deep yellowish- brown isotropic crystals of high refractive index. The pigment in the baked cherts is generally segregated into sharply defined clots and concentricshells leav- ing nearly clear irregularly shaped masses of quartz crystals between the pigmented clots. The pigment is generally confined to,chalcedony and cryptocrys- talline quartz of the finest grain size in the chert, whereas relatively unpigmented quartz is the coarsest in grain size, some crystals being nearly 1 mm in diameter. Silicate minerals are common products of baking. Unidentified microlites and unidentified long hairlike crystals, as well as larger crystals of hornblende, pumpellyite, epidote, aegerine, and stilpnomelane, are often found in these rocks. Hornblende with Z 2 dark to moderate yellow brown, X = Y = pale yellow brown to pale brownish yellow, Z /\ c = 15°, bire- fringence = 0.010 is common in chert found lying between pillows of basalt in a roadcut 2,000 feet north- west of Horseshoe Bay. Massive chert lenses are petrographically different from the thin-bedded chert that encloses them. Micro- scopic examination shows that massive chert consists largely of microcrystalline quartz and relatively coarse chalcedony spherules; the pigment appears to be in crystals larger than those in the typical red thin- bedded chert. Taliaferro and Hudson (1943, p. 260), however, found some massive chert to be partly opa- line. Irregular veins and clots of unpigmented rela- tively coarse anhedral to euhedral quartz crystals, as large as 1 mm in diameter, are found throughout the massive chert; so, the rock appears to be a breccia of older pigmented rock recemented with the clear coarse quartz. Radiolaria are not recognized with the hand lens and are obscure and sparse in thin sections examined under the microscope. Their apparent scarcity may be explained by the difficulty of seeing them against generally pale background of massive chert. Their scarcity, however, may be real and may be related to the origin of massive chert. Massive chert may contain a greater proportion of chemically precipitated silica than is found in typical thin-bedded chert. ' SHALE A few thin sections of shale interbedded with chert from the Twin Peaks area show that shale there con- sists mostly of (1) flakes and wavy veins of a micaceous mineral showing dark-reddish-brown (Y, Z) to mod- erate-reddish-brown (X) colors and (2) lens-shaped aggregates of tiny micaceous crystals that show a weak aggregate pleochroism in moderate brown hues. Pres- ent in smaller but variable amounts are veins and dis- seminated particles of dark-red-brown nearly opaque hematite and silt-size grains of quartz and colorless mica. The mica flakes and micaceous crystals, the lens- shaped aggregates, and the veins parallel the bedding, except in shale containing Radiolaria and spicules where their orientation near the fossils is parallel to the surface of the fossils. X-ray diffraction analysis of the shale indicates the presence of large amounts of mica, most of which is poorly crystallized or randomly interlayered with other micaceous minerals, moderate amounts of a chlorite- like mineral, and hematite, goethite, and quartz. No expanding-lattice clay mineral was detected. Feldspar is present in very small amounts or is absent. Radiolaria and spicules are sparse in most shale, but thin beds of shale containing as much as 70 percent Radiolaria and spicules are interbedded with Radio- laria-free shale. Radiolaria range in diameter from 0.2 to 0.05 mm. As in the chert, the Radiolaria are pre- served in microcrystalline quartz and chalcedony which contain red pigment inclusions in former chambers Of the test. Radiolaria replaced by calcium carbonate were also found in the chert of a phosphatic nodule near Laguna Honda. Taliaferro and Hudson (1943, p. 260) found that Radiolaria in their “Franciscan-Knoxville” radiolarian shales collected at a number of localities in the Coast Ranges are invariably replaced by calcium carbonate. Most shales also contain a small amount of carbon having relict plant-cell structures. Some red shales contain a small amount of light-bluish-white curved thin shells, as much as 1 mm in size, that resemble chitinous parts of arthropods and tiny shark’s teeth. These remains are preserved as carbonate fluorapatite. Green shale found between massive and thin-bedded chert on Sunset Heights in San Francisco consists of about 40 percent hydrous mica, 40 percent quartz, and 20 percent kaolinite. HYDROTHERMAL ALTERATION Radiolarian chert and shale are hydrothermally altered along faults and along greenstone bodies. Chert becomes either grayish orange or light gray, and shale generally is grayish orange. In some localities the rock 46 GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE is green from ferrous iron alteration products. In some chert sections the color changes and other effects of hydrothermal alteration are confined to bands along joints; these give the rocks a mottled appearance. Hydrothermal effects may follow the bedding of sec- tions 1—25 feet thick in the midst of a section of unal- tered beds or may cross the bedding (fig. 32). Altered chert beds generally preserve much of the hardness of fresh chert beds. Altered shale beds soften and are generally plastic when wet. X-ray diffraction analyses of the <2-micron fraction of hydrothermally altered shale show mica, kaolinite, and montmorillon- ite. The mica is poorly crystalline or randomly inter- layered with expandible clay minerals. In some places hydrothermal alteration has converted both chert and shale into a soft clayey material con- taining quartz, montmorillonite, kaolinite, mica, and the hydrous iron silicate hisingerite. The clay is plastic and weak when wet. At some wide fault zones the chert is thoroughly shattered and has the properties of a friable sandy material or a clayey, moderately plastic material. Clay minerals in fault gouge consist mostly of mica, with subordinate amounts of random mixed-layered mica- montmorillonite (mostly mica) and kaolinite. Fresh radiolarian chert near zones of hydrothermal alteration commonly contains veinlets of quartz and well-crystallized kaolinite. WEATHERING Most exposures of radiolarian chert and shale show virtually no weathering effects except for raveling of the shale towards the ground surface (fig. 33). The physical properties of the chert beds appear to be little affected FIGURE 32.—Irregular band of hydrothermally altered radio- larian chert of the Franciscan Formation within a section of unaltered chert. Light chert is altered; dark chert is fresh. Northwest Mount Sutro, University of California Medical Center, San Francisco. FIGURE 33.—Radiolarian chert and shale. Near the east side of US. Highway 101, three-fourths mile north of Golden Gate Bridge, Marin Peninsula. Note the thin soil cover and ravel— ing of thin-bedded chert and shale in foreground, thick chert beds in background. by weathering, although the slight disintegration of the interbedded shale decreases the shearing strength of the whole rock mass. At some exposures, weathering has leached out iron, removed some of the silica, and colored the surface of the chert light red brown to white to a depth of 1 mm. Radiolaria stand out in relief on such surfaces (Davis, 1918b, p. 261). In discussing oxidation of manganese ores in Franciscan chert, Talia- ferro and Hudson (1943, p. 270) evidently believed that manganiferous chert may be oxidized by weather- ing processes to a depth of several feet, although they gave no depth figures and stated that the maximum migration of manganese, leaving behind porous cores of silica, is only a few inches. RADIOLARIAN CHERT AND SHALE 47 ORIGIN Chert is a marine deposit, but its environment of deposition on the sea floor, the source of silica, and the manner in which the silica becomes chert are largely unsettled questions. The probability of a deep-sea origin for the gray- wacke sandstone and shale through turbidity current transportation of detritus also dictates a probable deep-sea environment for the radiolarian chert with which they are interbedded (Sanders and Swinchatt, 1957). The ultimate source of the silica is believed to be the greenstones of the Franciscan Formation. The close association in time and space of greenstone and radiolarian chert of the Franciscan Formation of the Coast Ranges is given by Taliaferro and Hudson (1943, p. 273—274) as the strongest evidence for this belief. They stated that chert did not begin to form until vol- canism started, that the maximum development of the chert coincides with maximum volcanism, and that where volcanic rocks decrease laterally and disappear, interbedded chert does likewise. In the San Francisco North quadrangle, chert is indeed generally found near greenstone and is absent or sparse where greenstone is absent. ORIGIN OF SILICA Silica in chert can be supplied by silica-secreting organisms or by inorganic processes. Bramlette (1946, p. 37—55) presented strong evidence for the derivation of porcelaneous and cherty rocks of the Miocene Mon- terey Formation in the California Coast Ranges from diatoms through a diagenetic alteration process that dissolves finely divided opal of diatoms in diatomaceous deposits and reprecipitates it nearby and that is accom- panied by moderate load and dynamic metamorphism. Taliaferro and Hudson (1943, p. 233—234) believed the silica is derived from submarine springs, which they expect are common during volcanism, and from the reaction of hot sea water and hot lava or pyroclas- tics. Rhythmically bedded chert and shale lenses may have been originally bottom oozes consisting of colloidal silica and small amounts of colloidal iron, aluminum, manganese compounds, and other substances, mixed with clayey material and the siliceous organisms that flourished in the silica-rich waters. When colloidal silica changes from a sol to a gel, it tends to free itself of impurities. Thus the shale interbedded with the chert represents layers of impurities segregated by flocculation of silica. In aging and syneresis, water is lost and the gel hardens rapidly (Taliaferro, 1933, p. 50; 1934, p. 221—227). Results of experimental investigations of the solu- bility and precipitation of silica by Krauskopf (1956; 1967, p. 166—170) and White, Brannock, and Murata (1956, p. 27—59) show that silica forms true solutions in water, mostly in the form of dispersed molecules of monosilicic acid, H4Si04. Amorphous and colloidal silica is more soluble than crystalline silica (tridymite, cristobalite, chalcedony); quartz has the lowest solu- bility. In supersaturated solutions, colloidal silica, in amounts equal to the excess over true solution equilib- rium concentration, forms slowly at low temperatures and rapidly near and above 100°C. The amount of solu- ble silica found in sea water (1—10 ppm) is far below the experimentally determined solubility of amorphous silica in sea water (100—110 ppm at 22°-27°C; 280—310 ppm at 85°—95°C, according to Krauskopf, 1956, p. 13). Removal of silica by Radiolaria, diatoms, and siliceous sponges probably accounts for much of this difference. The remarkable undersaturation of soluble silica in sea water would seem, at first glance, to be strong evidence against the inorganic origin of chert by flocculation of colloidal silica. But the undersaturation is true only of sea water that has been analyzed. Sampling has not been extended to sea water in regions of submarine volcanic activity where hot sea water is in contact with hot molten or solidified volcanic volcanic rocks and where springs are expected to be numerous and the silica content is expected to be high, especially in quiet basins. Determination of the release of silica, alumi- num, iron, magnesium, and other elements from molten lava to heated sea water should yield data especially pertinent to the origin of chert. The high temperature of such waters would greatly increase the solubility of silica, and cooling of the water would cause super- saturation and rapid formation of colloidal silica. Thus Krauskopf (1956, p. 23) suggested that the cherts commonly associated with pillow lavas may have an inorganic origin by flocculation of colloidal silica, although he added that Bramlette’s hypothesis of dia- genetic alteration for the opaline cherts of the Monterey Formation is in good agreement with experi- mentally determined behavior of silica. ORIGIN OF CHERT Both organic and inorganic processes may have formed the chert of the Franciscan Formation. The Franciscan cherts are closely associated in the field with contemporaneous volcanic rocks and also contain abundant Radiolaria. Part of the silica in Franciscan chert is obviously radiolarian skeletons and spicules. Thus the problem of the origin of the additional silica that fills and encloses the Radiolaria and spicules and that has no recognizable organic structure remains unsolved. The amount of silica is large in some cherts in which no fossils are apparent, but even those crowded with Radiolaria and spicules probably contain 48 GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE 30—50 percent of silica showing no organic structures. Hinde remarked (1894, p. 238) “the fossilization, which has been sufficient to obliterate most of the radiolarian structure, would completely destroy all traces of the smaller and more delicate diatoms,” and he suggested this process also occurred in the cherts. In addition, Bramlette (1946, p. 55) suggested “that the Radiolaria preserved in the cherts of the Franciscan Formation represent only the heavier-shelled forms of the Radiolaria or other siliceous organisms originally present.” This difference in behavior of silica in differ- ent organisms has been confirmed experimentally by D. E. White (oral commun., 1956) , who stated that the form and stability of silica vary among silica-secreting organisms and that the silica of diatomite is unstable, fairly soluble, and very readily crystallized to chalce- dony or quartz. He also suggested that in Radiolaria skeletons the silica of the spines is more soluble than the remaining silica. Thus some of the silica of the Franciscan chert not involved in organic structures may represent diatoms and other unstable organic silica remains whose organic origin has been obliterated. ORIGIN OF SHALE The shale that is rhythmically bedded with Fran- ciscan chert differs considerably from the shale that is interbedded with graywacke sandstone of the Fran- ciscan Formation. Individual shale beds in chert seldom reach 25 feet in length, whereas a shale bed in sand- stone extends 100 feet or more. Shale in chert generally has more total iron, commonly ferric, as hydrated or anhydrous iron oxide. Shale in sandstone has mostly ferrous iron in silicate minerals. Na20 is lower and K20 is higher in the shale in chert. There is a notable lack of feldspars, as shown by X-ray diffraction analysis. These peculiarities indicate to the writer that the shale in chert probably did not form by sedimentation of detritus. Its formation by the little-known colloidal process by which impurities in silica gel tend to be pushed aside and segregated into bands better explains its characteristics. Other than the fossils, most of the constituents of the shale, as well as of the chert, are thought to have been derived by rapid chemical precipi- tation from cooling sea water. The sea water was able to acquire these constituents in solution by reaction with hot molten or freshly solidified volcanic rocks (Bailey and others, 1964, p. 66—67). The precipitates would be mostly colloidal in size and would be mixed with skeletons of siliceous organisms, and possibly a small amount of detritus sinks to the sea floor. Most of the micaceous minerals may have formed incipient crystals during precipitation from the cooling sea water and may have grown later by diagenesis. The bulk of the deposit consisted of silica gel, which lost water, shrank, and coagulated under the weight of overlying sediments; eventually it hardened into chert. ORIGIN OF COLOR The red color of the chert and the shale is probably the original color of these rocks. It is caused by fine particles of ferric oxide, probably precipitated from sea water as colloidal ferric hydroxide. Two factors may have induced precipitation: (1) a lowering of tempera- ture and consequent lowering of solubility of Fe with respect to the minerals and (2) a change from acidic conditions of sea water in contact with erupting vol- canic rocks to the alkaline conditions of average sea water (Hem and Skougstad, 1960, p. 96). The high oxygen content of sea water at great depths (Sverdrup and others, 1946, p. 748, 753) would promote the oxi- dation of any ferrous ions in sea water, especially at the high pH of average sea water (Huber, 1958, p. 133 ). Nevertheless, shales associated with graywacke sand- stone deposited at the same depth of sea water contain more ferrous iron, in the form of silicates, magnetite, and pyrite, than ferric iron probably because of a local negative Eh (redox potential) created by the presence of detrital carbonaceous material. AGE Radiolaria and spicules are common (as much as 70 percent) in the chert and in some shale but are sparse in most shale. Radiolaria can be seen in chert by hand- lens examination of wet fresh fractures. They appear as dark-gray clear round or conical masses tightly cemented in the rock. On weathered surfaces they have relief. In shale they appear with the hand lens as tiny pellets or leave pits where they were detached. Most of them are between 0.5 and 0.05 mm in diameter. Deli- cate structures, such as slender external spines, are rare and are seen only in thin sections of chert or on chert surfaces that have been etched for a few minutes in hydrofluoric acid. Obscure chitinous material is found in some shale. Radiolaria in the cherts of the Franciscan Formation were first described by Hinde (1894), who identified several genera in cherts from Angel Island and from Buri-Buri Ridge, 6—10 miles south of the quadrangle. Preservation was too poor to permit identification of species. Hinde (1894, p. 237—238) noted a similarity in character of the rock, mode of preservation of the Radiolaria, and abundance of the genus Dictyomitra in the cherts of the Franciscan Formation and in the red radiolarian jaspers and chert of Jurassic and Cre- taceous age from the Tyrol, from Switzerland, from Hungary, and other places. RADIOLARIAN CHERT AND SHALE 49 Riedel and Schlocker (1956) described several genera of Radiolaria found in shale interbedded with red radiolarian chert of the Franciscan Formation at Belmont, 18 miles southeast of the San Francisco North quadrangle. The Radiolaria there are somewhat better preserved than those found in cherts in the San Francisco North quadrangle. Some are similar to Jurassic species of other parts of the world, others are similar to Cretaceous species. Radiolaria are not yet useful in dividing the Franciscan Formation. Pessagno (1970, p. 130) recognized tentative radiolarian zones in the Great Valley sequence and suggested that the zone may be used for the Franciscan Formation. Two typical red thin-bedded radiolarian chert and shale sections, 10—20 feet thick, are interbedded with sandstone and shale west of James D. Phelan Beach State Park. They are separated stratigraphically 400— 600 feet from the Douvilleiceras-bearing sandstone of Albian age. Part or all of the sedimentary rock sequence in this area is overturned, and it is not known whether the radiolarian chert and shale sections are older or younger than the Albian sandstone. OCCURRENCE MARIN PENINSULA On Marin Peninsula, radiolarian chert and shale occur in thick zones interbedded with sandstone and greenstone at several stratigraphic positions (fig. 34). Between Lime Point and Horseshoe Bay, the two radio- larian chert sections, separated by a thick sandstone section, vary in thickness from about 50 to more than 500 feet. They can be followed along the strike for about 2 miles, which, according to Taliaferro (1943, p. 149), is an exceptional persistence for chert beds of the Franciscan Formation. Several other thick sections of radiolarian chert lies stratigraphically above and below these sections (fig. 34) . ANGEL ISLAND AND VICINITY Relatively small bodies and lenses of chert are widely distributed on Angel and Belvedere Islands and on Tiburon Peninsula. Most of the chert shown in these areas is slightly to moderately metamorphosed, . although it retains most of the field appearance of bedded chert and associated shale. The chert north of Hospital Cove on Angel Island is the least metamor- phosed. These chert bodies are unsheared. They are pale reddish or greenish brown or pale blue. The largest mass is 2,000 feet east of Stuart Point. Sheared meta- chert is shown as metamorphic rocks. These rocks are brown and rich in aegerine and (or) stilpnomelane or are dark blue and rich in amphiboles of the glauco- phane-riebeckite series. FIGURE 34.—Franciscan Formation exposed on north shore of Golden Gate, Marin Peninsula. A, View between Golden Gate Bridge and Point Diablo. Cliff left (west) of bridge is greenstone. Cliff near left edge exposes radiolarian chert on left and greenstone on right. B, Close view of wooded ravine (see A for location) 06 mile west of Golden Gate Bridge. Radiolarian chert exposed as barren rock east (right) of the ravine and forms hogback west (left) of the ravine. Pyro- clastic greenstone exposed immediately west of ravine. Note dip to west in chert and pyroclastic rocks in 400-foot cliflz'. Colluvium generally underlies dark brush. SAN FRANCISCO The largest exposures of radiolarian chert in San Francisco are the central highlands area near the south border of the quadrangle. Sunset Heights, Mount Sutro, Twin Peaks, Mount Olympus, and Corona Heights owe their height and boldness in large part to the presence of radiolarian chert (figs. 35, 36, 37). The maximum thickness is exposed on Mount Sutro, where this unit may be as much as 1,600 feet thick. The same section is exposed on Mount Olympus and Twin Peaks. It appears to lens out rapidly westward and is sepa- rated by sandstone and greenstone from a younger radiolarian chert section, approximately 800 feet thick, exposed at Corona Heights and on 18th Street at Mis- sion Park. The sandstone underlying this younger sec- tion contains numerous lenses of radiolarian chert, many of them less than 50 feet thick and 400 feet long. Small lenses of radiolarian chert are also found inter- bedded with sandstone along the shore between Fort Point and Lands End. Somewhat larger radiolarian chert bodies are in the east half of Golden Gate Park. 5O FIGURE 35.—Central highlands area of San Francisco, viewed west from Potrero Hill. Peaks underlain by radiolarian chert in the Franciscan Formation are, from left to right, Twin Peaks, Mount Sutro, and Corona Heights (barren hill below and to right of small cloud). Although contortions are common in the radiolarian chert of Sunset Heights, persistent bedding attitudes indicate that this rock dips eastward, which would place it below sandstone of the Franciscan Formation exposed at the base of Mount Sutro at Laguna Honda Reservoir and also below the radiolarian chert of Mount Sutro. On the other hand, it may be part of the Mount Sutro chert section repeated by a normal fault now concealed by dune sand, west of Seventh Avenue. ,If the radiolarian chert of Sunset Heights is a separate section, it is at least 400 feet thick. Lawson (1914, San Francisco quadrangle, structure-sections plate) showed it to be 950 feet thick in Sunset Heights. METAMORPHIC ROCKS Metamorphic rocks of the Franciscan Formation include slate, phyllonite, phyllite, fine- to coarse- grained schist, and fine- to medium-grained granular hornfelslike or granulitelike rocks that lack foliation, for which the name granofels (Goldsmith, 1959, p. 109) is useful. On the basis of field appearance, rocks mapped as metamorphic rocks were judged to be of somewhat higher metamorphic grade than the semischists, meta- cherts, and metagreenstones described. Rock types mapped as metamorphic rocks in one area, however, may be mapped as other bedrock units elsewhere. Relict textures and structures that suggest parentage are present in many of the metamorphic rocks, but because details of mineralogy, chemical composition, and fabric of the rocks are lacking, ideas as to parentage and metamorphic processes are speculative. Metamorphic rocks are most prevalent on Angel and Belvedere Islands and on Tiburon Peninsula. Almost everywhere in these places—Quarry Point excepted— the elastic sedimentary rocks are semischists, and other Franciscan rock types are metamorphosed to varying GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE FIGURE 36.—Radiolarian chert and sandstone in the Franciscan Formation, 17th Street, Mount Olympus area, San Francisco, viewed northwest. ' FIGURE 37.—Radiolarian chert and shale, Sunset Heights, San Francisco, near Ortega Street and 14th Avenue. Dune sand lies on the chert at top of hill at elevation of 600 feet above sea level. degrees. Completely recrystallized schists are also common. Metamorphic rocks from Tiburon Peninsula north of this quadrangle are described by Lawson (1914, p. 7) and Taliaferro (1943, p. 162-166). Several bodies 2—10 feet in diameter in San Francisco are included on the geologic map (pl. 1) in the sheared rocks unit or in serpentine. Some offshore rocks north of Bakers Beach and west of Lands End are meta- morphic rocks within the Fort Point—Hunters Point or City College shear zones, respectively. On the basis of their field occurrence, the meta- morphic rocks are placed into three categories: (1) METAMORPHIC ROCKS 51 Rocks on serpentine borders; (2) rocks included within serpentine; and (3) rocks distant from known serpen- tine bodies. METAMORPHIC ROCKS ON SERPENTINE BORDERS The major metamorphic rock types near serpentine borders are fine-grained tough greenish-brown, bluish- gray, and dark-yellowish- and dark-reddish-brown phyllites, schists, and less-foliated rocks. Many of the rocks have little tendency to split along the foliation. Foliation generally strikes northwestward, parallel to the bedding of nearby strata. Near the south shore of Angel Island dips are northeastward. Minor crenula- tions have maximum amplitudes and wave-lengths of about 8 inches. They generally trend and plunge north- west, but they vary considerably over distances of less than 5 feet. Metamorphic rocks on serpentine borders include the following assemblages (minerals listed in decreas- ing order of abundance; minerals making up less than 5 percent of rock listed as minor) : Group I: Tremolite-albite-chlorite with minor pumpellyite, sphene, and oxidized pyrite cubes. Tremolite-albite-sphene with minor muscovite, chlorite, and oxidized pyrite cubes. Albite-stilpnomelane-tremolite with minor sphene and calcite. Tremolite (sodic) -albite-stilpnomelane-sphene. Tremolite (sodic)-albite-sphene with minor epi- dote, muscovite, chlorite. Albite-stilpnomelane-actinolite with glaucophane rims, chlorite with minor pyrite. Quartz-tremolite-stilpnomelane with minor mus- covite, sphene, magnetite, calcite, and garnet. Group II: Tremolite-chlorite. Tremolite. Group III: Quartz - muscovite - stilpnomelane - crossite schist (fine-grained, silvery to brownish-gray with dark-blue rosettes of crossite). Quartz-garnet-hematite-stilpnomelane with minor clinozoisite and glaucophane (fine grained, dark brownish gray) with thin micaceous layers of muscovite-chlorite-iron oxide. Quartz-albite-crossite schist with minor musco- vite, sphene, stilpnomelane, magnetite, oxidized pyrite cubes, and apatite (fine grained, silvery to dark bluish gray). Quartz-chlorite with minor muscovite (fine grained, green). Quartz-garnet with minor stilpnomelane, musco- vite, glaucophane, and sphene (fine grained, dark brownish gray). Stilpnomelane-quartz-albite with minor glauco- phane (fine grained, reddish brown, with thin blue glaucophane-rich streaks). Group IV: Pumpellyite-albite-sphene rock with minor clino- pyroxene (relict microphenocrysts) and chlor- ite (very fine grained, tough, slightly schistose, greenish gray). Clinopyroxene (diopside - jadeite) - albite - sphene rock with minor chlorite (very fine grained, tough, slightly schistose, greenish gray). Tremolite - albite - clinozoisite - epidote rock with minor sphene, stilpnomelane, and pyrite (fine grained, tough, slightly schistose, greenish gray)- Group II rocks are less abundant but more conspicu- ous than group I rocks. They are very fine grained white to silvery light-gray schists with small-scale crenulations (%—1/2-in. amplitude) and well-developed cleavage parallel to the foliation. Some are in elongated pods and veins in shear zones between serpentine and the darker rocks of group I. Tremolite and tremolite- chlorite rock are generally soft and talcose. Groups III and IV are not as common but form distinct meta- morphic assemblages. The largest exposed bodies of metamorphic rocks are on the borders of serpentine along the south shore of Angel Island (fig. 38). Inland to the northwest along the west side of a large serpen- tine body, however, the bodies are isolated and do not form a continuous belt adjacent to the serpentine. Along the south shore of Angel Island, metamorphic rocks border both sides of the serpentine dike (pl. 1). The smaller body, east of the serpentine, is mostly fine grained light-green rock consisting of the first two assemblages listed in group I. Schistosity in this body is shown by layers 1 or 2 inches thick, but individual layers are tough and only slightly schistose. The larger, western body consists of a large variety of metamorphic rocks and includes most of the assemblages in groups I—IV. Schistosity is developed to various degrees. Some rocks show closely spaced highly schistose mica- ceous layers; others, rich in quartz, albite, and other nonmicaceous minerals, are tough and show only mod- erate or little tendency to break along the schistosity. Rocks of group IV are present in minor volume. PARENT ROCK TYPES Metamorphic rocks along serpentine borders are believed to be metamorphic equivalents of sedimentary and igneous rocks of the Franciscan Formation. The highly tremolitic rocks of groups I and II may have 52 GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE FIGURE 38.——Serpentine body and metamorphic rocks on Angel Island. Upper photograph is view northwest across Raccoon Strait and along Tiburon Peninsula. Serpentine is light band in center partly covered with trees. Metamorphic rocks form headlands on shore at right. Lower photograph is of the same part of the serpentine body showing its ridge topographic expression, as viewed from a boat southwest of Knox Point (white lighthouse building on shoreline at left). Serpentine ridge is light-colored mostly barren ridge between the shore cliffs and the skyline. been subjected to magnesium metasomatism, or they may be metamorphosed serpentine or peridotite. Group III rocks are probably metamorphosed cherts or arko- sic wackes. Group IV rocks are probably metamor- phosed greenstones. A rock abundant near the west border of the serpen- tine along the south shoreline of Angel Island consists of alternating reddish-brown stilpnomelane-rich beds 1—3 inches thick and bluish- and brownish-gray quartz- rich beds cut by numerous quartz veinlets and pig- mented with crossite, glaucophane, garnet, and stilp- nomelane. Mineral composition and bedding strongly indicate that the rock was originally interbedded chert and shale. Thus, quartz veinlets in the quartz-rich layers appear to be relicts of quartz veinlets in the original chert. This metamorphic rock differs from typical Franciscan chert in that the quartz-rich layers and stilpnomelane-rich layers are nearly the same thickness, whereas unmetamorphosed chert layers are typically 3 inches thick and the shaly layers are only one-fourth inch thick. The difference in thickness may be caused by metamorphism. METAMORPHIC ROCKS WITHIN SERPENTINE BODIES A great variety of metamorphic rocks occur as inclu- sions in all the serpentine bodies. They are especially abundant on Angel Island but are sparse in San Fran- cisco. The inclusions vary in size from about 2 inches to 20 feet in diameter and include a variety of textures and compositions. They appear to have a random dis- tribution with respect to serpentine borders; foliation and lineation are random in some inclusions. All these features are compatible with the belief that the inclu- sions formed under varying physical and chemical con- ditions before reaching their present position. Similar tectonic inclusions in serpentine are described by Brothers (1954, p. 616) from the Berkeley Hills, 8 miles northeast of Angel Island. The inclusions in the San Francisco North quadrangle are believed to be metamorphosed rocks of the Franciscan Formation and metamorphosed serpentine and gabbro-diabase. Inclusions are classified here according to their min- eralogy and probable parentage into five groups: Group I: Quartz-crossite. Quartz-crossite-stilpnomelane. Quartz-hornblende (with crossite aureoles) - sphene-garnet with minor stilpnomelane and cut by veins of albite-stilpnomelane-muscovite. Quartz-stilpnomelane-hornblende (with glauco- phane and crossite aureoles) with minor sphene and albite. Quartz-albite-hornblende (with crossite aureoles) - . aegerine with minor sphene, stilpnomelane, and hematite. Group II: . Garnet-chlorite-vesuvianite with minor sphene. Garnet-chlorite-diopsidic augite-sphene. Garnet - chlorite - vesuvianite - calcite with minor magnetite and pyrite. Diopside-chlorite with minor sphene. Diopside-chlorite with minor sphene and garnet. Chlorite-diopside-zoisite-sphene. Clinozoisite-diopside-garnet-vesuvianite with minor sphene and chlorite. Diopside-garnet-prehnite-chlorite-sphene. Group III: Pumpellyite-pigeonite-chlorite-sphene with minor. stilpnomelane. Pumpellyite-albite-sphene and minor calcite and chlorite. ' Chlorite-sphene-andesine (An...) . Group IV: Homblende - albite - clinozoisite - diopside - sphene with minor stilpnomelane and muscovite. Clinozoisite-diopside-sphene with minor chlorite and stilpnomelane. METAMORPHIC ROCKS 53 Group V: Tremolite-chlorite. Actinolite-glaucophane-chlorite-pumpellyite with minor calcite, muscovite, and sphene. Albite-actinolite with minor chlorite and sphene. Chlorite-actinolite with minor sphene. Chlorite-magnetite-apatite. Chlorite-magnetite-sphene-apatite. Chlorite—sphene with minor stilpnomelane. Talc - chlorite - hornblende (sodic) with minor sphene, pyrite, and stilpnomelane. Rocks in group I are medium- to dark-bluish—gray or brownish-gray fine-grained somewhat brittle schists. Most of them contain more than 50 percent quartz and in addition are cut by numerous quartz veins. They are probably metamorphosed cherts or arkosic wackes. Rocks in group II are dark greenish gray to light yellowish gray, aphanitic to medium grained, and exceedingly tough. Typically, they are spheroidal inclusions 1 1/1—4 feet in diameter. They show no folia- tion. Most of them have a light-colored core and a 1—3- inch-thick medium- to dark-gray rim. The dark rim consists mostly of chlorite and magnetite. Garnet com- position, determined by use of Sriramadas’ (1957, p. 295—296) and Winchell and Winchell’s (1951, p. 493) graphs, shows grossularite-andradite (40—70 percent grossularite) and garnet-hydrogarnet of the same gros- sularite-andradite range, but with 1—1.2 moles of water. Many rodingites show a relict porphyritic texture with phenocrysts replaced by metamorphic minerals, al- through some of the clinopyroxene appears to be pri- mary phenocrysts. One inclusion contained euhedral plagioclase laths replaced by chlorite, garnet, and vesu- vianite in a groundmass that shows structures identical to those found in tachylitic, pyroclastic, and pillow greenstone preserved mostly by cloudy garnet-hydro- garnet (of grossularite-andradite composition). Green- stone structures are also preserved, mostly as chlorite and magnetite, in the dark rim. Rocks in group II are believed to be metamorphosed igneous rocks and possibly sediments of the graywacke type. Their mineralogy indicates that calcium was added to them. Their composition and their occurrence as inclusions in serpentine suggests that they are rodingites. Rodingites have been described throughout the world as calcium-enriched gabbros or diorites intru- sive into serpentine. Rodingites on Angel Island do not intrude serpentine; rather, they are tectonic inclusions in serpentine and are intruded by serpentine (antigor- ite) veins. Thus the rock from which they were altered to rodingite is older than serpentine (Schlocker, 1960). The source of additional calcium needed to form rodingite and the conditions under which calcium was concentrated in the cores of the inclusions and mag- nesium was concentrated in the dark rim are not known. Calcium is thought to come either from solu- tions left after crystallization of ultrabasic “magma” or from pyroxenes in peridotite and may have been released during serpentinization (Turner and Ver- hoogen, 1951, p. 488). The presence of calcium-bearing pyroxenes such as diallage (diopside) in the serpentine of the San Francisco area indicates that the latter source is possible here. The diopside-garnet-prehnite- chlorite-sphene rock and the diopside-chlorite-zoisite- sphene rock were inclusions in serpentine and land- slides north of Bakers Beach. Rocks of group III are grayish-green aphanitic to fine-grained exceedingly tough unfoliated metagreen- stones. They are generally cut by blue veinlets of crocidolite. Rocks in group IV are green to greenish-gray fine- to medium-grained granoblastic metadiabase-gabbros. They are commonly cut by pumpellyite, quartz, and albite veins. Rocks of group V include types that are rich in mag- nesium minerals. They have various colors that range from dark greenish gray to pale pearly green and vari- ous textures that range from aphanitic hornfelsic at one extreme to coarse-grained porphyroblastic schist- ose at the other. They may have been subjected to magnesium metasomatism. Parentage is unknown, but some of them may be metaserpentines or metaperido— tites. The actinolite-glaucophane-chlorite-pumpellyite rock was found in serpentine in San Francisco near’the intersection of Euclid and Masonic Avenues. METAMORPHIC ROCKS DISTANT FROM KNOWN SERPENTINE BODIES Some metamorphic rocks have erratic distributions within semischists and relatively unmetamorphosed rocks of the Franciscan Formation of Angel and Belve- dere Islands and Tiburon Peninsula. They are particu- larly abundant on Angel Island, within and near the centrally located greenstone body, although a number of them are found near Blunt Point, near a small expo- sure of pillow greenstone (pl. 1). They are generally elongated bodies less than 1 to more than 5 feet thick and mostly too small to show on the geologic map (pl. 1). Schistosity, where shown, is more or less parallel to the bedding of nearby stratified rocks. Some of the metamorphic rocks occupy shear zones, and high shear- ing stresses may have been important in their forma- tion. A hydrothermal origin is also suggested by the following characteristics: (1) Crocidolite on joints in unmetamorphosed unsheared greenstone; (2) veins of albite, quartz, calcite, pumpellyite, amphiboles, chlor- ite, stilpnomelane, and opaque ore minerals in meta- 54 GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE morphic rocks; (3) rapid changes in texture and min- eralogy over short distances; and (4) concentration of metamorphic rock bodies in and near the largest green- stone body exposed on Angel Island. (For more infor- mation on these processes, see Turner and Verhoogen, 1951, p. 244, 472.) Similar observations were made by Berg (1956, p. 1581). She used these as evidence that hydrothermal metamorphism formed glaucophane schists and eclo- gites in the Franciscan Formation near Healdsburg, Calif, 60 miles northwest of San Francisco. Neverthe- less, Borg (1956, p. 1563, 1581) concluded that the metamorphism was not accompanied by metasomatism, after she had compared chemical analyses of unmeta- morphosed parent rock types and the metamorphic equivalents and found them to be similar. The analyses given in table 5 show that this is also true of unmeta- morphosed sandstone and jadeitized sandstone of Angel Island. The rocks not directly related to serpentine bodies are further segregated according to their mineralogy and (or) parentage into the following groups: Group I: Quartz-crossite with minor stilpnomelane. Quartz-crossite-aegerine. Quartz-aegerine-stilpnomelane with minor rie- beckite and sphene. Quartz—aegerine-magnetite with minor stilpnom- elane. Quartz-garnet-stilpnomelane with minor musco- vite, hornblende (sodic) , and hematite. Group II: Blue amphibole (glaucophane-riebeckite series)- lawsonite - pumpellyite - stilpnomelane - albite - chlorite with minor sphene Blue amphibole (glaucophane-riebeckite series)- lawsonite—sphene with minor stilpnomelane Glaucophane - pyroxene ( diopside - jadeite) -chlor- ite-sphene with minor clinozoisite, muscovite, and stilpnomelane Clinozoisite - chlorite - lawsonite - stilpnomelane - pyroxene (diopside-jadeite)-sphene with minor muscovite, cut by veins of lawsonite-stilpnom- elane-chlorite with minor muscovite Glaucophane - lawsonite - albite - sphene cut by quartz and calcite veins Hornblende (hastingsite)-lawsonite with minor muscovite and sphene, cut by veins of lawsonite with minor chlorite and by veins of albite with minor hornblende. Tremolite (sodic)-lawsonite-pyroxene (diopside- jadeite) with minor chlorite, stilpnomelane, glaucophane, and lepidocrocite. Group III: Glaucophane-albite-chlorite-muscovite with minor sphene. Glaucophane - chlorite - pumpellyite - sphene with minor muscovite, stilpnomelane, lawsonite, and apatite (garnet porphyroblasts completely altered to chlorite, muscovite, pumpellyite, sphene, and stilpnomelane). Glaucophane-clinozoisitic epidote-stilpnomelane with minor muscovite and sphene, cut by veins of calcite. Albite-hornblende (with glaucophane aureoles)- muscovite with minor chlorite, clinozoisite, gar- net, pyrite, and lepidocrocite. Albite—hornblende-chlorite with minor sphene, leucoxene, ilmenite, and magnetite, cut by veins of albite, hornblende, and chlorite. Albite-amphibole (sodic hornblende to tremolite) with minor chlorite, muscovite, and sphene, cut by veins of coarse-grained albite. Group I rocks are predominantly fine grained meta- cherts. Quartz makes up three-fourths of most of them. Those rich in crossite are blue; those rich in aegerine are yellowish or reddish brown or grayish red. Some rocks contain both minerals as segregated masses or as yellowish and reddish brown rosettes streaked with blue. The last assemblage listed in group I is mostly light gray with subordinate brown flakes. Some rocks in group I are interbedded with relatively unmetamor- phosed chert, especially on Belvedere Island, where blue semischistose metacherts are interbedded with colorless to red nonschistose chert. Group II rocks are predominantly fine- to coarse- grained metagreenstones that are within, and at the borders of, the large greenstone body on Angel Island. The first two assemblages listed are blue fine-grained breccia fragments in a sheared partly metamorphosed greenstone that it enclosed within relatively unmeta- morphosed greenstone breccia. The glaucophane-law- sonite-albite-sphene rock is a blue fine-grained tectonic inclusion more than 20 feet in diameter in the shear zone west of Lands End. Group III rocks are fine- to coarse-grained blue or grayish-green generally micaceous completely recrys- tallized schists. On Angel Island they are spatially related to the large greenstone body and crop out at the unnamed point 3,000 feet north of Blunt Point. Parentage and conditions of metamorphism are uncer- tain. Because of the lack of chemical analyses, guesses about their parentage cannot be very selective. Their association with greenstone and sandstone and their mineralogy suggest that they are probably metamor- phosed greenstone or arkosic wacke. The presence of late stilpnomelane and late pumpellyite suggests that SHEARED ROCKS 55 chlorite zone conditions prevailed for some of them at a late stage of metamorphism. TECTONIC SETTING OF METAMORPHIC ROCKS Metamorphic rocks of Angel Island are part of an irregular northwestward-trending zone of metamorphic rocks, occurring mostly as tectonic inclusions, that follows the Tiburon Peninsula syncline for at least 6 miles and includes Tiburon Peninsula and Belvedere Island. Perhaps this belt was an exceptionally strong axis of downwarping in which the Franciscan rocks were metamorphosed to varying degrees, depending on local conditions of shear, water-vapor pressure, temper- ature, and chemical environment. The tectonic inclu- sions within and on the border of serpentine evidently were emplaced during shearing. Lack of schistosity in the numerous rodingite inclusions in serpentine sug- gests that they were enriched in calcium after reaching their present position. Metamorphic rocks within semi- schists and relatively unmetamorphosed rocks of the Franciscan Formation in most places appear to be tectonic inclusions, but exposures are not complete enough to confirm this for all occurrences. . Metamorphic rocks in San Francisco are tectonic inclusions within the Hunters Point—Fort Point and the City College shear zones. Ransome (1894, p. 211, 222) believed that the meta- morphic rocks of Angel Island were formed by “local contact metamorphism” along intrusive borders of greenstone and serpentine. He explained the greater abundance of metamorphic rocks on the west side of the serpentine dike as due to greater intensity of meta- morphic processes on the upper side of the dike. His report on the geology of Angel Island first clearly described the transition of the unmetamorphosed rocks into metamorphosed rocks and the close spatial rela- tions of the metamorphic rocks and the mafic and ultramafic intrusive rocks. Early thinking on the origin of the Franciscan metamorphic rocks was also indi- cated by Ransome (1894, p. 211): As the existing literature on Coast Range geology makes no mention of “glaucophane schists” arising from local contact metamorphism, and generally assigns them, together with the radiolarian cherts and much of the serpentine, to widespread regional metamorphism, the results arrived at in this paper have been to a certain extent forced upon the writer against certain preconceived notions drawn from reading. Taliaferro (1943, p. 168, 182) believed that the metamorphic rocks are the result of local pneumato- lytic metamorphism or metasomatism accomplished by emanations from mafic and ultramafic intrusives and that the schistosity is relict bedding. The writer believes that metamorphism was accomplished by one or more of the following: Hydrothermal solutions, metasomatism, local shear, or hydrostatic pressure. HYDROTHERMAL ALTERATION AND WEATHERING The metamorphic rock body exposed about 750 feet northeast of Knox Point is altered to a grayish-orange crumbly schist composed mostly of muscovite, vermic- ulte, quartz, and albite, and of a small amount of glaucophane. Small patches of partly altered bluish glaucophane-rich schist within the altered schist sug- gest that some or all of the vermiculite was derived from glaucophane. The schist and underlying green- stone are believed to be hydrothermally altered. Weathering of metamorphic rocks has been slight. SHEARED ROCKS Zones of intense shearing that include several types of bedrock are distinguished on the geologic map (pl. 1) from zones of intense shearing in but one rock type. These shear zones generally consist of a mixture of hard blocks of bedrock, from less than 1 inch to 25 feet or more in diameter, in a soft intensely sheared matrix. The largest of these blocks is about 0.4 mile by 0.6 mile, in the Potrero Hill area. The blocks are generally rounded and surrounded by slickensided shear sur- faces. The matrix is cut by abundant closely spaced shear surfaces. It consists generally of incompetent serpentine or shale that has broken down and generally moved plastically. Other incompetent material, such as hydrothermally altered rock, is found locally, but sheared shale is the only matrix material in many zones (fig. 11). Sheared serpentine and sheared shale resemble each other superficially, and they are difficult to distinguish in the field. Because of this resemblance, serpentine has been erroneously reported in previous literature in some areas. Shearing of Franciscan rocks is greatest in shale, especially at and near serpentine borders, and the width of shear zones varies considerably. In some places, shear zones between the Franciscan Formation and serpentine are gradational, are more than 100 feet wide, and consist of an intimate mixture of serpentine and other rock types. In other places, shear zones are sharply defined and are only 3—5 feet wide, and the Franciscan rocks beyond are relatively undisturbed. Serpentine in many shear zones is converted to black waxy bentonite veins consisting predominantly of montmorillonite mixed with fragments of unaltered serpentine. The color of shear zones is generally black or gray tinged with green. The color is determined by the sheared matrix. In sheared shale, carbon in thin dis- seminated streaks and films is believed to be partly responsible for the black and gray color. Chlorite is formed, in part, during shearing and gives the green tinge. 56 GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE OCCURRENCE The largest bodies of sheared rocks form the ridge between the US. Mint and the intersection of Presidio Avenue and Geary Boulevard. These rocks are pre- dominantly clastic rocks and serpentine, but they include small bodies of radiolarian chert, greenstone, gabbro-diabase, and metamorphic rocks. Some of the elastic rock bodies are more than 1,000 feet long. Rock- type relationships are similar to those on the northeast slopes of Potrero Hill, where tectonic inclusions of sandstone occur in serpentine. Franciscan rocks, how- ever, appear to be more abundant than serpentine. Depicting rock types within shear zones on the geologic map (pl. 1) generally is not feasible because the zones are very complex and mostly concealed by homes and streets in this area. The sheared rock relationship south of California Street and west of Presidio Avenue is considerably generalized from a detailed field map at a scale of 1:2,400 prepared by M. G. Bonilla and the writer when this area was being subdivided. In a shear zone in the Point Lobos—Lands End area, a large variety of Franciscan bedrock is represented, including a few glaucophane-bearing metamorphic rocks with subordinate amounts of silica-carbonate rock and serpentine. The shear zone is tentatively thought to be the northern extension of the City Col- lege fault. (See p. 92.) Bodies of sheared rocks also occur on the north slope of Twin Peaks, Lone Mountain, on the west slope of Mount Sutro, and at Point Diablo. SERPENTINE Serpentine derived from peridotite occurs in San Francisco, on Angel and Belvedere Islands, and on Tiburon Peninsula. The peridotite was largely harz- burgite; a small part of it was dunite. The present posi- tion of much of the serpentine is believed to be due to tectonic movement at relatively low temperatures. Most of the serpentine is sheared, but in a few places it is largely massive and unsheared. A few small bodies of clinopyroxenite and gabbro-diabase are found in the serpentine. Serpentine in San Francisco is exposed in a shear zone about 1 1/2 miles wide that extends from the south- east corner of the quadrangle across the city to the shoreline southwest and east of Fort Point. Other occurrences are below the south tower of Golden Gate Bridge, at James D. Phelan Beach State Park, in shear zones in the Lands End area, and in Lincoln Park. A serpentine body, about 600 feet thick, crosses the southwest part of Angel Island (fig. 38). Small ser- pentine exposures occur southwest of this body, along the southwest shore of Belvedere Island, and on Tibu- ron Peninsula. In the San Francisco North. quadrangle, as in the Coast Ranges generally, large bodies of serpentine form ridges (fig. 39). Serpentine and mixtures of ser- pentine and sheared rocks of the Franciscan Formation form the ridges of Potrero Hill, Alamo Square, and much of the high area of Fort Scott in the Presidio and continue for half a mile northwest of Fort Point as a submarine ridge to a point about midway in the Golden Gate channel (Carlson and others, 1970; Carlson and McCulloch, 1970). On Angel Island, serpentine is gen- erally exposed in ridges (fig. 44) and spurs except at the northwest end of the island where it underlies a valley. Small bodies of serpentine do not generally form ridges. Serpentine is apparently resistant to erosion, despite its generally sheared character. This is evidently due to its stability in the weathering environment and its consequent lack of a thick soil mantle. It is also due partly to its property of retaining cohesiveness after shearing by yielding to tectonic forces in generally plastic manner, instead of by shattering. It yields easily, however, to strong and continuous erosion proc- esses, such as wave attack at the base of the cliffs. Most of the serpentine clifis south of Bakers Beach are involved in landslides. Another possible explanation for the tendency of serpentine to form ridges is that some Coast Ranges serpentine bodies appear to be under stress and are moving upward and laterally over adjoining more com- petent rock. MEGASCOPIC FEATURES Most serpentine is strongly sheared and fractured (fig. 40). Typical exposures include spheroidal knobs of hard serpentine with a polished, slickensided rind in a waxy, crumbly, thoroughly sheared matrix (fig. 41). FIGURE 39.—Shoreline along Bakers Beach, the Presidio, Fort Point, and vicinity, view northeast. The cliffs south of Fort Point (far left) are huge landslides mostly in serpentine. The hill on the skyline is mostly underlain by serpentine. Bakers Beach is in the middle distance. Raised beach sands were found in the excavation for sewers for the houses on the far right on the distant slope of the Presidio. Franciscan sand- stone exposed in near cliffs. * SERPENTINE 57 FIGURE 40.—Serpentine in a landslide. Fort Point Rock area, San Francisco, viewed southwestward from Golden Gate Bridge. The serpentine was strongly sheared before land- sliding. The knobs vary in diameter from a fraction of an inch to 6 feet or more. The sheared matrix generally con- stitutes 30 percent or more of the entire rock. In some places, bands of serpentine several hundred feet wide consist largely of sheared serpentine and a few un- sheared nodules. In a few exposures, however, the ser- pentine consists of massive blocks, 2 feet or more in size, and less than 10 percent sheared serpentine. Such serpentine is shown separately on the geologic map (pl. 1) for the Potrero Hill serpentine mass and for the small serpentine body on Divisadero Street near Duboce Avenue and is also present within other ser- pentine bodies. The color of serpentine varies widely and is evidently dependent on the composition of the parent peridotite, on the serpentinization process, and on the degree and type of alteration and shearing. Most commonly, coher- ent knobs and massive serpentine derived from harz- 534—039 0 - 74 - 5 FIGURE 41.——Serpentine nodule in sheared serpentine. West end of Crissy Field, Presidio, San Francisco. burgite are bluish gray mottled with dark green and brown. Some especially hard ones are black, or green- ish or bluish black. Somewhat crumbly ones are grayish green, and the abundant joints are black or dark blue. Hard, massive serpentine thought to be derived from dunite is mostly grayish green, grayish yellow green, or mottled grayish green and reddish brown. The pre- dominating surface color of a knob, generally lighter than the interior color, is light greenish or bluish gray, but some knob surfaces are black or dark blue. The hard knobs with a black interior may have a white or yellow rind. The color of the sheared matrix surround- ing the knobs, also generally lighter than the interior of the knobs, is mostly light greenish or bluish gray, or grayish yellow. Strongly sheared and altered serpen- tine may be dark gray or white. On Angel Island many slickensided ellipsoidal knobs are embedded in sheared serpentine. These knobs are tough dark-gray to grayish-green serpentine that shows a very fine to fine-grained drusy to sugary 58 GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE appearance on a fresh fracture. Hard inclusions of metamorphic rocks are exceedingly common in sheared serpentine of Angel Island, also, but they are sparse in sheared serpentine in San Francisco. MINERALOGY AND PETROGRAPHY The spheroidal knobs and blocks of serpentine derived from harzburgite are generally studded with green, submetallic bronze-brown, or dark-brown crys- tals as much as 1 cm in diameter. These crystals have a good cleavage or parting, and they constitute 5—15 percent of the rock. They are either unaltered or partly altered pyroxene, or they are bastitic serpentine com- pletely replacing pyroxene. Where the pyroxene is largely fresh, relict olivine appears as tiny corroded crystals in the cores of mesh-texture serpentine. The pyroxene is orthorhombic, generally enstatite. Palache (1894, p. 166—167) , Ransome (1894, p.221), and Talia- ferro (1943, p. 154) also reported the presence of a small amount of the clinopyroxene diallage. The pyrox- ene is completely serpentinized in some bastite pseudo- morphs, and so the original pyroxene could not be identified; however, the clinopyroxene content is believed to have been small, and thus the peridotite prior to serpentinization was harzburgite. Serpentine believed to have been derived from dunite shows no relict pyroxene. It is not as common as serpentine derived from harzburgite. It is present on the west side of Potrero Hill. Microscopic examination of thin sections shows that most serpentine from San Francisco consists predomi- nantly of fibrous serpentine minerals. They commonly show mesh texture with length-fast alpha serpentine borders and length-slow gamma serpentine cores. The description and illustrations of this texture and its variations given by Francis (1956, p. 201—226) for a serpentine body in Scotland are equally applicable to the serpentine of this quadrangle. In some San Fran- cisco serpentine, the cores of the meshes are isotropic. This material has been referred to previously as an amorphous mineraloid “serpophite,” but Francis (1956, p. 219) explained that the isotropism is caused either by a matted growth of gamma serpentine fibers or by the 2V of the serpentine mineral passing through 90°. Alpha serpentine handpicked by Francis (1956, p. 208—210) yielded an X-ray diffraction pattern of chrysotile. Several color and textural varieties of hard nodules from the serpentine of the San Francisco North quad- rangle were investigated by X-ray difiraction. The cri- teria given by Whittaker and Zussman (1956, p. 107— 126) were used to distinguish serpentine minerals. The results, given in table 8, show that clinochrysotile, lizardite, and orthochrysotile are the predominating serpentine minerals in most serpentine in San Fran- cisco. Antigorite predominates in the serpentine sampled on Angel and Belvedere Islands. Brucite is absent except in some serpentines, where it may be present in trace amounts. Table 8 also gives bulk den- sity of serpentines. The bulk density of serpentines from the Potrero Hill and US. Mint areas ranges from 2.1 to 2.4. In the Fort Point area serpentine partly altered to montmorillonite has a value of 1.14, whereas serpentines containing unserpentinized enstatite reach values of 2.65. The antigorite-bearing serpentines of Angel and Belvedere Islands have a bulk-density range of 2.44—2.66. Mesh-texture serpentine is evidently derived from olivine, for some of it contains tiny corroded olivine relicts in the cores of the meshes and magnetite vein- lets, 0.01—0.03 mm thick, which divide the rock into polygonal areas and which are believed to have been derived by exsolution along fractures in olivine at the time of serpentinization. The areas outlined by the magnetite veinlets generally include five to 10 meshes, but the magnetite veinlets uSually follow the border of the outer meshes of the group. Nontronite and pos- sibly bowlingite or stilpnomelane are present in some veinlets instead of magnetite. It is not known whether they are alteration products or magnetite or whether they formed in lieu of magnetite during serpentiniza- tion. Opaque iron ore dust is found in the cores of some of the meshes. Chromite or picotite appears sparsely in irregular masses, some of which, especially the iron- rich opaque chromites, have rims of nontronite or, more commonly, kammererite. BASTITE Altered pyroxene or “bastite” serpentine is readily recognized in thin section by (1) the presence of closely spaced straight parallel veinlets of magnetite, bowlingite, or nontronite that follow cleavage or part- ing fractures in the original pyroxene, (2) one or two common extinction directions of the serpentine within the altered pyroxene, and (3) lack of mesh texture that is common in the matrix surrounding replaced pyrox- ene. An X-ray diffraction powder pattern of a serpen- tinized pyroxene, handpicked from a brown serpentine from the US. Mint area, showed mostly spacings of lizardite. Zussman (in Whittacker and Zussman, 1956, p. 123) reported that two of three bastites he studied gave X-ray diffraction powder patterns of lizardite; the third bastite gave a pattern of chrysotile. RELICT OLIVINE AND ORTHOPYROXENE Optical measurements (table 9) of the relict olivine and orthopyroxene confirm the presence of iron in the original peridotite, although the FeO/MgO ratio, indi- SERPENTINE 59 TABLE 8.-—Partial mineral composition and bulk density of serpentine [Mineral composition based on X-ray diffraction and thin section examination; magnetite and chrome-spinel present in all serpentine. M=major constituent; m=minor constituent: Tr.(7)=identification of minor constituent is uncertain; n.d.=not determined] a .3 33 u. "' fl *6 ‘2‘ g Bulk 3 E- b E 2.3 g ._ a: Sample No. Description density :3 E .g g 3.5 3 53 .5 g g g §.e=ss§§§-sei .... z: a = a g .c: = :—'1 >. '3 A o 0 <1 m «a a: o m o n. S Potrero Hill area: SF— 237: —1525 .......... Light yellow green, compact, porcelaneous. mesh texture ............ 2.1—2.3 M M 2133.. ...Dark greenish gray, mesh texture .. ._ 2.41 M M 274.. ...Light yellowish gray, asbestiform fibers, brittle, 11%; in. long ..................................... n.d. M 289.. ..White, waxy to fibrous compact veins n.d. M 319. .. Grayish green, compact, mesh textur .. 2.31 M M m Tr. (?) 1119.. Light blue green, opaline .............................. 2.30 M M 1501.. ...Dark yellowish green. bastitic and mesh texture ................................................ 2.40 ............ M M 2025 ...................... Grayish yellow green, compact, chalcedonic .................................................. 2.30 M m M ............ 2050 ...................... Mottled dark green and yellowish brown, mosh texture, compact, hard .................... 2.28 M M 2213 ...................... Dark greenish gray, strongly sheared, crumbly ........................................................ 2.36 M M m 322 ...................... Brown, resinous, strongly sheared, crumbly .......................................................... 2.37 M M U.S. Mint area: SF— 174 ...................... Mottled dark brown and dark gray, bastitic, hard 2.37 m M M m m ........................ 1947.. ...Pale yellow green. mesh 2.01 M M 1950.. ..Pale olive, mesh texture, moderately firm to friable ............ 2.07 M M M 1951.. ..Dusky yellow green, r 2. 2 M M m 1953.. ..Greenish gray with pale green yellow mesh cores ........................................ 2.28 M M Tr. (7) 1954 ...................... Mottled dark green and dark gray, bastitic .......................................................... 2.42 M M 1955 ...................... Pale blue, chalcedonic rind on bastitic serpentine 2.18 M M m 1958 ...................... Mottled greenish yellow, olive, an greenish gray, bastitic ................ 2.21 m M Fort Point—Presidio area: ...Light greenish gray, friable 1.14 M M m M Black, bastitic, hard .............. 2.58 Tr.(?) M 'l‘r.(?) M ...Dark greenish gray, bastitic, white powdery coating ..... 2.34 M M M 1m _ 1949 ...................... Black, bastitic, hard 2.65 M M Tr.(?) m Angel Island and Belvedere Island: SF— 696 ...................... Mottled light yellow green and dusky blue, ‘ sheared but hard .......................................... 2.62 m m M m 2209 ...................... Mottled pale greenish yellow and black, sheared but hard .......................................... 2.58 m M B—26 ...................... Mottled pale yellow green and dusky blue, sheared but hard .......................................... 2.54 m M m 51 ...................... Mottled dark greenish gray and light yellow green, sheared but hard. 2.66 m m M S—10.. ...Dark gray, fine grained, hard ...... .. 2.65 M m 11.. .. Mottled moderate yellow green and black, sheared but hard .......................................... 2.52 m m M m 12 ...................... Mottled pale green, dusky blue, and white, ha rd ................................................................ 2.52 .................................... M m 44 ...................... Mottled medium gray and yellow green, ................................................................ 2.44 M M M 57 ...................... Mot:tled dark gray and pale yellow green, ................................................................ 2.63 m m M 10n coating. TABLE 9. —0ptical properties of olivine and orthopyroxene BLUE COLOR OF SERPENTINE and suggested composition Ref m, , d 2v d The characteristic blue coloring of much serpentine r we in ex an . . . Mineral Locality and composition composition is 1n thin films and crusts along shear surfaces. Its Olivine ................ Divisadero St. and Not determined .............. 84°; MgSiO = ‘ ‘ ' ' ' ‘ ' Duboce Ave. 75 Dermal compos1t10n was not investigated In detall, but 1t Orthopyroxene .. do ..A111511;a5i=01.653;p t 2 80;:1 Mgsioa: appears to be a finely disseminated mixture of magne- 3= ercen ; percen . . . o - eggggflgfg- c m tlte crystals, smaller than 2 microns 1n dlameter, and s= er e . _ _ Orthopyroxene ..South edge of Alpha=1.669— Not determined. serpentlne mlnerals- Potrero Hill. MgSiOs=88 percent; gamma=1.6‘76 MgSiOa=90 percent. 1From charts of Poldervaart (1950, p. 1073) . 2From charts of Kuno (1954, p. 40). cate by optics of enstatite, is lower than that indicated by olivine. ANTIGORITIC SERPENTINE Serpentine composed largely of antigorite is found on Angel and Belvedere Islands as hard rounded knobs within soft sheared serpentine or as sheared, but co- herent, serpentine containing tight slickensided joints 60 GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE throughout. The hard knobs are mottled dark-gray and pale-yellow-green fine-grained rock resembling basalt in hand specimen appearance. Under the microscope they are seen to consist mostly of plates of antigorite that average 0.1 mm in diameter and 0.006 mm in thick- ness. The largest plates are 0.4 mm in diameter and 0.05 mm in thickness. Veins and vague clots and segre- gations of antigorite crystals of the same general size make up the rock. Some rocks appear to show a vague foliation of the antigorite plates; others show a rec- tangular arrangement in which some plates are per- pendicular to others. Magnetite crystals are generally larger than in mesh-texture serpentine, though in some serpentine they appear to outline meshes or are parallel to relict olivine 0r pyroxene crystallographic directions. In other antigoritic serpentine magnetite is in large isolated crystals or clots as much as 2 mm in diameter surrounded by 2—5 mm of magnetite-free antigorite. Chlorite appears to pseudomorph pyroxene and con- tains parallel bands of magnetite euhedra and antigor- ite plates. Evidence for the partial conversion of mesh-texture serpentine to antigoritic serpentine is seen in thin sec- tions of some serpentines from Angel Island consisting of a matrix of antigorite plates within which walls of the meshes appear collapsed and broken into subparal- lel groups containing very little mesh core material. The supposed collapsed mesh-wall material is similar to the sheared mesh-texture described by Francis (1956, p. 218—220). X-ray diffraction analysis of ser- pentine consisting partly of collapsed mesh walls shows the presence of antigorite, clinochrysotile, and lizardite (entries S—11 and S—44, table 8). VEINS IN SERPENTINE Late veinlets of cross-fiber chrysotile are common in the mesh-texture serpentine but are sparse in the anti- goritic serpentine of Angel and Belvedere Islands. These veinlets vary from dark-green through light-yellow- green silky asbestiform fibers to white and pale-blue compact porcelaneous material. X-ray diffraction anal- ysis shows clinochrysotile is present in all and ortho- chrysotile and lizardite in some. The porcelaneous rinds on hard serpentine knobs are also largely clinochryso- tile. Many knobs have a strongly fibrous rind, as much as one-half inch thick, of fibers alined parallel to the knob surface. In a few places this material has the appear- ance of cross-fiber chrysotile asbestos, but the fibers lack the flexibility of asbestos and break when bent. The fibrous rinds generally give the X-ray diffraction pattern of splintery antigorite (Whittaker and Zuss- man, 1956, p. 121), though a few are mixtures of chrysotile and lizardite. The fibrous rinds may have formed by shearing of cross-fiber chrysotile veinlets, which evidently are planes of weakness that become shear surfaces during tectonic movement. Some rinds show gradation from cross-fiber chrysotile veinlets to parallel-fiber splintery antigorite. The clinochrysotile rinds may have formed during shearing of massive ser- pentine and may not be related to the presence of pre- viously existing veinlets. On many serpentine knobs, late cross-fiber chryso- tile veinlets, about 2 mm in maximum thickness, are oriented radially, at right angles to the knob surface. Such veinlets characteristically wedge out 1 or 2 inches from the surface of the knob. A pleochroic green and brownish-yellow mineral, Chlorite or ferrostilpnome- lane, accompanies the chrysotile in some veins. These veins appear to have formed by filling of shrinkage fractures, but no cause for such shrinking is known. Many knobs are cut by three-dimensional rectangular networks of late subparallel chrysotile veinlets (fig. 42). Veins of carbonates and magnetite are common in serpentine. The common carbonate veins are hydro- magnesite (Mg.CO;.(OH)2'3HzO) and magnesite (Mg- C03). They are conspicuously white veins and zones of nodules that reach 3 inches in thickness. Xonotlite, Ca6(Si6017) (0H2), (a:,8:1.582:0.002; 721.593: 0.002) , occurs as radial fibrous sheaves resembling pec- tolite in veins with carbonates. Calcite (CaCOB) in brown and white veins is com- mon in some places. Aragonite (CaCOa) is found as bladed crystals on joints in serpentine nodules at sev- eral places on the east slope of Potrero Hill. Aragonite also occurs as anastomosing cross-fiber veinlets, as much as one-half inch thick, and as euhedral crystals in the serpentine near Fort Point Rock. Aragonite is also reported in serpentine at Hoboken, N.J. (Palache and others, 1951, p. 190). The crystallization of ara- gonite rather than calcite from aqueous solution is evidently favored by the presence of magnesium salts and a temperature of BOO—70°C (Palache and others, 1951, p. 191). Such conditions were evidently met dur- ing hydrothermal alteration of the serpentine. Along Webster Street, at the US. Mint, sockets remaining in sheared serpentine matrix after the re- moval of hard knobs have a 14 -inch-thick lining of mag- nesite, possibly derived from chrysotile or splintery antigorite. Thin sections of serpentine from Potereo Hill show veinlets of cross-fiber chrysotile grading to fibrous magnesite. SHEARED SERPENTINE In San Francisco the sheared matrix enclosing the hard ellipsoidal knobs of serpentine contains the same serpentine minerals that are found in the knobs. Soft sheared matrix of antigoritic serpentine knobs on Angel SERPENTINE 61 FIGURE 42.—Clinochrysotile veinlets in bastitic serpentine. Fort Point, San Francisco. Island and Belvedere were not analyzed. Montmoril- lonite is common and in some sheared serpentine may be a major constituent. Talc is present locally. A light- blue to White strongly sheared friable serpentine is found in several places in the quadrangle. It swells and becomes plastic when wet and consists mostly of clino- chrysotile and pyroaurite. No montmorillonite could be found in this material. The swelling and plasticity seems to be related to the clinochrysotile which forms a flufiy sediment and colloidal suspension when agi- tated in water. CHEMICAL COMPOSITION The approximate chemical composition of serpen- tines was obtained by semiquantitative emission spec- trochemical analysis (table 10). The nickel, chromium, cobalt, and scandium contents are compatible with the results of Faust, Murata, and Fahey (1956, p. 318—320) on ultrabasic rocks of eastern United States and of Europe. Page (1968) and Page and Coleman (1967) gave partial chemical composition of serpentine from Tiburon Peninsula and complete chemical composition of serpentine minerals from other localities. 62 GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE TABLE 10.—Approximate semiquantitative spectrochemical analyses of serpentine and nontronite veins in serpentine [Analyst: H. W. Worthing. Looked for but not found: As. Au, Be, Bi. Cd, Ce, Cs, Dy, Er, Eu, Gd, Ge, Hf, Hg, Ho, In, Ir, La, Lu, Nb, Nd, 0s, P, Pd, Pr, Pt, Rb, Re, Rh, Ru, Sb, Sm, Ta, Tb, Te, Th, Tl, Tm, U, W, Yb, Zn. M=major constituent > 10 percent] 1 2 3 4 5 6 7 Si M M M M M M A] 0 3 0.3 0.3 0.3 0.07 1.5 Fe 7 7 7 7 7 7 7 Mg M M M M 3 1.5 Ca 015 07 .7 .15 .03 .07 .15 Na 0 0 0 .03 .015 .03 .15 K 0 0 0 0 .7 .7 Ti .015 .0015 .003 .003 .0007 .07 .07 Mn .07 .07 .15 .15 .7 .7 7 Ag 0 0 0 0 0 0 000015 B .003 .007 .003 .007 0 .003 .007 Ba .0003 .00015 0003 .00015 .0003 .0015 .003 Co .007 .007 .007 .007 .007 .003 .003 Ct .07 .07 .07 .07 .07 .03 .015 Cu .0003 .0007 .0015 .0007 .0007 .0007 .0015 G9. 0 0 0 0 0 .00015 00015 Mo 0 0 0 .0007 .0007 Ni 3 3 .3 .3 3 .3 .3 Pb .00015 0 0 0 .0015 Sc .0003 .0003 .0003 .0007 .0003 .0015 .0015 Sn 0 0 0 .0003 0 .0003 .0003 Sr 0 0 .0007 .0003 0 .0003 .0003 V .003 .003 .0015 .003 .003 .003 .003 Y 0 0 0 0 0 0 .0003 Zr .0007 .0015 .0007 .0015 .0015 .003 .007 1. Serpentine; mostly antigorite; Angel Island (sample No. SF~1604) . 2. Serpentinized harzburgite; no primary minerals: shattered bluish nodule; U.S. Mint, San Francisco (sample No. SF-—1956) . 3. Serpentinized harzburgite; tough brownish nodule. Most of enstatite is unserpentinized; small amount of primary olivine; U.S. Mint, San Francisco (sample No. 81371957). 4. Serpentinized harzburgite; tough rock. Most of enstatite is unserpentin- WEATHERING AND HYDROTHERMAL ALTERATION Soil development on serpentine is very slight, and only in a few places does the surficial mantle, generally less than 1 foot thick, appear to be soil formed by weathering processes. The white or yellow rind on black knobs of hard harzburgite serpentine may have formed during weathering. The serpentine at and near Fort Point is altered to hydromagnesite, pyroaurite (MgsFe2C03(OH)m'4H2 O), coalingite (MngegC03(OH)24'2H2O), and nes- quehonite, MgC03'3H20 (Mumpton, 1965) . The hydro- magnesite forms white nodules as much as half an inch in diameter in a talcose matrix of serpentine and pyro- aurite. Coalingite occurs in golden brown veinlets 1—2 mm thick in serpentine largely altered to pyroaurite. Nesquehonite forms a white soft surface efllorescence at moist places in serpentine. Eakle (1901, p. 316) reported barite and gypsum in fissures in serpentine near Fort Point. Sheared serpentine at the Hunters Point area is altered to a white soft talcose mixture of pyroaurite and clinochrysotile. The altered serpentine is in bodies as much as 3 feet thick. When wet the mix- ture is plastic. Mixtures of epsomite and hexahydrite commonly occur as a white soft powdery efflorescence on pro- tected surfaces of serpentine. Chromite, shiny black on fresh fractures, is sparsely scattered in the serpentine in crystals as much as 4 mm in diameter. Locally, in serpentinized dunite, it is concentrated in thin bands of blebs 1—10 mm in diameter. Millerite and pyrite are ized; small amount of primary olivine; late veins of chrysotile and stilpnomelane; Fort Point,San Francisco (sample No. SF—2056) . 5. Serpentinized dunite showing mesh texture; no primary minerals; Potrero Hill near San Francisco Hospital (sample No. SF—2133) . 6. Nontronite vein in serpentine; Potrero Hill, 18th and James Lick Freeway (sample No. SF—1851). 7. Nontronite vein in serpentine; Ellis and Gough Streets, San Francisco (sample No. SF—2134) . commonly disseminated on sheared surfaces in serpen- tine, especially near contacts with the Franciscan For- mation; they have not been found together. Octahedra of magnetite as much as 2 mm in diameter commonly stud shear surfaces. Gray and brown montmorillonitic clay minerals are common in serpentine. At many places in San Francisco, serpentine contains a variety of alteration products in the form of opaline spheroidal nodules, mammillary and botryoidal masses, and crusts. In hand specimen some of these are color- less, glassy, and transparent. Others are white, vitreous, and opaque; pale yellowish brown porcelaneous to earthy; or mottled blue, brown, and green and opaline with a greasy cast. Many are coated with a thin blue opaline film. The clear glassy ones are opal; the others contain large amounts of opal. The white ones also con- tain montmorillonite as well as manganese oxide dend- rites in addition to opal. The pale-yellowish-brown por- celaneous to earthy ones contain larger mounts of montmorillonite admixed with opal. Earthy nodules in some places have been largely destroyed, leaving thoroughly pitted serpentine. The mottled opaline ma- terial is a mixture of opal and serpentine minerals. Serpentine containing the opaline material generally shows other signs of alteration, such as gossanlike cellu- lar iron oxide-rich segregations and abundant soft waxy brown segregations of montmorillonite. Alteration to opal appears to be related to weathering. At many places serpentine is hydrothermally altered as well as strongly sheared along its contact with the SERPENTINE 63 Franciscan Formation. A common product of these processes is a dark-gray bentonite which is moderately hard when dry but exceedingly plastic when wet. This material consists largely of montmorillonite. Mont- morillonite masses vary in thickness from 1 or 2 inches to more than 20 feet. The large bodies of this material contain knobs of serpentine and Franciscan rocks. In some places, such as on Parker Avenue near DeAnza Street, white talc segregations are associated with the montmorillonite. Hydrothermal alteration within serpentine bodies has also changed serpentine into soft crumbly porous brown-to-white material of bulk density less than 2. The alteration zones vary from distinct veins, 1/2—1 foot in thickness, to bodies several hundred feet in size hav- ing indistinct borders and containing segregations of fresh hard serpentine. Alteration of the sheared matrix of serpentine is apparently more pronounced and more widespread than alteration of the knobs. Nickelian nontronite is found as brown crumbly veins 1—12 inches thick; semiquantitative X-ray spec- trochemical analyses by the writer indicate that the nickel content of the nontronite is several times greater than that in unaltered serpentine. Semiquantitative analyses of two nontronite veins are given in table 10 (samples 6, 7). The analyses indicate that the altering solutions added Al, Na, K, Ti, Ba, Ga, Mo, Sc, and Zr to the serpentine and diluted or removed Mg, Co, and Cr. Silica-carbonate rock, a product of extreme hydro- thermal alteration of serpentine, is rare. It is an exceed- ingly tough rock consisting of a complex network of veins and masses of quartz, chalcedony, opal, carbon- ates (generally magnesite), and limonite. Silica-car- bonate bodies, 5—15 feet in diameter, are seen at the U.S. Mint and at the Lands End landslide area. SHAPE OF SERPENTINE BODIES At one stage of the field mapping, shear orientation in the serpentine seemed to give a clue to the manner of intrusion and the shape of the serpentine. Shear orientation is best shown by the elongation of sheared hard knobs. In the soft matrix between the knobs, shear orientation is influenced by the knobs and is gen- erally parallel to the surface of the knobs. In general, however, shear orientation varies greatly over short distances, and no meaningful pattern evolved except, perhaps, in Potrero Hill, where most of the shearing dips eastward or is horizontal or vertical and could be interpreted to suggest that the serpentine was intruded by low-angle shearing or thrusting (fig. 43). Unfortunately, because of poor exposures, the struc- ture of the Franciscan Formation adjacent to the ser- pentine belt of San Francisco is unknown. Lawson (1895, p. 450—453; 1914, p. 6) believed the serpentine between Potrero Hill and Fort Point consists of two nearly flat lying sheets separated by a continuous sheet of sandstone about 160 feet thick and “in places so warped that they are more or less discordant with the stratification planes of the Franciscan rocks” (1914, p. 6). No field evidence was found for Lawson’s con- tinuous sheet of sandstone. Instead, the sandstone bodies in the serpentine are believed to be tectonic inclusions completely within serpentine (fig. 44). FIGURE 43.—Sheared serpentine containing large augens of altered and sheared gabbro-diabase. West side of Potrero Hill, James Lick Freeway near 19th Street, San Francisco. FIGURE 44.—Serpentine, light gray, on the left and sandstone, dark gray, on the right, separated by a shear zone containing both rock types. Boys are at the edge of the serpentine and are left of sheared serpentine, sandstone, and shale. The sandstone is a huge tectonic inclusion in serpentine. East side of Potrero Hill, San FranciSco. Iowa Street, south of 20th Street. 64 , GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE Irwin’s (1964, p. C—1, C—9) suggestion that serpen- tine belts mark westward-overriding low-angle thrust faults may very well apply to the Hunters Point—Fort Point and City College shear zones. The hypothesis would be strengthened if the Franciscan Formation lying northeast of the Hunters Point—Fort Point shear zone should prove to be considerably different in age from those west of the zone. No evidence to support an age difference was found. The border of the serpentine body on Angel Island appears to dip southwestward, whereas the adjoining sandstone dips northeastward. Ransome (1894, p. 219) described this body as a dike dipping about 55° south- westward. Conclusions could not be reached regarding the sub- surface shape of the serpentine bodies in the quad- rangle. ORIGIN Microscopic examination shows that most of the ser- pentine was derived from harzburgite and some from dunite. Evidence is sparse on the early intrusive rela- tions of the peridotite and serpentine and on the time of serpentinization. The peridotite and serpentine in the quadrangle are classified as of the alpine type because they generally occur with spilitic volcanic rocks, radiolarian chert, and graywackes in an orogenic zone (Hess, 1955, p. 392; Turner and Verhoogen, 1951, p. 239-240). The peridotite and serpentine are believed to come from the upper mantle, from which they were raised into the crust and eventually to the surface along deep faults or folds that extended into the mantle (Irwin, 1964; Hess, 1955, p. 403; DeSitter, 1956, p. 360—361). Bailey, Blake, and Jones (1970) believed the peridotite and serpentine were derived from the oceanic mantle and became the basal part of the ophi- olite sequence of the oceanic crust upon which the Great Valley sequence and the Franciscan Formation were deposited. They stated that since middle-Creta- ceous time the Franciscan Formation has been thrust (subducted) under the serpentine of the oceanic crust upon which the Great Valley sequence was deposited. The serpentine bodies of the quadrangle, however, do not appear to be part of an ophiolite sequence, for they lack an overlying layer of mafic volcanic rocks. The serpentine bodies apparently lie wholly on or within the Franciscan Formation, with the possible exception of the serpentine in the shear zone of Lands End which, as part of the City College shear zone, separates the Great Valley sequence of San Bruno Mountain and Point Lobos from the Franciscan Formation to the northeast. (See fig. 62.) Serpentine in the Fort Point— Potrero Hill—Hunters Point shear zone may have been derived from the upper mantle, for the shear zone may be a northeasterly dipping thrust fault created by underthrusting (due to sea-floor spreading) the pre- dominantly radiolarian chert and greenstone block southwest of the zone below the predominantly gray- wacke and shale block on the northeast. (See fig. 62.) Although the borders of serpentine bodies in this quadrangle are poorly exposed and hydrothermally altered, some evidence was found to suggest that the serpentine reached its present position by tectonic stresses acting on more or less solid serpentine. Con- tacts between serpentine and the Franciscan Forma- tion, wherever they were seen, are strongly sheared. Blocks of Franciscan Formation lying within serpentine have sheared slickensided exterior surfaces. The ran- dom attitude of bedding planes in such blocks suggest that they are tectonic inclusions. Furthermore, the common occurrence of serpentine as a sheared matrix enclosing ellipsoidal slickensided knobs is best ex- plained as a response of solid material of low strength to shearing stresses. Finally, the adjoining rocks of the Franciscan Formation do not show the metamorphism that would be expected at the high temperatures needed to maintain a molten peridotite or serpentine magma (Turner and Verhoogen, 1951, p. 244—252). Serpentine forms by hydration of peridotite by chem- ical reactions similar to the following for olivine and water (Turner and Verhoogen, 1951, p. 250) : 5MggSiO4 + 4H2022Mg38i205 (OH) 4 + 4MgO + SiOZ (olivine) (water) (serpentine) If MgO and SiO2 are removed in solution, the volume of olivine is almost the same as the volume of the serpen- tine formed from it. If the MgO and SiO2 contents remain the same before and after serpentinization, a volume gain of more than 50 percent is possible. The preservation of cleavage and parting of serpentinized enstatite in the harzburgite knobs in San Francisco indicates little or no volume change on a microscopic scale during serpentinization. Coleman (1971, p. 910) suggested that both constant-volume and constant- composition processes may operate in serpentinization. He finds that the most common products of serpentini- zation of alpine-type ultramafics are clinochrysotile, lizardite, antigorite, brucite, and magnetite. Coleman (1971, p. 908) gave reactions for serpentinization of olivine that show that SiO2 must be added or MgO removed in order to convert olivine to serpentine by hydration. The excess MgO can be removed from the reaction by forming brucite. Judging from the absence of brucite in serpentine derived from dunite in the quadrangle, SiO2 was added or MgO was removed dur- ing serpentinization. After using data from laboratory investigations of the system MgO-SiOZ-FeO-Fezog- H30, Coleman suggested that most serpentine derived from alpine-type ultramafic rocks in orogenic zones probably formed between 100° and 300°C and that GABBRO 65 antigorite-bearing serpentine may have formed be- tween 3000 and 550°C in deeper levels of the crust than those at which lizardite and clinochrysotile form. Barnes and O’Neil (1969) suggested that serpentiniza- tion is taking place at present at and near the surface. AGE Serpentine in the quadrangle is probably slightly younger than the rocks in which it was injected, but almost no time data are available. The serpentine at James D. Phelan Beach State Park probably reached its present position by tectonic injection into sandstone containing Douvilleiceras sp. of Early Cretaceous Albian age. Thus, the serpentine injection here may be post-Albian. The presence of rare detrital grains of picotite and orthopyroxene in some Franciscan sandstone means that ultramafic rocks were exposed, though scarce, dur- ing Franciscan time. In his study of the Coast Ranges, Taliaferro (1943, p. 153—154) found sills and plugs of serpentine in the Franciscan Formation and also in and nearly to the top of the Late Jurassic Knoxville Formation (Great Val- ley sequence of Bailey and others, 1964). He believed they “were emplaced as peridotite before the sediments had been uplifted from the basin of deposition,” but also stated that “some of the stratigraphically higher sills were largely serpentinized before complete emplacement***.” In a study of the Klamath Mountains and northern Coast Ranges, Irwin (1964, p. 0—1) suggested that ultramafic rocks, including serpentine, were emplaced during the Nevadan (Late Jurassic) and Coast Range (probably Late Cretaceous) orogenies. PYROXENITE Several masses of pyroxenite as much as 10 feet across were found in the serpentine on the south slope of Potrero Hill and along the shore north of Bakers Beach. The pyroxenite is a tough grayish- to yellowish- green coarse-grained rock that contains crystals as much as 7 mm long. It consists largely of diallage with small amounts of tiny euhedral to subhedral magnetite crystals. The diallage is slightly altered to chlorite and leucoxene. Pyroxenites have a hypidiomorphic texture, modified by postsolidification forces to a cataclastic texture. This texture is shown by breccia zones between some of the diallage crystals and by veinlike breccia zones. A tough greenish rock on the shore about half a mile south of Fort Point is regarded as a metapyroxenite. The rock contains relict pyroxene that is ragged and has been largely replaced by chlorite and by smaller amounts of clinozoisite and sphene under chlorite-zone metamorphic conditions that may have existed during serpentinization. A tough pale-green rock encased in altered serpen- tine, one-eighth mile east of Fort Point, consists of diopside, chlorite, and prehnite. It may be an altered gabbro. Ransome (1894, p. 221—222) found pyroxenite in the serpentine on Angel Island. He stated that the pyroxene crystals are “10 mm or so in length***” and identified them as diallage. Crandall (1907, p. 19) found both diallage and enstatite in pyroxenite from the Presidio area. GABBRO Small bodies of fine- to medium-grained gabbro are widely scattered in the serpentine. They vary in size from 5 feet in diameter for small equidimensional bodies to elongate bodies 100 feet by 20 feet. Only the larger ones are shown on the geologic map (pl. 1). Gabbro is dark gray where freshest and brown where weathered. Most bodies have sharp slickensided con- tacts with serpentine; the small bodies are rounded by shearing, and the elongated ones are randomly ori- ented. They are probably tectonic inclusions. Because some of the bodies have a linear arrangement and a decrease in grain size from their centers outward, Palache (1894, p. 172) suggested that they are intru- sive bodies that were disrupted by movement of the serpentine. At most places no contact effects are seen on the adjoining serpentine. The gabbro is tougher and less fractured than the serpentine and tends to crop out in relief 2—5 feet above the serpentine. On Angel Island metagabbro is so tough, so fracture free, and so rounded that it is difficult to obtain a hand specimen. At one locality on Potrero Hill, gabbro altered mostly to prehnite is in sharp con- tact with fresh gabbro. , The texture of the gabbro varies from allotriomorphic granular to subophitic; rocks with the latter texture have the field appearance of a diabase. The predomi- nating constituents are plagioclase laths (white to colorless), which generally make up one-half to two- thirds of the rock, and ferromagnesian minerals (dark), which rarely make up more than half of the rock. The gabbro of Potrero Hill in San Francisco has a slight foliation. The plagioclase is labradorite, Aneo, mostly altered and replaced by chlorite, muscovite, zoisite, and unidentified clay minerals. The ferromagnesian mineral is predominantly hornblende, in generally equidimen- tional crystals about 1 mm in diameter (ZAC=20°; X:pale brown; Y:light brown; Z=moderate to slightly greenish brown to light brown). All the hom- blende may be secondary after pyroxene, for a few crystals of colorless pyroxene (ZAC:48°), partly 66 GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE altered to hornblende, are identified in thin sections. Part of the ferromagnesian minerals is replaced by chlorite and nontronite. Present in small amounts are stumpy apatite prisms, rounded blebs of monazite, and magnetite and ilmenite filling the space between plagio- clase and hornblende crystals. A medium-grained gabbro at the Presidio, San Fran- cisco, is about 90 percent hornblende (Z AC:21°; X:pale brown; Y:light brown; Z:light bluish green). Most of the remainder is clinozoisite, actinolite, epidote, and albite. Monazite blebs are abundant in the hornblende. Actinolite needles have grown into the albite from the borders of hornblende crystals. Xonotlite and pumpellyite veinlets are common in gabbro. In a sheared and altered unidentified “basic rock” inclusion in serpentine near Fort Point, Eakle (1901, p. 316) found veins of pectolite and datolite. The rock was not found by the writer, but it has been described (Murdock .and Webb, 1956, p. 137) as an “altered diabase dike.” On Angel Island a fresh meta- gabbro consists of about two-thirds andesine laths (An42) and one-third pigeonite. The pigeonite gener- ally has rugged edges where it is partly replaced by actinolite and chlorite; in one body the pigeonite was completely replaced by actinolite. The andesine is crowded with needles and stumpy prisms of a colorless to pale-brown amphibole. About 2 percent leucoxene replaces skeletal ilmenite. Skeletal pyrite is also pres- ent. The gabbros of Angel Island and Tiburon Penin- sula are also cut by pumpellyite and chlorite veinlets. Palache ( 1894, p. 173—178) found orthorhombic and monoclinic pyroxene in some of the gabbro of Potrero Hill and named the rock hypersthene diabase. He also found the hornblendic variety, which he classed as epidiorite. Gabbro generally crops out as fresh tough rock. In some places it has a weathering mantle, 1—3 feet thick, consisting of material high in nontronite and vermiculite. The gabbro may represent acidic segregations of a periodotite magma richer than‘usual in Si02, CaO, and NaZO. If such segregations formed they probably had a lower temperature of crystallization that the perido- tite and thus may be slightly younger than the perido- tite. If serpentinization quickly followed crystallization of peridotite, gabbro may also be younger than serpen- tine. This possible age difference may explain why it is not serpentinized. As seen today, most of the bodies are tectonic inclusions in serpentine. Because the gabbro of Angel Island shows no con- tact efiects on the adjoining serpentine, Ransome (1894, p. 227—231) suggested that it is older than the peridotite. Palache (1894, p. 172) suggested that the gabbro bodies of Potrero Hill are parts of one or more continuous dikes or sills that intruded the serpentine or peridotite and were broken into the present detached masses by movement of the serpentine. Some rodingites on Angel Island and the prehnitized gabbro of Potrero Hill are metamorphosed gabbro. SURFICIAL DEPOSITS Surficial deposits of late Pleistocene and Holocene age cover approximately 80 percent of the land area of the quadrangle. They have a maximum thickness of approximately 300 feet. The Tertiary Period is not represented in the quadrangle, other than the possible occurrence of Pliocene deposits in the oldest deposits below San Francisco Bay. The oldest of the surficial deposits, the Colma Formation, is a complex of Pleisto- cene coastal sediments, which include marine estuary deposits, beach deposits, and—at higher elevations— eolian, stream, and colluvial (slope debris) deposits. Dune sands cover more than half the city of San Fran- cisco. They are the most widespread surficial deposit and reach thicknesses of 150 feet. Most dunes were actively moving in historic time. Most hill slopes are mantled with debris (colluvium) derived from underlying rock and represent in part old landslide deposits. Ravines are partly filled with col- luvium from adjoining slopes. Modern alluvium is uncommon in the quadrangle because of the absence of large streams. Alluvium is shown on plate 1 only in small deposits on Twin Peaks where it is related to a slightly older than modern drainage system that existed there when rainfall was greater than it is now. Modern landslide deposits are widely distributed on bedrock hills and along sea cliffs, though many of them are too small to show on plate 1. Mud and clay of San Francisco Bay are unconsoli- dated sediments containing clay- and silt-size detritus and large amounts of water. They have accumulated to thicknesses greater than 100 feet in the bay along the east shore of San Francisco and along the north shore of Sausalito and Belvedere Island. Artificial fill covers more than 3 square miles and is as much as 60 feet thick in some places. It consists of dune sand, colluvium, spoil from excavations and quarries, bay mud, and the general garbage of an urban area. COLMA FORMATION The Colma Formation is a group of unconsolidated sandy estuarine and coastal deposits of Pleistocene age. It was first named by Schlocker, Bonilla, and Radbruch (1958). Its type locality is south of the town of Colma, approximately 6 miles south of the quadrangle. Typical COLMA FORMATION 67 exposures are on the west side of El Camino Real extending 0.7 miles southeast of the intersection of El Camino Real with Hickey Boulevard. The lower part of the Colma Formation is exposed in depositional contact on the Merced Formation, approximately 3.6 miles south of the quadrangle. This exposure, designated a reference locality, is on the east side of the road leading to Thornton Beach State Park, about 0.15 miles south- west of the intersection of Alemany and Skyline Boule- vards and between 200 and 225 feet above sea level. The Colma probably records several different periods of sedimentation related to periods of relatively high sea level. The isolated exposures within the San Fran- cisco North quadrangle can be correlated with each other and with the type Colma on the basis of gross similarity in physical properties and stratigraphic position. The Colma Formation was first recognized by the author and his colleague, M. G. Bonilla, in the Lake Merced area near the Pacific Ocean, about 1 mile south of the quadrangle. There the Colma is a nearly flat lying friable sand that lies unconformably on the steeply tilted fossiliferous marine Merced Formation of Pliocene and early Pleistocene age and unconform- ably below Pleistocene(?) and Holocene dune sand. In the San Francisco North quadrangle, Widespread poorly consolidated bedded sand deposits are corre- lated with the Colma Formation rather than with the Merced Formation of the Lake Merced area because they are approximately horizontally bedded and lie immediately below latest Pleistocene and Holocene deposits. They also lack the shale beds and marine invertebrate megafossils that are common in the Mer- ced Formation. For practical mapping purposes, silt, clay, and poorly sorted rubble intercalated with bedded sand were placed in the Colma Formation. Silt, clay, _ and rubble not associated with bedded sand were gen- erally mapped as slope debris and ravine fill, though they may have been deposited contemporaneously with part of the Colma Formation. MEGAS CO PIC FEATURES Bedding in the Colma Formation is generally easy to discern, for individual beds are 1—3 inches thick. Some of the Colma deposits, however, such as the one at map locality 6 in the Presidio (pl. 2A), are obscurely bedded. Bedding is revealed mostly by variations in the proportion of the silt-clay matrix relative to sand grains and the resulting variations in cohesiveness and cementation. Cohesiveness increases as the proportions of silt and clay increase; however, the silt-clay matrix does not generally exceed 20 percent by weight. The variations in cohesiveness are soon revealed in fresh cuts by rainwash sculpturing. The sandy beds are more readily eroded than the clayey beds, which stand out in relief a fraction of an inch to several inches from the sandy beds. Bedding is accentuated in some places by color differences which probably are caused by varia- tions in the permeability of individual beds and by resulting differences in the amount of iron oxide stain deposited by interstratal solutions. The source of the iron is largely grains of iron-bearing heavy minerals concentrated in thin nonpersistent layers. Bedding may appear as an abrupt change in sand grain size. Rarely, a few pebbles, as much as 1 inch in diameter, are sparsely distributed in a one-pebble-thick layer in the midst of clayey sand. Bedding is rarely shown by clay beds. Thinly bedded clayey sands in the Colma Formation commonly display a slight waviness of bedding with an amplitude of 1—6 inches and a wavelength of %—3 feet (fig. 45). The waviness probably records uneven com- paction rather than the environment of deposition. Crossbedding or inclined torrential bedding inter- calated with near-horizontal bedding is especially com- mon on the south slopes of Twin Peaks (fig. 46). Locally crumpled bedding between parallel nearly hori- zontal undisturbed beds 1—2 feet apart probably records subaqueous slumping. The Colma Formation is generally yellowish or red- dish brown. On Angel Island it is predominantly light gray, as are beds on the west slope of Russian Hill uncovered in the excavation for the Broadway Tunnel. On Twin Peaks the Colma Formation is mostly yellow- ish or reddish brown, but cuts disclose gray unce- rnented sand deposits resembling channel fillings, 15— 25 feet wide. In a few places in San Francisco, the top 3—5 feet of the Colma Formation immediately under- lying dune sand is gray to black, owing to organic car- bonaceous material. Richard J anda (written commun., 1967) believed that the fresh Colma Formation is light gray or light brownish gray. The common yellowish- and reddish-brown colors, he believed, are due to weathering. He also believed soil stratigraphy may be useful in dividing the formation. The Colma Formation is chiefly moderately sorted fine to medium sand with small to moderate amounts of detrital silt and clay (pl. 2B-E). Beds of silty clay 1/2—5 feet thick and coarse rubble containing gravel as large as cobbles are interbedded with the sand in a few places. Locally, rubble beds make up a minor part of the formation. By using the method described by Inman (1952), Inman and Chamberlain (1955, p. 109), and Folk and Ward (1957 ) , 13 samples of the Colma Formation were analyzed; the results are given on plates 2D and 2E. They show the following ranges in characteristics: 68 GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE FIGURE 45.—Colma Formation (QC) overlying greenstone of the Franciscan Formation (KJ g), Sutro Reservoir, San Francisco. The light-colored streak dipping to left is inter- preted as a zone of considerable ground-water chemical activ- ity that altered the Colma Formation to clay and ironstone, possibly an old soil developed on a surface cut on older rocks of the Colma Formation to the right and later buried by younger Colma to left. Photograph below is a closeup view showing bedding and the inclined clay-ironstone zone. FIGURE 46.—Rubb1e deposits and crossbedding in the Colma Formation. In upper photograph contact with greenstone of the Franciscan Formation is 3 feet to left of man. Lower photograph is closeup view of well-bedded sand overlying poorly bedded rubble deposit containing large greenstone pieces. The rubble may be an old landslide deposit; the sand is a stream or lake deposit. South slope Twin Peaks, San Francisco. Median diameter: Md 0.183—0.270 mm Mdgb 2.45—1.89 phi units Sorting: Trask So 1.18—1.85 Phi deviation 17¢ 0.44—1.91 Skewness: Trask Sk 0.625—1.019 Phi aqb + 0.02—0.76 Conclusions on the significance of these characteris- tics are given elsewhere. COLMA FORMATION 69 COMPOSITION AND PHYSICAL PROPERTIES Most of the sand grains of the Colma Formation are (in decreasing order of abundance) rock fragments, quartz, and feldspar (pl. 2F). Rock fragments are generally locally derived. For example, those in the Colma Formation at map locality 2 on Angel Island consist mostly of metamorphic rocks similar to those found in the underlying Franciscan Formation. Sand grains from locality 9 in San Francisco consist mostly of chert and from locality 11A of altered greenstone, nontronitic clay, and some chert. Heavy minerals make up 10—25 percent of the sand grains (pl. 2F). The pre- dominant heavy minerals (in decreasing order of abun- dance) are green hornblende, augite, clinozoisite-epi- dote, brown hornblende, hypersthene, and ilmenite. Locally, dark layers of heavy minerals 1/16—% inch thick are abundant. Particles smaller than 2 microns consist of varying proportions of montmorillonite, chlorite, and mica and random and regular mixed layered clay min- erals of all three. In most sands, black, carbonaceous silt-size material, probably representing plant remains, is present, generally in amounts‘less than 1 percent. In a few exposures in San Francisco, dark-gray to black peat, clayey sand, or sandy clay occurs at the top of the Colma Formation immediately below dune sand. Peaty layers less than 1 foot thick were also penetrated by a few borings in the Colma Formation. Rubble beds were derived locally and consist of whatever rocks were available from nearby Franciscan Formation exposures in Colma time. Most sand grains in the formation are medium size and coarser, are generally subangular to subrounded, and have polished surfaces. About 5—15 percent of the grains are angular; an equal portion are well rounded and have frosted surfaces. Feldspar, heavy minerals, serpentine, and rock fragments—including chert and microporphyritic igneous rocks—are generally better rounded than quartz; however, a small number of quartz grains also are well rounded and have frosted surfaces. Cementation and porosity of the sands of the Colma Formation vary with the clay content. Most sands of the Colma Formation are friable or only weakly cemented. A 25-foot-thick deposit of gray fairly clean sand interbedded with brown clayey sand at Sutro Reservoir is uncemented. Porosity of the Colma For- mation sand is generally moderate to high because of the moderately good sorting, the large proportion of subrounded and rounded grains in the sand sizes, and the lack of enough fine sizes to fill the spaces between sand grains. Some porous sands are cemented only at points of contact of the 1/2—1-mm-thick clay shells around each grain. In a few places the clay matrix com- pletely fills the space between sand grains, and the sediment has low porosity. Sand grains and clay cement are generally iron stained. Oxidation of black layers of iron-bearing heavy minerals produces a tough, low- porosity ironstone 1—3 inches thick. WEATHERING Widespread orange and brown iron stains, with green and gray patches, and sporadic ironstone layers are evidence for oxidation and reduction which, in surficial deposits such as the Colma Formation, are probably related to weathering processes. Most of the clayey material in sand of the Colma Formation appears to be detrital, though a small amount of clay may repre- sent alteration of sand grains. Evidence that the clay is detrital is indicated by the freshness of the feldspar and rock grains and by fairly abrupt variations in clay content from bed to bed. Soil development varies, but it usually appears to be slight to moderate, for distinct soil horizons generally are present only as a 1-foot- thick light- to medium-gray top layer. In a few places the soil-forming process has been intense, such as on the west side of Russian Hill and near Lombard and Divisadero Streets, where clayey sand grades upward into a sandy silty clay layer about 10 feet thick. The bottom half of the sandy silty clay layer is compact, but the top half is vesicular, has columnar structure, resembles loess in appearance, and is evidently the B soil horizon. The A horizon is poorly represented by small 1-foot-thick patches of gray clayey silty sand. Weathering is also shown by etching of pyroxenes and hornblende. Almost all pyroxene grains in the Colma Formation have well-developed, long, delicate sawtooth terminations. In contrast, the pyroxene grains in the modern beaches and dunes have only tiny, short sawtooth terminations or no such terminations. Horn- blende grains in the Colma Formation generally show good cleavage faces, in contrast with grains in modern beach and dune sand which are generally well rounded. Strong etching of pyroxene comparable to that found in the Colma Formation was illustrated by Bradley (1957, p. 434, fig. 10). Though he examined modern beach sands, he found etched pyroxene only in old marine deposits lying on an elevated 100-foot marine terrace at Santa Cruz, Calif. The dark-gray to black, carbonaceous clayey mate- rial found at the top of the Colma, underlying dune sand in a few places in San Francisco, may represent an old soil that developed in or near local marshes or lakes. OCCURRENCE The Colma Formation is widely distributed on Angel Island and in San Francisco. It was not recognized on 70 GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE Tiburon and Marin Peninsulas or on Belvedere Island, although Quaternary deposits at these localities may be of the same age as the Colma Formation. The thick- ness of the Colma Formation varies greatly over short distances and is difficult to estimate. On many slopes its nearly horizontal bedding may be exposed in a con- tinuous section through elevation differences of 100 or 200 feet, yet thicknesses measured at right angles to the bedding or the slope at individual points are con- siderably less. Thicknesses are less variable and gen- erally are greater in broad valleys at low elevations than on the slopes. The Colma Formation is exposed discontinuously on the east and southeast shores of Angel Island (fig. 47) , where exposures range in altitude from sea level to about 325 feet. It may also be represented in the unconsolidated deposits that fill the valley of Hospital Cove. Maximum observed thickness in the quadrangle is 75 feet. FIGURE 47.—Col_ma Formation on the southeast slope of Angel Island, about 250 feet above sea level. In San Francisco the Colma Formation is exposed discontinuously from about sea level to an elevation of about 550 feet, and it is recognized in boreholes more than 100 feet below sea level. The most extensive expo- sures are on the southwest and northeast slopes of the ridge occupied by the Presidio Golf Course and reser- voir, where the deposits are confined to elevations below 280 feet (pl. 1). Sands of the Cohna Formation near Mountain Lake are at least 30 feet thick. In the Presidio north of Kahn Playground, small youthful valleys appear to be cut entirely in the Colma Forma- tion, which is at least 60 feet thick. On the slopes between the Presidio and Russian and Nob Hills, the Colma Formation is found in small sporadic exposures below about 200 feet elevation and in boreholes. The top of a 42-foot-thick clayey sand, believed to be the Colma Formation, was penetrated in a boring at 42.2 feet below mean sea level at Fort Mason near the southeast corner of the pier at the end of Laguna Street. The formation was found to be more than 30 feet thick in the excavation for the twin vehicu- lar tunnels on Broadway near Hyde Street. The Colma Formation is believed to be present on the lower slopes and valleys in eastern San Francisco below an elevation of 100—200 feet, in the area between North Point and the south edge of the quadrangle. At the Ferry Building in a boring below 102 feet of mud and clay, a sand layer, 38 feet thick, is correlated with the Colma Formation. Small patches of the Colma Formation rest on the bluff behind Bakers Beach and on the sea clifls west- ward and southward towards the Cliff House. These are slightly consolidated nearly horizontally bedded sand. Most of these patches are too small to show on the geologic map (pl. 1 ). In the valley north of the Cliff House, the formation is 30 feet thick. Southeast of the Cliff House the Colma Formation overlies sandstone of the Franciscan Formation and is overlain by dune sand. Here, the basal part consists of coarse sandstone rubble derived from the underlying Franciscan Forma- tion. The Colma Formation is at least 40 feet thick; it appears to thicken southward, and it dips gently south- ward toward Golden Gate Park. On the north slope of Mount Sutro, the Colma For- mation is exposed north of Parnassus Avenue. It was penetrated south of Parnassus Avenue in borings made to investigate foundation conditions below the Uni- versity of California Medical Center Building (Moffet Hospital). South of Mount Sutro and west of Twin Peaks, the formation is found at elevations as high as 550 feet. On the south slope of Twin Peaks, the forma- tion fills a channel cut in greenstone of the Franciscan Formation (fig. 48). In the excavation for Sutro reser- COLMA FORMATION 71 FIGURE 48.—Colma Formation filling a channel cut in green- stone. South slope of Twin Peaks, San Francisco. voir, a 75-foot-thick section of the formation was exposed (fig. 45). ORIGIN The Colma Formation appears to have been deposited mostly by water and gravity and, to a lesser extent, by wind in a variety of coastal environments. At eleva- tions above 250 feet, the Colma Formation commonly is massively bedded or steeply crossbedded and is believed to consist mostly of stream, colluvial, and eolian deposits (figs. 46, 48). In contrast, at lower elevations the Colma Formation consists mostly of moderately well sorted sands, usually with persistent horizontal stratification but in some places with inter- calated steeply inclined crossbeds. The lower sediments probably accumulated in an extensive complex of shallow bays, inlets, and channels such as would exist if sea level were 35—100 feet higher than now. Deposi- tion in water is also suggested by local contorted layers, 1-2 feet thick, interbedded with undisturbed beds; the contortions appear to have formed by slumping in water. Intercalations of bedded moderately sorted marine sands with clay and with massive poorly sorted con- tinental sediments indicate fluctuations in the trans- portation and sedimentation processes, in the strength of water currents at the site of deposition, in the loca- tion of the shoreline, and in the character and amount of material supplied to the basins of deposition. The environment of deposition of the Colma Forma- tion can be inferred from the size-distribution diagram (pl. 2D). Most of the data shown are for marine sands believed to have been deposited at the lower elevations. Data for dune sand and modern beach sand are also shown on the diagrams. Dune sands are distinguished from beach sand by their smaller median grain diam- eter and by better sorting. (095 for dune sands is less than 0.5; «4: for beach sands is greater than 0.5.) Sedi- ments of the Colma Formation—whether the whole sediment or only sand fraction is considered—have about the same range in median grain diameter as dune sands; however, they are generally not as well sorted, and their frequency distribution curve is gen- erally skewed towards finer grain sizes. Stewart (1958, p. 2589, 2595) found that marsh sediments of lagoon borders have sorting and skewness values comparable to those of the Colma Formation, though his samples were of finer grain (median phi diameter about 3—6) and the Colma does not contain enough organic matter to be a marsh sediment. Inman and Chamberlain (1955, p. 114, 119) referred to sediments of median phi diam- eter about 3—4, sorting ad) values greater than 1.5, and positive skewness, 01¢, values as transition sand and suggested that such sediments probably result from basic differences between modes of transport of sand and of silt and clay. They found transition sand about 14—1 mile offshore from a beach north of Point La J olla, San Diego County, Calif., and within shallow bays behind barrier beach sand islands near Rockport, Tex. The grain-size distribution of the Colma Formation may be due to deposition and mixing of silt and clay in estuaries to which sand was transported before or during the existence of the estuary. The sites of Colma estuaries, before the rise in sea level that created them, may have been covered by dune sand, by the Merced Formation, or by sand brought in by the ancient Sacra- mento-San Joaquin River. The similarity in ranges of median diameters of the sand fraction of the Colma Fermation and of dune sand suggests that some sand may have reached Colma estuaries by wind. Sand of median diameters like those of the Colma Formation may also have been brought by water currents less vigorous than breakers on a beach. The poorly consolidated Merced Formation is rich in sand grains of sizes found in the Colma Formation (M. G. Bonilla, oral commun., 1961) and is the most likely source of much of the Colma Formation. The Merced Formation was deposited over a wide area west of the present shoreline and as far north as Point Reyes Peninsula and is approximately 5,000 feet thick. Its initial volume appears to have been large enough to have supplied detritus for a large part of the Colma Formation. The folding of the Merced probably brought it above sea level and made it available as a source of the Colma. The predominance of hornblende in the Colma Formation in San Francisco (p1. 2F) is evidence that its main source was not the local Franciscan For- mation, in which hornblende is scarce (table 3). How— 72 GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE ever, the content of hornblende and other minerals in the Merced Formation is similar to that in the Colma Formation and supports the belief that some or most of the Colma Formation was derived from the Merced Formation or from the same source as the Merced For- mation. Where the Colma Formation lies on the Mer- ced Formation, west of Lake Merced, it is difficult to distinguish one formation from the other. Another possible source of the Colma Formation is the detritus carried by the Sacramento—San Joaquin River. Drainage from the Great Valley into the San Francisco Bay area is believed to have started by early Pleistocene time and to have been the source of the upper member (Pleistocene) of the Merced Formation (Hall, 1965, p. 152, 153). Because sea level may have been high when the Colma was deposited, the ancient Sacramento—San Joaquin River may have reached the estuary that existed then far upstream from the site of deposition of the Colma Formation on the San Fran- cisco Peninsula, much as the present river reaches tide- water east of Suisun Bay. Thus the silt and clay sizes in the Colma Formation may have been partly con- tributed by the ancient Sacramento—San Joaquin River drainage from the Great Valley of that time. The sandiness of the Colma Formation may be explained also by its deposition in basins and channels swept by tidal currents. Sands are deposited where tidal currents are vigorous, such as those in the central parts of San Francisco Bay and in parts of Golden Gate channel. Muds are deposited along the fringes of the bay where currents are weak. Muddy sands are deposited under intermediate conditions or where rocky headlands furnish sand that mingles locally with muds (Louderback, 1939, p. 784—793; 1951, p. 91; Inman and Chamberlain, 1955, p. 119). The few frosted and well-rounded sand grains in the Colma Formation suggest wind action on a beach, or they may have been derived from the Merced Formation. The Colma Formation was thought previously to have been deposited in an offshore marine environment (Lawson, 1895, p. 463; Ashley, 1896, p. 354) or in swamps, lagoons, lakes, flood plains, and dune fields (Martin, 1916, p. 226). The author believes the Colma Formation now found below an elevation of 200 feet was deposited in a complex estuary swept by vigorous tidal currents. A eustatic rise of sea level and a local rise of the crust after deposition explain the presence of marine Colma deposits of late Pleistocene age several hundred feet above sea level. Sea level should have been higher than now whenever less glacial ice existed than exists today. During at least one interglacial period, ice was less extensive than it is today (Louderback, 1951, p. 86; Kuenen, 1955, p. 201; Kuenen, 1950, p. 539). Hopkins, MacNeil, and Leopold (1960) and McCulloch, Taylor, and Rubin (1965, p. 446) showed that in the Nome and Kotzebue Sound areas, Alaska, the rise in sea level during Sangamon time was only about 35—40 feet. Hopkins, MacNeil, and Leopold (1960, p. 49) showed that in the Nome area the greatest eustatic rise of sea level during a Pleistocene (preKansan(?)) interglacial period was approximately 100 feet. Thus the marine deposits in the quadrangle presently above an elevation of 100 feet probably reached their present position partly by tectonic uplift. If sea level were 35—100 feet higher than it is today, sandy sediments of the Colma Formation could have accumulated on beaches or in shallow marine water near the shore, as well as in basins several miles off- shore covered by sea water several hundred feet deep. Thus the Colma Formation in the Presidio area, which is now more than 200 feet above sea level, may have been deposited at the same time as the Colma Forma- tion now lying more than 100 feet below sea level at the Ferry Building. Smith (1960, p. 160) suggested that the Colma For- mation records depositional episodes related to several different sea levels. He separated a unit from the Colma Formation which he believed represents beach and dune deposits related to his Colma marine terrace. He traced this marine terrace for more than 15 miles at elevations between 200 and 300 feet above sea level on the bay side of San Francisco Peninsula south of Lake Merced. Above this lower terrace he found terraces 400 feet and 500 feet above sea level. Deposits on the 400- foot terrace resemble the Colma Formation. No deposits were found on the higher terrace. Smith believed the terrace between 200 and 300 feet above sea level was cut during the high sea level of the Sanga- mon Interglaciation. Lawson (1895, p. 463) recognized the beds here called the Colma Formation as his “Terrace Forma- tions.” He described these beds as occurring from an altitude of about 750 feet down to sea level and thought they represented marine deposits. In the San Francisco Folio, Lawson (1914, p. 15) abandoned the term “Ter- race Formations” but mentioned a deposit in the Lake Merced area of “light-yellow sands, about 200 feet thick, which probably lies unconformably upon the Merced***” and suggested that it may be the correla— tive of the Alameda Formation, a Pleistocene marine and continental formation found on the east shore of and in San Francisco Bay. He explained that he did not show the deposit in the Lake Merced area on a geologic map because it could not be easily distin- guished from the underlying Merced Formation. Marine megafossils have not been found in the Colma Formation at lower elevations where the formation is COLMA FORMATION 73 believed to have been deposited in a marine environ- ment. Perhaps strong currents and a sandy bottom were unfavorable for marine life, as they are today. Evidence for the presence of leaching acidic solutions is found in decomposed sand grains bearing iron such as magnetite, which occur almost everywhere in the Colma Formation. Such acidic solutions could decom- pose marine shells, which are largely calcium carbon- ate, more readily and more completely than they could decompose magnetite and other iron oxides. Smith ( 1960, p. 153) also suggested leaching by ground water as an explanation for the lack of fossil shells in marine sediments of Colma age on former wave-cut platforms 2—25 miles south of the quadrangle. CORRELATION WITH NEARBY DEPOSITS Sand lying below bay mud along the bay shore east of San Francisco is included in the Colma Formation, although it could not be traced continuously from the Colma Formation on the Pacific Ocean and Golden Gate channel sides of San Francisco Peninsula. On their geologic cross section that extends from the bay onto San Francisco near the south border of the quad- rangle, Trask and Rolston (1951, p. 1085) designated the sand below bay mud as the Merritt Sand. It was designated the “sand layer” in the Islais Creek area at the southeast corner of the San Francisco North quad- rangle by Radbruch and Schlocker (1958); they cor- related the upper part of the layer with the Merritt Sand and Temescal Formation. On the east side of San Francisco Bay, Radbruch (1957) showed that the marine type Merritt Sand grades into and interfingers eastward with alluvial-fan deposits of the Temescal Formation, which she found to be the same unit as the upper member of the San Antonio Formation. Similar facies are found on the west shore of the bay in San Francisco where the marine Colma Formation of lower elevations—here correlated with the Merritt Sand— interfingers with and grades into the continental Colma Formation of mostly higher elevations. Pleistocene marine deposits are also above sea level on the east side of San Francisco Bay (Louderback, 1951, p. 86). Similar deposits were not seen on the Marin Peninsula, though small patches of them were seen in the vicinity of Point Bonita west of the quad- rangle. They are lacking from Marin Peninsula because such unconsolidated deposits could easily have been removed by erosion of the steep slopes when sea level dropped. The same cause may explain the absence of the Colma Formation on the steep slopes on the north side of Angel Island. Some of the deposits at elevations of 200 feet and lower on Yerba Buena Island, desig- nated “reworked colluvium” by Radbruch ( 1957), resemble the Colma Formation and may have a marine origin. Abraded marine and nonmarine diatoms and sponge spicules are present, but no specific identifications have been made. Part or all of the diatoms could have been derived from older sediments. Vertebrate remains are reported from what may be the Colma Formation. Hay (1927, p. 5, 210) reported that a bone now lost, appar- ently a fragment of a tibia of a large ground sloth, was found before 1853 at a depth of 23 feet on Pacific Street between Kearney and Montgomery Streets. Stock (1925, p. 201—202) reported that a humerus of Mylodon (probably Paramylodon, a ground sloth) was found in excavations for the Twin Peaks Tunnel at a depth of 60 feet, about 300 feet west of 18th Street. Hay (1927, p. 210) concluded that the material from the two locali- ties comes from Lawson’s San Antonio Formation and represents the Aftonian Interglaciation. Peabody (1945, p. 60—63) described a vertebrate fauna of late Pleistocene age from a deposit that he believed to be slightly older than Lawson’s “Terrace Formations” (the Colma Formation of Schlocker and others, 1958, and of this report). The bones were found above Mussel Rock on the ocean shore 6 miles south of the quadrangle. Peabody ( 1945, p. 60) reported that the bones are in a stream deposit of gray sand and gravel resting on the Merced Formation and overlain by Lawson’s “Terrace Formations” of reddish-brown horizontally bedded sand. Field mapping in the Mussel Rock area by M. G. Bonilla (oral commun., 1958) indi- cates an obscure relationship of the fossiliferous sand and gravel and the overlying reddish-brown sand to the Colma Formation. Savage (1951) listed other vertebrates from the San Francisco Bay region to which he assigned an “undifferentiated Pleistocene” age. They were found in deposits that may be the same age as the Colma Formation. A tree identified by Rowland W. Brown as a juniper or redcedar, probably J uniperus californica, was found in the excavation for the Broadway Tunnel on the west slope of Russian Hill near Hyde Street in a sand thought to be the Colma Formation. Duplicate carbon- 14 age determinations by J. L. Kulp in 1954 gave an age greater than 30,000 years (sample designation: Lamont No. 227). The Colma Formation is tentatively assigned a late Pleistocene age. AGE RELATION TO THE ANCIENT SACRAMENTO RIVER The shape of the exposed and buried parts of the bedrock surface shown on the bedrock-surface map (pl. 3) is strikingly similar to the pattern of valleys and ridges made by stream erosion. Most of the valleys 74 GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE are tributaries of the large valley, now Golden Gate channel, that was carved between San Francisco and Marin Peninsulas by the ancient Sacramento River which emptied into the Pacific Ocean at various dis- tances west of Point Lobos. The ancient channel of the Sacramento River and some of its tributaries is shown on maps by Trask (1956, p. 17). The bedrock- surface map of central San Francisco Bay by Carlson and McCulloch ( 1970) shows channels in bedrock that may represent old stream channels. Louderback (1951, p. 82) suggested that Golden Gate channel was carved perhaps in early Pleistocene time or perhaps during a middle Pleistocene disturbance inasmuch as it main- tained its flow across the slowly rising land west of the Great Valley. Thus, some of the sedimentary deposits mapped as Colma Formation may have been trans- ported by the ancient Sacramento River before the crustal rise or at an early stage of the rise. Sea level is estimated to have been about 228—406 feet below present sea level during the Wisconsin Gla- ciation and 325—390 feet below present sea level dur- ing a pre-Wisconsin glaciation (Flint, 1947, p. 437; Moore and Shumway, 1959, p. 373; Curray, 1965, p. 725; Milliman and Emery, 1968). The present shore- line would shift 34—39 miles westward if sea level were lowered to the present 300-foot-depth contour (US. Coast and Geodetic Survey, 1941), and this lower base level would result in Vigorous erosion of the present bottom. The present-day buried valleys on San Fran- cisco and Marin Peninsulas must have been carved when sea level was lower, during a maximum glacial stage, or earlier during a middle Pleistocene disturb- ance, as postulated by Louderback (1951, p. 82). The Colma Formation and its age equivalent, the Merritt Sand, are the upper part of the natural fill in these buried valleys. For example, the bedrock floor of the buried valley below the Ferry Building is more than 270 feet below sea level; however, the bottom of the Colma Formation at this place is 127 feet higher. Thus, at the Ferry Building the Colma Formation was deposited after the valley was carved, and it subse- quently was filled with 127 feet of sediments older than the Colma. It is also likely that several periods of ero- sion and filling during successive glacial and inter- glacial stages intervened between the time of the carv- ing of the valley and the deposition of the Colma Formation. After its deposition, the Colma Formation was also cut by valleys graded to a sea level lower than the present level. RELATED DEPOSITS Some Quaternary deposits around San Pablo, Sui- sun, Tomales, and Monterey Bays (fig. 1) are tenta- tively correlated with the Colma Formation. The Millerton Formation of late Pleistocene age at Tomales Bay consists of interbedded marine and continental beds (Weaver, 1949a, p. 99—103; Weaver, 1949b, p. 51—52). The marine beds contain a large molluscan fauna, and the continental beds contain a large flora. Weaver (.1949b, p. 52) believed the fossils “indicate a climate slightly warmer than that which prevails today, possibly representing an interglacial epoch.” Richards and Thurber (1966, p. 1092) deter- mined an apparent age of 55,000 years by the Thm/ U234 method on mollusks from the Millerton Formation and suggested that the mollusks “could easily have been deposited during the last interglacial stage or earlier.” The Millerton Formation contains seven ver- tebrate genera, including Bison (Weaver, 1949a, p. 103), and is therefore of Rancholabrean mammalian age (Savage, 1951, p. 289), which spans the Illinoian Glaciation, the Sangamon Interglaciation, and the Wis- consin Glaciation (Hibbard and others, 1965, p. 514). Fossiliferous marine sand and associated sandy clay and gravel, assigned to the Millerton Formation by Weaver (1949a, p. 101), are found in nearly horizontal beds in San Pablo and Suisun Bays. Weaver (1949a, p. 103—106) also named several nearby continental deposits, the Huichica, Glen Ellen, and Montezuma Formations, which are partly contemporaneous with the Millerton Formation. The Aromas Red Sands of Allen (1946, p. 18, 43—45) , exposed over a wide area near Monterey Bay, are simi- lar in field appearance and lithology to the sands of the Colma Formation. Allen believed “They were laid down by the action of both wind and waves, on a low- lying plain, as lagoonal deposits, sand dunes, and bars. After uplift, the oxidation and solution of the magne- tite in the sand resulted in the development of red colors, cementation, and reprecipitation of hematite.” Although no fossils have been found, Allen (1946, p. 45) suggested that the Aromas is “at least as late as middle Pleistocene in age.” OLDER BEACH DEPOSITS In May 1952 a fresh-appearing beach sand about 100 feet above sea level was exposed below a 5—10-foot cover of dune sand in a deep trench on the Presidio Military Reservation. The sand is pale yellowish brown (IOYR 6/2), loose, and fine to medium. It is believed to be a beach deposit because of the nearly horizontal bedding and the persistence of thickness and inclina- tion of the beds for several hundred feet (fig. 49). These features are in marked contrast to corresponding features of dune bedding (Thompson, 1937, p. 747— 751; McKee, 1953, p. 20, 25). In many places the bed- ding resembles Thompson’s (1937, p. 732) type A OLDER BEACH DEPOSITS 75 FIGURE 49.—Raised older beach deposit covered by dune sand. Southwest corner of the Presidio, San Francisco. Bakers Beach is one-fourth of a mile to the left. A, The raised beach deposit below dune deposits as exposed in north wall of sewer trench. Elevation approximately 150 feet above sea level. B, Closeup view of the raised beach deposit showing regu- larity of bedding. C, Closeup view of dune sand covering raised beach deposit. Note crossbedding in dune sand. cross-lamination found on the upper foreshore of modern California beaches, where laminae of low land- ward dip are truncated by an erosional surface of low seaward dip. In the Presidio the dip of the bedding of the raised beach deposit increases to as much as 20° eastward, suggesting a facies change that may repre- sent backshore beach deposits (McKee, 1953, p. 14, 19). These sands, in turn, grade into sands ‘showing classic dune crossbedding. Grain-size distribution of sand from the raised beach is similar to modern beach sands and modern dune sands (pl. ZC—E). Most of the deposit has a median diameter near the lower limit of the medium sand size, but sands of maximum eastward dip (pl. 2A, map local- ity 19, sample 1434) have a median diameter near the lower limit of the coarse size. Sorting is good, except for the coarse sand, which is only moderately sorted. The shape, roundness, and frosting of grains of the raised beach sand are more similar to those of dune sand than to those of modern beach sand; mineral composition is similar to that of both modern beach sand and dune sand. Intensity of etching of pyroxenes is between that shown by dune sand and that shown by modern beach sand. About 1—2 percent of the pyroxene grains show long sawtooth terminations with blunted points. Most of the pyroxene grains show only tiny sawtooth terminations or show none. AGE The older beach deposit, like the Colma Formation, was probably deposited during an interglaciation when sea level was higher. Properties of the grains and their size distribution suggest that the raised beach sand consists mostly of dune sand blown from a beach west of it and below it and that the dune sand entered the beach environment during a high stand of the sea. The raised beach sand may be younger than the Colma Formation, for it appears to be completely unaltered and unconsolidated and contains tiny shell fragments, whereas sands of the Colma Formation are generally weathered to a darker brown color, have a fair cohesion, and have no shell fragments. The contrast in etching of pyroxene grains is also striking, for the strong etching seen in almost all pyroxene in the Colma Formation is rare in pyroxene of the raised beach sand. On the other hand, the raised beach sand may be a relatively unweathered part of the Colma Formation, the weath- ered part having been removed by wind action. The stratigraphic relations of the raised older beach sand and the Colma Formation were not directly ob- served and remain obscure. The Colma is exposed at the same elevation 2,500 feet east of the raised beach sand and also near sea level 1,000 feet west of it. If the 76 GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE raised beach deposit is younger than the Colma Forma- tion, why is it not more extensive and why is it not found elsewhere above the Colma? Perhaps some deposits thought to be dune sands, penetrated in bore- holes, are part of the raised beach deposit. Because it is unaltered, the raised beach sand appears to be younger than marine‘ sands on the Half Moon Bay marine terrace, described by Smith (1960, p. 46). If the raised beach sand is younger and if Smith’s (1960, p. 186) Sangamon age designation for the Half Moon Bay terrace is correct, the raised beach sand is Wisconsin in age. Sea level during Wisconsin time, however, was lower than present sea level (Hop- kins, 1959, p. 7). Consequently, if the raised beach sand is indeed Wisconsin in age, it reached its present elevation by uplift rather than by deposition during a sea level higher than present sea level. On the other hand, the raised beach sand may have been deposited during Sangamon time, which was the latest time when sea level was substantially above present sea level. If the sand is Sangamon in age, then the Colma Forma- tion and possibly the Half Moon Bay terrace are related to earlier interglaciation. ' MODERN BEACH DEPOSITS Sands and coarser beach deposits are numerous along all the shores except the east edges of San Fran- cisco and Sausalito. Most beaches are small and are confined to coves and inlets. The largest, Ocean Beach, extends for nearly 8 miles south of the Clifi House (figs. 50, 51). The next largest, Bakers Beach, is more than half a mile long, west of the Presidio. Beach deposits that at one time bordered nearly all the shore between Fort Point and Telegraph Hill are now largely buried under artificial fill. Beach deposits vary in thickness and areal extent, depending on the nature of the waves and the supply of sediments, both of which vary from hour to hour and season to season. An excess of sediments permits beach building, and an inadequate supply of sediments for some wave conditions causes beach erosion. Energetic storm waves may cause considerable erosion in a few hours (Trask, 1959, p. 21). Deposits on the smaller beaches probably are not more than 20 feet thick, but those along Ocean Beach and Bakers Beach may be considerably thicker. Beaches of predominantly coarse gravel and cobbles with some sand and fine gravel are found in a few coves on both sides of the Golden Gate and on Tiburon Peninsula and Belvedere and Angel Islands. The prin- cipal materials of gravel beaches on the north side of the Golden Gate channel are radiolarian chert and greenstone. Some beach deposits consist of boulders FIGURE 50.——Ocean Beach, viewed southeastward from the Cliff House area. The seawall serves as a barrier to the eastward movement of windblown sand. FIGURE 51.——Ocean Beach looking northward from a point about 1 mile south of the San Francisco North quadrangle. Note slight development of berm. Photograph taken in February 1962. of intact greenstone pillows that have become detached and fallen from exposures above the shore. Rocks 5—10 feet in diameter are common on the beach between Fort Point Rock and Bakers Beach and west of Lands End where debris of sheared rocks or serpentine has fallen to the shore. Locally abundant are concrete pieces derived from construction waste dumped into the sea. Trask (1959) examined Ocean Beach at five stations (pl. 2A) at 2—6-week intervals from July 1956 to June 1957 and found many changes during that time. Beaches generally were built upward and seaward dur- ing summer and fall and were eroded during winter and spring. Sand, the predominant beach deposit, is yellowish brown (10YR 6/4) to light gray (N7) speckled with a few white shell fragments and abundant dark-gray, green, and brown grains. The median diameter of sand MODERN BEACH DEPOSITS 77 grains is in the medium to coarse size ranges. Sorting is moderate as measured by the phi deviation measure, 0(1), which ranges from 0.60 to 0.95 (pl. 2C—E). The Trask sorting coefficient ranges from 1.34 to 1.55. The phi skewness measure, 01¢, shows that most of the analyzed sands have a size-frequency curve skewed slightly to moderately on the larger grain size side of the mode. Trask (1959) gave median diameters and Trask coefficients of sorting for about 500 sand samples col- lected along Ocean Beach on various parts of the beaches at his stations (pl. 2A) during 1 year. The average median diameters ranged from 0.191 mm (fine) to 0.497 mm (medium). These extremes were obtained from sands collected on one beach (station J) in Sep- tember (fine) and April (medium). The median diam- eters were generally larger in winter and spring than in summer and fall. Average median diameter for the year at all the stations were in the medium sand size range. At any given time of sampling, grain size did not vary significantly at different parts of the beach. The sorting values suggest a progressive increase in sorting to the south along Ocean Beach. Sorting was best in summer and fall. Sand grains are mostly quartz, feldspar, and rock fragments. The rock fragments from the sample taken at map locality 15 (p1. 2A) consist mostly of chert, quartzite, porphyries, felsite, and some serpentine. Heavy minerals, chiefly ilmenite, chromite, green and brown hornblende, augite, and hypersthene, make up about one-fifth to one-third of the sand. Black sands— natural concentrates of heavy minerals—are found from time to time on Ocean Beach about 1 mile south of the quadrangle. More than half of the dark grains have a secific gravity below 2.82; these include chert, quartzite, porphyry, felsite, and serpentine. A small amount of gold has been recovered from these sands (DeGroot, 1890, p. 545—547; Day and Richards, 1905, p. 1188; Davis, 1949, p. 107). Modern mollusk and echinoid shells are abundant locally on the beach sur- face, but they are generally rare in the sand below the surface. Flat rounded cobbles of fossiliferous sand- stone of the Merced Formation are occasionally found on Ocean Beach. Sand along the north shore of the Golden Gate are mostly dark brown and consist mostly of radiolarian chert and greenstone grains. Most of the quartz and feldspar grains are slightly elongated, are subangular to subround, and have pol- ished surfaces. A small percentage are angular or well rounded and have dull pitted surfaces. Some grains show these features on only part of the grain. The dark grains, Whether heavy or light, are generally better rounded than quartz and feldspar grains; a large per- centage of the dark grains are well rounded and have polished surfaces. Many of the heavy mineral grains, chiefly hornblende, pyroxene, and zircon, are rod shaped. Pyroxene grains generally have tiny or blunted sawtooth terminations, a few have long, delicate saw- tooth terminations, and a few have no such termina- tions. SOURCE OF SAND ON OCEAN BEACH The chief sources of sand on Ocean Beach and in the related dunes that extend 6 miles eastward are the poorly consolidated Merced Formation along the shore south of this quadrangle and the Colma Formation and younger Quaternary rocks. These sources are exposed for several miles along the shore to the south in steep cliffs that reach heights of 500 feet. Active landsliding of these cliffs furnishes large volumes of sediment to the shore at their feet (Bonilla, 1959, p. 29; 1960, land- slide location map). Direct evidence for the derivation of beach sand from these older formations is furnished by the sand grains. The similarity in mineral composition of sand of Ocean Beach and sand of the Colma Formation (pl. 2F) and Merced Formation suggests these formations are sources of the beach sand. Hornblende is abundant in the beach sands and indicates that the local hornblende- poor sandstone of the Franciscan Formation is not an important source. Hornblende, however, as well as other minerals, may have reached Ocean Beach from other sources, such as the quartz diorite of Montara Moun- tain 9 miles south of San Francisco. Bradley (1957, p. 434) indicated that pyroxene grains are not etched in the active beach environment. The abundance of etched pyroxene grains, some strongly etched, on Ocean Beach indicates a derivation from older sand deposits such as the Colma and Merced Formation, in which almost all pyroxenes are strongly etched. In the past, beach sand may also have been derived from the vast Sacramento—San Joaquin drainage area. During Quaternary glaciations sea level is thought to have been 228—406 feet lower than now (Moore and Shumway, 1959, p. 373; Hopkins, 1959, p. 7; Kuenen, 1955, p. 197; Kuenen, 1950, p. 537; Flint, 1947, p. 437). The shoreline during these times would have been about 30 miles west of Ocean Beach. Thus, the Sacra- mento River would have carried sediments into the sea to form a delta many miles west of the Golden Gate. Sand from the river would probably have furnished much material for beaches near its mouth and for dunes east of the shore. With rising sea level, the locus of deltaic, beach, and dune deposition would. have shifted eastward. The delta then would have retreated to its present location east of Carquinez Strait as the ocean covered the lower course of the river. The wide sandy sea-bottom plain between the present shoreline and the 78 GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE Farallon Islands is probably a relict of these former deltas, beaches, and dunes. Some or much of the sand on Ocean Beach, Bakers Beach, and the beaches between Fort Point and Black Point and the sand in the offshore bar across the mouth of the Golden Gate may, thus, be derived from land far to the east. Although direct measurements of sand drift have not been made, some northward movement parallel to the shore probably has occurred for several reasons. Waves frequently strike Ocean Beach at an angle from the south (Trask, 1959, p. 24). Richard Janda (written commun., 1967) also suggested that a north-moving current may be related to strong prevailing winds from the southwest and west and the relatively straight north-south trend of the coast. The US. Coast and Geodetic Survey (1943, p. 32—33; Gilbert, 1917, p. 69) reported a weak offshore current that flows south- southeastward, following the trend of the coast, but near the shore the Survey reported a weak current that flows northward, except when winds blow southward. The north-flowing inshore current in the San Francisco area is explained in part as an eddy, formed in the lee of the Point Reyes salient of the coastline, counter to the general southward circulation of the California part of the eastern Pacific Ocean (Davis, 1933, p. 19—20; Howard, 1951, p. 105). Moore (1965, p. 55—56) concluded from a study of heavy minerals in recent coastal sediments west of Marin and San Francisco Counties that (1) longshore transport is only of local importance at the present time; (2) sand in more than 90—120 feet of water originates from an earlier period of sedimentation than that of the modern coast; (3) the bay bar sand is derived from San Francisco Bay; (4) the beaches south of the Golden Gate owe their composition at least partly to offshore sands mixed with local sediments by the surf; and (5) longshore movement south of the bay bar is to the south. Yancey and Lee (1972) placed the modern beach sediments of the west and north coast of San Fran- cisco in their hornblende-augite-hypersthene heavy- mineral assemblage, which they believe results from mixing of sediments derived from volcanic, meta- morphic, and sedimentary rocks of the Central Valley of California. DUNE SAND Dune sand underlies more than half of the city of San Francisco, but the only other occurrence in the quad- rangle is a small deposit behind a small sandy beach on the east shore of Angel Island south of Quarry Point. Prevailing westerly winds swept the sand from Ocean Beach, Bakers Beach, and possibly from a former beach west of Black Point as far eastward as the former Yerba Buena Cove, which was between Rincon and Telegraph Hills and has been artificially filled. Winds from the east created the small dune on Angel Island. Dune sand is more than 600 feet above sea level at Sunset Heights and Grand View Parks and covers the 575-foot summit of Buena Vista Park. On Nob Hill, near its east border, dune sand is nearly 300 feet above sea level. Dune sand in the Civic Center area (fig. 52) and surrounding parts of Hayes Valley reaches thicknesses greater than 100 feet. Dune sand, 5—40 feet thick, covers the west slope of Mount Sutro north of Kirkham Street. Maximum thickness of dune sand is approxi- mately 150 feet. Elevation differences of 75 feet were measured from dune crests to interdune troughs in the area now occupied by the Mark Twain School and A. P. Giannini Junior High School on Ortega Street prior to the development of that area in 1952, and the total thickness of sand was greater than 100 feet. Dune sand mantling the prominent hills and ridges is much thicker on the east (lee) side than on the west and is thin or absent on the crests. Dune sand is absent from the west slopes of Mount Sutro on the lee side of the Grand View—Sunset Heights Parks ridge. The extensive area of hills and lowlands east of the Mount Sutro—Mount Olympus—Twin Peaks highland was also protected from windblown sand. At the present time, manmade barriers confine most wind transportation of sand to a narrow zone near Ocean Beach. Active dunes were observed by the author in 1937 in the Sunset District west of 33d Ave- nue and as late as 1951 in the area west of Sunset Boulevard and south of Ortega Street (figs. 53, 54). They were of the transverse-ridge type (Cooper, 1958, p. 27—49; 1967, p. 42—52). The US. Coast Survey map FIGURE 52.—Dune sand exposed in excavation for Brooks Hall. Civic Center, San Francisco. DUNE SAND FIGURE 53.—Sand dunes in Sunset District of San Francisco. Notch in skyline in center of photograph is site of San An- dreas fault in Marin County. View northwest toward Bolinas Bay. FIGURE 54.—Unstable embankment of dune sand on the lee side of Sunset Heights, near Moraga Street and Funston Avenue, San Francisco, left by excavation for pad for house. Sand was blown over top of bill, which is 666 feet above sea level. View west. of San Francisco published in 1853 (fig. 62) shows ridges that probably are longitudinal dunes in the pres- ent Civic Center—Market Street area. The dunes near Aquatic Park are shown in a photograph taken about 1878 (fig. 55). Dune sand is fine to medium and well sorted (pl. 2C—F). Median diameter decreases sharply from beach sands to adjoining dune sands. A decrease in median grain diameter and an increase in sorting were expected from west to east across the peninsula; however, no consistent variation was seen in the dune sand samples. Dune sand in the central part of the peninsula generally shows positive phi skewness, whereas samples closer to the Pacific Ocean beaches and samples closer to the 79 FIGURE 55.—Sand dune in Aquatic Park area near intersection of Columbus Avenue and Beach Street, San Francisco. Photograph taken in 1878 or earlier looking northwest to Golden Gate from Telegraph Hill. (Photograph from Pro- fessor William S. Cooper, Boulder, Colo.) San Francisco Bay show negative phi skewness or zero phi skewness. Because variations in parameters are expected in different parts of a single dune, the incon- sistency may be due to a failure to select samples from the same parts of dunes, though a negative phi skew- ness of the western dune sand may be an inheritance from beach sands. It is also possible that samples were taken from different generations of dunes; these may have formed under slightly different conditions of wind and source of sand. Dune sands show considerable local variations in shape, roundness, surface texture, and mineralogy. As in the beach sands, most grains are generally equant in shape, though such heavy minerals as hornblende, pyroxene, and zircon are rod shaped. Most grains are subround or subangular. Polished surfaces are more common than dull pitted surfaces. In all dune deposits, however, some well-rounded grains with dull pitted surfaces are present; in a few deposits they are predom- inant. In general, the larger grains are more rounded than the smaller grains, and dark grains are more rounded than light-colored grains of the same size. The dark grains have greater roundness because they are generally of greater specific gravity or are more easily scratched than the light-colored grains (Pettijohn, 1949, p. 428). Textures of rocks that are now grains in the dune sand may have some influence on rounding, for dark-red-brown fine chert grains in the dune sand are generally more rounded than grains of single quartz crystals of the same size and specific gravity. Magne- tite, ilmenite, chromite, sphene, monazite, and apatite 80 GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE are especially well rounded, even in sizes smaller than 100 mesh (0.149 mm in diameter), whereas in the same deposit quartz and feldspar grains of the same size are generally angular or subangular. Roundness of quartz grains between 50 and 70 mesh (0295—0206 mm in diameter) was determined by O. M. Schmidt, using Power’s (1953) visual method for beach sands (pl. 2A, E, samples 16888, 1690) and dune sands (samples 1293, 811, 1483, 1253, 2001, 1249). Roundness increases sharply between beach sands and the nearest dune sands and increases slightly and grad- ually from west to east in the dune sands. The sharp increase in roundness between beach sand and dune sand was also observed in other areas by Beal and Shepard (1956) , who suggested that the wind tends to select the more rounded grains for transportation. Pettijohn (1949, p. 422) also suggested that shape sorting may be more important than abrasion in caus- ing a greater roundness of dune sands compared with sands from which they were derived. Pyroxene grains with sawtooth terminations are common in the dune sands near Ocean Beach. Accord- ing to R. C. Gervacio, graduate student at Stanford University (oral commun., 1957 ) , such pyroxene grains progressively decrease in abundance eastward. Bradley (1957, p. 434) found a progressive increase of etching of pyroxene grains with increasing time of subaerial weathering in a marine terrace deposit at Santa Cruz, Calif. He indicated that etching of pyrox- ene does not take place on active modern beaches. Mineral composition of sand grains in two dunes from the quadrangle and from a dune at Edgemar, 8.7 miles south of Point Lobos, is given on plate 2F. Feld- spar and rock fragments, chiefly chert and quartz in varying proportions, make up 70—85 percent of the dune sands. The remaining grains are heavy minerals, in approximate order of abundance, hornblende, clino- zoisite-epidote, ilmenite, magnetite, augite, hyper- sthene, and chromite. Hornblende, chromite, and ilmen- ite are the most common heavy minerals in the Edge- mar dune sand. A very slight iron staining of the sur- face of quartz and feldspar grains and the abundance of dark grains give dune sand its prevailing color of moderate yellowish brown, though active dunes com- monly are light gray. ORIGIN AND AGE The extensive dunes in San Francisco are due to a favorable combination of wind, supply of sand, coastal topography, and limited plant cover. Strong eastward onshore winds are frequent along the west shore of San Francisco Peninsula. The greatest source of dune sand, Ocean Beach, is continuously replenished by longshore currents and by the surf carrying sand shoreward. Bakers Beach, too, is probably supplied by sand from the surf and tidal currents. Some of the sand for Bakers Beach and the beach east of Fort Point may come from the large area of sand waves on the bottom of the bay between Angel Island and San Francisco, formed by the action of tidal currents moving from the bay to the ocean (pl. 1). The topography east of Ocean Beach and Bakers Beach is either nearly level, such as at Golden Gate Park, or rises gently. The level topography and the present—and possibly past—absence of a forest of large trees offer no marked obstruction to wind and sand movement within at least 2 miles of the shore. Dune sand above the high cliffs between Point Lobos and Bakers Beach was probably blown by occasional east winds from dune sand lying on the gentle slopes in back of the beaches. Two sections of dune sand separated by bay mud and clay are penetrated in borings in the Market Street area, east of the Civic Center. This indicates a long history of dune activity. Intensive development of dunes in San Francisco took place more than once dur- ing the Holocene and perhaps in the Pleistocene (Cooper, 1967, p. 49—50). However, belts of progres- sively older dunes away from the beach such as those in the Santa Maria coast area, 230 miles south of San Francisco, were not seen in San Francisco (Woodring and Bramlette, 1950, p. 116; Cooper, 1958, p. 136). Numerous accounts of San Francisco during and fol- lowing the gold rush of 1849 illustrate and discuss dunes between Nob and Rincon Hills and in the west- ern part of San Francisco near the present Golden Gate Park (Brewer, 1930, p. 499; Weinstein, 1967, p. 38—39). Cooper (1958, p. 130; 1967, p. 113—115) suggested that during periods of rising sea level, dune develop- ment is especially active. He believed that the coastal dunes of California, Oregon, and Washington are related to major sea level changes—a rise in sea level, some 5,000—6,000 years ago, and a succeeding period of vir- tually stable sea level. He also suggested that sea level changes from earlier Pleistocene glaciations would have influenced dune activity. Thus, it is probable that dunes of San Francisco are not all of the same age. For example, some of the dunes on the east side and central part of the peninsula may be older than dunes closer to the ocean on the west side of the peninsula. SLOPE DEBRIS AND RAVINE FILL Most slopes in the quadrangle are mantled by uncon- solidated debris derived from weathered underlying material and transported by mass-movement processes. These deposits have been mapped wherever they are known to be more than about 5 feet thick. Locally, this unit includes soils developed on bedrock as well as SLOPE DEBRIS AND RAVINE FILL 81 minor amounts of alluvial, eolian, and landslide mate- rials. Slope debris thickens progressively downhill to an observed maximum of 18 feet; locally, it probably attains thicknesses three or four times as great. Most of the ravines are partly filled with material derived from the adjoining slopes. These deposits vary in thickness from a few feet to more than 30 feet. Locally, especially on Marin Peninsula, they are inter- bedded with and grade into stream alluvium. In many places throughout the quadrangle, roadcuts in gently rounded, mature slopes have revealed deep, steep- sided ravines that have been completely buried by slope debris (fig. 10). Slope debris and ravine fill consist of a mixture of angular rock fragments in a matrix of sand, silt, and clay (pl. 23). The physical properties of the slope debris are partly dependent upon the underlying bed- rock. On many slopes in northern San Francisco, where the bedrock is predominantly sandstone and shale of the Franciscan Formation, most of the material is in the sand-to-clay size range with minor amounts of gravel-size material. Elsewhere in San Francisco and on Marin Peninsula, where radiolarian chert is a com- mon bedrock and where slopes are generally steeper and transportation of surface materials more vigorous, slope debris and ravine fill contain a large percentage of gravel- and cobble-size radiolarian chert pieces and appreciable quantities of greenstone and sandstone (fig. 56). Slope debris on or below greenstone or sand- stone and shale bedrock is generally much thicker than that derived mostly from radiolarian chert and shale bedrock. Slope debris may include deposits of different ages. Some roadcuts in slope debris on Marin Peninsula FIGURE 56.—Slope debris on weathered greenstone of the Fran- ciscan Formation. Gravel pieces are chert derived from up- slope. Note slight soil development at top. 534-039 0 - 74 - 6 show unconformable layers of distinct lithologies. In many areas slope debris is apparently the youngest material of the slope. On some of the lower slopes and along the wide shelving shores, however, the ravine fill is overlain by marine deposits and is probably late Pleistocene in age. Material identical in lithology with the modern slope debris has also been observed below and interbedded with the Colma Formation. Material in the buried ravines is somewhat older than modern slope debris. The buried ravines were evidently formed during a period of vigorous erosion, such as would occur at the time of a lower Pleistocene sea level when, also, the intensity and amount of precipitation was probably greater than now. The ravines were probably then filled during the early part of the succeeding interglacial period. Transportation of material downslope and into ravines was mostly by such colluvial processes as creep, mud flowage, and debris flowage. SAN FRANCISCO BAY SEDINIENTS San Francisco Bay and the Golden Gate channel, which cover more than half the quadrangle, are under- lain by modern sediments, though in places these are only a few inches thick. Bay sediments also underlie areas of artificial fill along the fringes of the bay. Sedi- ments and sedimentation in San Francisco Bay are currently (1969) under investigation by the US. Ge- ological Survey. This report summarizes results of other investigations and results of the author’s examination of San Francisco Bay sediments. Sediment studies in San Francisco Bay have been reported by Sumner, Louderback, Schmitt, and J ohns- ton (1914), Gilbert (1917), Packard (1918), Bailey (1921), Miller, Ramage, and Lazier (1928), Louder- back (1939; 1951,p. 88—93), Trask and Rolston (1951), Trask (1956), Arden (1961), George Porterfield, N. L. Hawley, and C. A. Dunham (written commun., 1961), Einstein and Krone (1961), Krone (1962; 1963), Arnal and Conomos (1963), Kvenvolden (1962), US. Army Corps of Engineers (1963), Smith (1963; 1966), Conomos (1963), Treasher (1963), Mitchell (1963), Storrs, Selleck, and Pearson (1965), Slater (1965), Reese (1965), Means (1965), Gram (1966), Selleck, Pearson, Glenne, and Storrs (1966), Meade (1967), and Quinterno (1968). Data in this paragraph on sediment volume and movement is taken from Smith (1966, p. 10) and Sel- leck, Pearson, Glenne, and Storrs (1966, p. 34). The annual sediment inflow to the bay system, including the delta area east of Suisun Bay, is estimated to be between 6.2 and 8 million cubic yards. The combined San J oaquin—Sacramento Rivers supply about 85 per- cent by weight of sediments that enter the bay; much of the remainder is carried into the bay by Napa River 82 GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE and Alameda Creek. About 1.8 million cubic yards of sediment is deposited each year in the delta area. More than half the total sediment volume enters the bay during February, March, and April. Ebb tides tend to move sediments towards the Golden Gate; approxi- mately 30 percent of the sediments entering the bay is swept out to sea. Smith (1966, p. 15) gave the following data on grain size of sediments. The approximate grain-size distribu- tion of sediments entering the bay from all streams is 57 percent clay, 29 percent silt, and 14 percent sand. The composition of sediments in suspension in bay water is 57 percent clay and 43 percent silt. Much of the sand-size material is deposited in the delta and in Suisun Bay, but some is carried into San Pablo Bay and southward. . Colloidal clay particles suspended in fresh water flocculate where the salinity is approximately 1,000 parts per million. The interface between fresh water and water of this salinity value varies in position depending on the volume of fresh water entering the bay. During periods of severe storms and high runoff, the fresh-water—saline-water interface may reach the Golden Gate, but it is generally in San Pablo Bay or in the adjoining Carquinez Strait. A large proportion of suspended sediments settle out in San Pablo Bay because of flocculation of colloidal particles and because of the large areas of relatively quiet water there compared to Carquinez Strait and Suisun Bay. Deposition in the bay system is generally greatest at water depths of 10—30 feet and outside of tidal chan- nels. Deposition also takes place at times in water less than 10 feet deep, but these shallow-water deposits are generally eroded by water made turbulent by wind. Part of the material kept in suspension by tidal cur- rents and other turbulent water, and deposits stirred up, eroded, and resuspended by these waters, are car- ried out to sea through the Golden Gate. Or they are carried to the south arm of the bay, where they form thick deposits in places where the water is quiet. Sand and locally gravel, therefore, are deposited in the tidal channels. Bedrock, as well as gravel and sand, may be exposed where strong ebb currents are concen- trated into narrow powerful currents by shoreline pro- jections. In the San Francisco North quadrangle, such areas are found near Campbell and Stuart Points on Angel Island, Peninsula Point on Belvedere Island, east and south of Alcatraz Island, and in the Golden Gate channel. Smith (1963, figs. 6D, E) presented data indi- cating considerable scour in the period 1855—95 in much of the area between Belvedere Island and San Fran- cisco west of Alcatraz Island and in the period 1850— 1956 along the east shore of San Francisco between China Basin and Hunters Point. Vigorous tidal currents prevent deposition of silt and clay in much of the part of San Francisco Bay lying within the quadrangle. Tidal currents are especially strong and widespread in this part of the bay because this area is the distribution point of the flood tidal cur- rents and the joining point of the ebb tidal currents passing through the Golden Gate channel. Bottom sampling by the US. Coast and Geodetic Survey (1959) yielded rock, gravel, and sandy sediments over much of this area. Mud and clay deposits, however, are thick in other places where tidal currents and water turbu- lence have been weak during the Holocene. Trask (1956, p. 18) showed the general distribution of present-day sediments in San Francisco Bay, based on US. Coast and Geodetic Survey charts. The author found about 1 foot of gravel overlying Franciscan sandstone and shale near the northwest end of Rac- coon Strait. The gravel pieces, 1—6 inches in diameter, consist largely of Franciscan sandstone and some gas- tropod shells. Most gravel pieces were coated with Bryozoa. Holocene mud and clay deposits in the San Fran- cisco North quadrangle reach thicknesses of more than 80 feet along the east shore of the city of San Fran- cisco, including the part now covered by artificial fill (Radbruch and Schlocker, 1958; Goldman, 1969a, p]. 4). These fine-grained deposits are more than 120 feet thick about 2,500 feet northeast of the Ferry Building. Mud is more than 100 feet thick in the center of Rich- ardson Bay at the northwest corner of the quadrangle. Mud is about 40 feet thick near the shore of Sausalito and may be 40—60 feet thick in the northeastern part of the quadrangle, east of Angel Island. TRANSVERSE SAND BARS Soundings by the US. Geological Survey and the US. Coast and Geodetic Survey suggest that between Angel Island and San Francisco, sand is distributed in transverse bars or giant ripple marks resulting from turbulent interaction of ebb and flood tidal currents (Gibson, 1951; Shepard, 1952, p. 1909; Carlson and others, 1970). The approximate boundary of the area of transverse sand bars is shown on plate 1. Detailed soundings of a small area about 1 mile southwest of Alcatraz Island show sand bars oriented approximately due north to N. 300 E. Bar crests are at depths of 50—60 feet, and the average distance between crests is 240 feet. Their height is approximately 6 feet. In Raccoon Strait similar bars are 17 feet high. The swifter westerly ebb current is believed to make the east slopes of the bars gentler than the west slopes. Shallow bor- ings by the US. Bureau of Mines and the State of California Division of Bay Toll Crossings, observed by the author in 1965, show that the youngest deposit SAN FRANCISCO BAY SEDIMENTS 83 located about 1 mile northeast of Alcatraz Island is medium sand at least 8 feet thick. BAY MUD AND CLAY Bay mud and clay are being deposited in San Fran- cisco Bay at the present time. In their natural state they are generally olive gray (5Y 3/2) to dark bluish gray (SB 3 / 1), and they range from soft, plastic, and nearly fluid to moderately firm—about like modeling clay. When dried, the color lightens typically to light greenish gray (5GY 7/1) or light olive gray (5Y 6/1), and the material becomes fairly tough and somewhat brittle. Bedding is not readily apparent in some mud and clay, but in others it is shown by sand partings, 1—3 mm thick, separating clay layers 0.5—20 cm thick or by flattened plant remains. The general particle-size content of the bay mud and clay is 30—60 percent clay- size particles (<2 microns), 30—65 percent silt-size particles (<62—>2 microns), and 1—10 percent sand- side particles (< 2,000—> 62 microns). Lenses and irregular segregations, 1 cm to several meters thick, of fine to coarse sand or mollusk shells are common in the mud and clay in some places. The sand generally has a low silt and clay content. The shell lenses range from a 1-inch-thick layer of a few isolated clam shells in mud or sand, generally both valves being present, to concentrations of shells 3 feet thick in a clay or sand matrix. Shell lenses are mainly in the top 30 feet of the bay mud and clay, according to Hart (1966, p. 43). The shells are mostly pelecypods and minor amounts of gastropods. Although radiocar- bon analysis indicates that shells in thickest lenses are approximately 2,400 years old, the shells are the same species as those living in the bay today (Packard, 1918, p. 236—244). Shell layers as much as 3 feet thick, com- posed entirely of whole or fragmentary separated pele- cypod valves, are found on beaches along the bay south of San Francisco. They were probably concentrated by strong waves and tidal currents that washed away the mud and silt. Buried and preserved, such lenses in time will become coquinas. Goldman (1969b, fig. 1) showed the distribution of shell deposits in San Francisco Bay. Organic remains such as diatoms, peaty plant frag- ments, Radiolaria, Foraminifera, sponge spicules, Bryo- zoa coccoliths, mollusk shells, and so forth are common and even abundant at some localities. Diatoms are especially common and make up 20 percent by volume of some mud and clay. Carbonaceous plant remains commonly make up 3—10 percent of the volume of bay muds and clays. Conspicuously peaty sediments, in scattered lenses about 1 inch to about 1 foot thick, con- tain brownish-black layers of plant remains 1/16—1/1 inch thick and as much as 5 inches long. A persistent peat zone is found at the base of the bay mud and clay in the vicinity of the San Mateo Bridge, commonly as a concentrated layer less than 1 foot thick but locally as a disseminated zone 4 feet thick (Story and others, 1966, p. 48—49) . Water content in bay mud and clay ranges from about 30 to 92 percent; percentages between 50 and 60 percent are common in the uppermost 60—100 feet. The high percentage of water and the plasticity of the sedi- ments stem from their montmorillonite content. When wetted, montmorillonite absorbs water, expands, and becomes plastic. X-ray diffraction analysis of particles smaller than 2 microns from these sediments, from near the east shore of San Francisco and from Richardson Bay, shows about 50—60 percent montmorillonite, about 20—30 percent mica, and about 10—20 percent chlorite. Smaller percentages consist of mica and chlorite ran- domly interlayered with montmorillonite or regularly interlayered montmorillonite and mica (1:1), quartz, and plagioclase. The presence or absence of a small amount of kaolinite could not be ascertained because of the difficulty of detecting small amounts of kaolinite in the presence of much larger amounts of chlorite. A moderate amount of kaolinite, however, was found by the author in mud collected north of Suisun Bay. The silt-size fraction of the mud and clay also has a high montmorillonite content, but it differs from the clay- size fraction in having a much larger proportion of non- clay minerals—mostly quartz and plagioclase feldspar and minor potassium feldspars and organic remains. Sand grains are mostly quartz and plagioclase feldspar and minor potassium feldspar, rock fragments, and shell fragments. Preliminary examination of bay muds from a limited number of localities from depths of 1—30 feet shows that the sand and silt grains generally include 1/2—5 percent heavy minerals (specific gravity >286). These miner- als include, in approximate order of abundance: green hornblende, augite, tremolite-actinolite, pyrite, brown hornblende, micas, clinozoisite, epidote, chlorite, glau- cophane, jadeite, lawsonite, hypersthene, sphene, oxy- hornblende, garnet, zoisite, zircon, apatite, tourmaline, calcite, and anatase. Pyrite, crystallized around diatoms and other organic remains, occurs in drusy spheres, 10—30 microns in diameter, intergrown spheres, and elongate barbs. Irregular round grains, resembling glauconite in form, green or brown in color, and consisting of aggregates of tiny translucent crys- tals, make up about 1—3 percent of the heavy minerals. Analyses by X-ray diffraction and optical microscopy showed that some of them are ankerite. Gram (1966, p. 131) reported 3.1—23.5 percent glauconitic material in sand-size grains of bottom samples obtained from an area between Hunters Point and the San Mateo Bridge. The jadeite, which was abundant in one sample, is the 84 GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE variety showing strong dispersion of the bisectrices and a texture of crystal aggregates similar to the jadeite in metasandstones of Angel Island and the Diablo Range. A program of heavy-mineral analyses of bay sediments from widespread localities and depths would yield information on sediment source and geologic history of the bay. ENGINEERING PROPERTIES OF BAY MUD AND CLAY Use of the bay mud and clay for foundation purposes demands careful engineering design to avoid shear failures and differential settlement problems. The problems are related to the high content of both water and the swelling clay mineral montmorillonite, which together cause low shearing strengths, high voids ratios, low specific gravity, high consolidation under load, and high drying shrinkage. Mitchell (1963, p. 26) sum- marized these and other properties in a table that shows a range of shearing strength of 100—1,200 pounds per square foot. Engineering properties of bay mud and clay are also discussed in Goldman (1969a, p. 20—23), Lee and Praszker (1969), Seed (1969), Steinbrugge (1969), Schlocker (1969, p. 25), Wigginton (1969), San Francisco Bay Conservation and Development Commission (1967), Treasher (1963), US. Army Corps of Engineers (1963), Roberts and Darragh (1963) , Langston, Trask, and Pask (1958), Radbruch and Schlocker (1958) , California Department of Public Works (1955), Lee (1953), and Trask and Rolston (1951) (fig. 57). Treasher (1963, p. 23) found a marked increase in strength of bay mud 40—70 feet below the top of the mud. The lower, slightly stiffer mud is slightly over- consolidated. According to him, the increased strength is the result of desiccation during a glacial period of low sea level. The top 2—4 feet of bay mud loses water and gains strength in areas that have been diked off and drained. STRATIGRAPHIC CORRELATION, FOSSILS, AND AGE Bay mud and clay lie on other Quaternary deposits or on bedrock (fig. 56; pl. 1, section C—C’). Trask and Rolston (1951, p. 1085, 1086) showed bay mud lying on Merritt Sand in an east-west cross section at the southeast corner of the San Francisco North quad- rangle and bay mud overlying undifferentiated Merritt Sand, Posey, and San Antonio Formations in a cross section parallel to and 300 feet north of the San Fran- cisco-Oakland Bay Bridge. The foundation excavation for the Headworks Building of the Southeast Sewage Treatment Plant, near the shore of the bay half a mile south of the quadrangle, and logs of nearby boreholes indicate that the bay mud and clay lie unconformably on the eroded surface of sand deposits that may belong FIGURE 57.—Bay mud, showing desiccation cracks, lying under artificial fill (dune sand), Foundation area Crown-Zellerbach Building, Bush and Battery Streets, San Francisco. to the Colma Formation. These sand deposits are mas- sive, brown, and clayey (described by the author in Radbruch and Schlocker, 1958). They may be part of the same fossil soil zone that is developed also on sand below bay mud and clay found by Story, Wessels, and Wolfe (1966, p. 49) in numerous boreholes in the Vicinity of the San Mateo Bridge. Clays older than the soft uppermost bay mud and clay (informally called younger bay mud and clay by Radbruch and Schlocker, 1958) lie below or are interbedded with the sand deposits (Radbruch and Schlocker, 1958). Older bay clays also lie below a sand layer in the area between Rincon Hill and Telegraph Hill in San Francisco, but according to Treasher (1963, p. 21) this sand forms scattered lenses in San Francisco Bay rather than a continuous blanket on older bay clay. In some areas the younger bay mud and clay (Treasher’s semicon- solidated mud) directly overlie older, stiffer bay clay. The older bay clay is distinct from the overlying younger clay and mud because it is generally highly overconsolidated, from 1.5 to 3 tons per square foot (Trask and Rolston, 1951, p. 1096—1097). ALLUVIUM 85 Fossils found in the younger bay mud and clay are of species living today. They include mollusks and a large assemblage of microscopic flora and fauna. Pre- liminary work on diatoms obtained from boreholes in eastern San Francisco shows both marine and fresh- water species in some mud samples and only marine species in others. Kenneth E. Lohman (oral commun., 1955) briefly examined the diatoms and suggested that the muds containing both marine and fresh-water dia- toms were deposited under brackish-water conditions. For two core samples of bay mud obtained by the Corps of Engineers, US. Army, about 1 mile east of the quadrangle and 2 miles south of its north boundary and collected 11 and 20.5 feet below the sediment sur- face, Kvenvolden (1962, p. 1645) reported the follow- ing radiocarbon dates from unspecified organic mate- rial: 6,210:175 and 7,925:810 B.P. (before present), respectively. Shell and peat in bay mud obtained 6—9 miles south- east of the quadrangle were dated by Story, Wessels, and Wolfe (1966, p. 47). Radiocarbon ages ranged from 2,420:180 to 7,360:320 B.P. for samples 2—50.5 feet below the top of the bay mud. These radiocarbon ages confirm the belief that the modern deposition of mud began after the Wisconsin Glaciation about 14,000 years ago, when sea level began to rise with the melting of ice (Milliman and Emery, 1968, p. 1121—1123). Treasher (1963, p. 23) suggested that semiconsoli- dated bay mud was deposited during a pre-Wisconsin interstadial high stand of sea level, was slightly over- consolidated by desiccation during a brief exposure to air during the subsequent Wisconsin Glaciation, and was inundated again and covered by the uppermost bay mud after the Wisconsin Glaciation. Older bay clays may have been deposited during periods of high sea level associated with the Sangamon and earlier interglaciations. Great volumes of the uppermost soft mud were no doubt deposited during the days of hydraulic mining of gold when huge amounts of debris were fed into the Sacramento River. Gilbert (1917, p. 48—49) stated that the chief deposit in the bay from mining debris is mud. He also estimated that inthe period 1849—1914, 50 million cubic yards of sediments, mostly of grains smaller than fine sand and mostly derived from mining debris, reached the ocean. As early as 1862, many years before hydraulic mining reached [its peak, Brewer (1930, p. 295) wrote as follows: Previous to 1848 the river (the Sacramento) was noted for the purity of its waters, flowing from the mountains as clear as crystal; but, since the discovery of gold, the “washings” render it as muddy and turbid as is the Ohio at spring flood—in fact it is perfectly “riley”, discoloring even the waters of the great bay into which it empties. ALLUVIUM Alluvium, which is material transported and deposited by running water, is only sparsely present in the quad- rangle. Alluvium is distinguished with difficulty from slope debris and ravine fill, which were transported by creep and landsliding. Evidently streams are too small to transport enough slope debris and ravine fill to create substantial deposits of alluvium. In San Fran- cisco the high porosity of the dune sand may have inhibited substantial surface flow. Alluvium that formed in the last few hundred years may be repre- sented by the deposits in the valley at Horseshoe Bay and in the southwest-trending valley of an intermittent stream in the west-central part of Marin Peninsula. The bed of the largest intermittent stream in San Francisco, Lobos Creek, is dune sand, artificial fill, and Colma Formation. Valleys on the east flank of Twin Peaks contained streams that flowed into the bay when the city was founded in 1776, but the stream beds now are covered with slope debris or artificial fill. Several deposits of alluvium were mapped on the west slopes of Twin Peaks (fig. 58). They have a maxi- mum observed thickness of 15 feet and are interbedded with and grade laterally and vertically into slope debris and ravine fill. Most of these deposits are related to slightly older drainage systems than the present ones and are now moderately dissected. Similar alluvial deposits occur in many other parts of the quadrangle, but they are not shown on the map (pl. 1) because they are largely or entirely concealed by other surficial deposits or by artificial fill and manmade structures. Most of the alluvium is composed of medium silty clayey sand; clean medium sand occurs locally. Allu- FIGURE 58.—Alluvium (Qal) lying on Colma Formation (QC).’ Greenstone (Kjg) at right. West slope of 'I\Nin Peaks, San Francisco. 86 vium is generally better sorted than slope debris and ravine fill. Chert and greenstone pebbles are locally abundant in the alluvium, especially near the heads of the valleys. LANDSLIDE DEPOSITS Landslide deposits are widespread throughout the quadrangle, and they indicate that landsliding is one of the principal agents of erosion. The deposits mapped as landslides, however, are rather local features that were formed by relatively recent landsliding. Landslide topographic features, such as shear-failure scarps or hummocky topography, are visible, even though some of them are considerably modified by erosion. Spoon- shaped cliffs like the one just west of the Golden Gate Bridge tower (fig. 19) may have been shaped by landsliding. In the downward and outward movement of slope- forming material that constitutes landsliding, part of the potential energy of the slope is converted to kinetic energy and friction; the subsequent landslide deposit has correspondingly less potential energy, and so many slopes are more stable after landsliding than before. Thus, if a slope is to remain unstable over a long period of time, the landslide deposits must be removed. In the San Francisco North quadrangle, instability is usually perpetuated by rainwash and stream erosion, espe- cially during large storms, that rework and move land- slide deposits downslope or completely remove them from the landslide-sculptured slope. For this reason the activity of landsliding is directly proportional to the activity of other agents of erosion. Consequently, some other Quaternary deposits, especially some of the Colma Formation and slope debris and ravine fill, are reworked landslide deposits, but they are older than the landslide deposits designated as such on the map (pl. 1). Though landslides are widely distributed in the quadrangle, the small volume of landslide deposits shown on the geologic map (pl. 1) may seem to belie the important role of landsliding as an agent of erosion. Local construction activities have obscured many of the distinctive topographic features of landslides. On the other hand, abundant manmade structures afford numerous known survey points for measuring even slight landslide movements. Landslide processes are succinctly discussed by Vames (1958, p. 42—45) , who pointed out the following: The process of landsliding is essentially a continuous series of events from cause to effect. * * * Very seldom, if ever, can a slide be attributed to a single definite cause. The process leading to the development of the slide has its beginning with the formation of the rock itself, when its basic physical properties are determined, and includes all the subsequent events of crustal movement, erosion, and weathering, until some action, perhaps trivial, sets a mass of it GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE in motion downhill. The last action cannot be regarded as the one and only cause, even though it was necessary in the chain of events. Because most slides involve failure of slope materials under shear stress, a logical account of causes of land- sliding would deal with (1) factors that contribute to high shear stress and (2) factors that contribute to low shear strengths. These factors are outlined in detail by Vames (1958, p. 42-45). Removal of lateral and underlying support is the most common factor that contributes to high shearing stress in a slope. It includes actions of erosion, of previous landsliding, and of man. A variety of factors leads to low shear strengths, includ- ing the initial state of the slope material, subsequent changes brought on by weathering, and other physico- chemical reactions. In addition, water is an important factor in slope stability, for it contributes both to high shear stress and low shear strength. The numerous landslides in the quadrangle owe their existence to several factors: an irregular hilly terrain with gentle to steep slopes underlain in many areas by unconsolidated surficial deposits or by badly sheared and shattered bedrock, a relative abundance of highly plastic and swelling clay in all the foregoing materials, occasional periods of prolonged rainfall, occasional earthquakes, and the continuous disturbance and alter- ation of the original terrain by man. The distribution of landslides in the quadrangle is shown on plates 2 and 3. Slopes underlain by strongly sheared Franciscan rocks, by serpentine, and by the sheared-rock map unit are especially prone to land- sliding. The largest landslides are in sheared serpentine FIGURE 59.—Steeply cut slopes in serpentine, Potrero Hill, San Francisco. Cut face to right of stairway has remained intact for many years except for minor sloughing and raveling. Slope beneath and to left of stairway was probably the site of a small landslide. LANDSLIDE DEPOSITS 87 and in Franciscan rocks in the Lands End area and in the sheared serpentine south of the Golden Gate Bridge. Wave action periodically reactivates some of these slides by removing the supporting material at their bases. Serpentine, contrary to popular belief, is not especially prone to landsliding everywhere, because the large serpentine body of Potrero Hill is almost com- pletely free of landslides (fig. 59). Harding (1969, p. 66) pointed out that “areas of obvious potential insta- bility can be mapped on a regional scale, however, the corollary, mapping whole hillside areas as stable on the basis of rock type, cannot safely be done.” Many landslides too small to be shown on the geo- logic map (pl. 1) are numerous in the slope debris and ravine fill, especially along roadcuts. Soil moisture in ravine fill is believed to be a primary cause for sliding, although the actual movement may be triggered by other factors. CHARACTERISTICS OF LANDSLIDES Slides range from rockfalls at one extreme to mud- flows at the other; earthflows and debris flows are the commonest types. Many landslides, however, are com- plex; their upper parts are rockslides, debris slides, or rotational slumps, and their lower parts are earthflows or debris flows (figs. 60, 61). A few rockfall deposits are found below steep quarry faces and along steep rocky shores. Characteristics of landslides are given on plate 3. FIGURE ESQ—Landslide in center of photograph is a slump and debris slide in its upper part and a debris flow in its lower part (No. 93, pl. 3). Note fracturing of sandstone of the Franciscan Formation. South Bay part of Golden Gate, San Francxsco, looklng southeast. (Barney Peterson, photographer, used with permission of the San Francisco Chromcle.) 88 GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE FIGURE 61.-—Landslide in which soft sheared sandstone and shale, below radiolarian chert, is moving downhill as a rota- tional slump and debris slide (No. 51, pl. 3). Toe of slide above roadcut has bulged the surface and is calving onto road. One mile northwest of Lime Point, Marin Peninsula. The frequency of landsliding differs greatly for the various lithologic units. About 60 percent of the land- slides involve surficial deposits and weathered and sheared bedrock. More than half the landslides in bed- rock, including sheared and weathered rock, involve sandstone of the Franciscan Formation, 21 percent involve greenstone, 8 percent involve radiolarian chert, and 9 percent involve serpentine. Bonilla (1960) obtained similar results in an analysis of landslides in the San Francisco South quadrangle. These percent- ages, however, do not take into account the size of slides, nor the relative area of exposure of each type. Serpentine slides are among the largest in the quad- rangle. Landslides involving radiolarian chert are gen- erally of small size. A more satisfactory statistical com- parison of the tendency of various materials to slide would evaluate dimensions or volume of slides and the relative abundance of the various lithologic units. Proper evaluation of slope stability can be made only by considering such local factors as degree of shearing and chemical alteration, steepness of slope and ground- water conditions, as well as the initial state of the rock. The frequency of types of landslide movement fol- lows. Each type of movement in a complex landslide is counted as a separate landslide in the tabulation. Type of movement Percent of total landslides Earthflow ...................................................... 28 Debris flow .................................................... 24 Debris slide ................................... 22 Rotational slump ............................. 13 Rockfall ............................................. 5 Sand flow .................................... 3 Block glide .............................. 2 Mudflow ....................................... 2 Debris avalanche ........................................ 1 100 The slope angle before sliding ranges Widely for sandstone of the Franciscan Formation, but it averages 35°; for greenstone it averages 31°. These averages, of course, cannot be used to design slopes at individual construction sites. The lower slope angle of greenstone probably reflects the tendency of greenstone to alter by weathering and related processes more readily than does sandstone. Greenstone alteration products, more- over, consist mostly of easily sheared clay minerals including large proportions of swelling clay. Landslide deposits vary widely in composition. Because earthflows and debris flows are the commonest types of slides, most deposits consist of heterogeneous unstratifiedfmixtures of rock, sand, silt, and clay in proportions that vary within each landslide and from one landslide to another. In the large landslide half a mile north of Lime Point on Marin Peninsula (pl. 3, Nos. 12, 13, 48, 52), most of the landslide debris is stabilized, though some of it near the bases of the slides has moved since about 1940. One of the earliest published accounts of slope insta- bility was that of Captain Beechey who visited San Francisco in 1827. In his “Narrative of a Voyage to the Pacific and Beering’s Strait” (Beechey, 1831, p. 345), he reported the following at Fort Point: “The fort, which we passed upon our right, mounts nine guns, and is built upon a promontory on the south side of the entrance, apparently so near to the precipice, that one side will, before long, be precipitated over it by the gradual breaking away of the rock.” ARTIFICIAL FILL The practice of creating land by dumping artificial fill on the gently shelving tidal flats along the east and north margins of the San Francisco Peninsula was begun before 1850. Flatland has been at a premium since the Gold Rush first made San Francisco a center of growth and development. More than 3 square miles of the most valuable land in San Francisco originated in this way (pl. 1; fig. 62). The average thickness of the fill north of China Basin is about 10 feet; south of China Basin it reaches a maximum thickness of about 60 feet. Similarly, large areas of land have been made in the Sausalito-Tiburon area. The thicknesses of arti- ARTIFICIAL FILL FIGURE 62.——Map showing shoreline of San Francisco in 1853, present shoreline, and areas formerly covered by water that are now artificially filled (shaded). Base modified from Chart 627, US. Coast and Geodetic Survey (formerly US. Coast Survey). 89 90 GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE ficial fill at several places are shown on the geologic map and cross sections (pl. 1). The earliest artificial fills consisted of nearby dune sand and, less frequently, colluvium and weathered rock, which were simply dumped on the soft wet bay mud. A seawall, built to prevent the fill from being washed away now extends more than 3 miles from Fort Mason to China Basin and supports the Embarcadero. The first sections of the wall were made by building a long ridge of rock and earth. Subsequent better designed sections consist of a concrete wall supported on piles set in a rock embankment. For a long time, no restrictions were placed on the type or quantity of fill; spoil from excavations, debris from quarries, dune sand, and mud dredged from the bay were dumped indis- criminately. Some parts of the fill area were also used for public dumps; so, fairly large deposits of manmade debris were added to the mixture, including miscella- neous discarded objects of a large urban population, such as shoes, bottles, broken pottery, and bed frames. The Marina Park fill area in San Francisco was created in 1913 for the Panama-Pacific Exposition. Treasure Island was created from sandy bay sediments obtained southwest of Blunt Point and piped to the site by hydraulic transport. The site was enclosed by sheet piling. Maps showing historical development of reclaimed land are found in “Future Developments of San Fran- cisco Bay Area 1960—2020” (U.S. Office of Area Devel- opment, 1959, pl. 19) , which also includes a discussion of reclamation of marshland, tideland, and submerged land (p. 7 5—94), and in “The San Francisco Bay Area —A Metropolis in Perspective” (Scott, 1959, p. 37). The hilly parts of the quadrangle are also dotted with thousands of small fills made in the course of road and building development. Most of these fills are too small to be shown on a map at a scale of 1 : 24,000, but the largest ones are shown. They are generally along the downhill sides of major roads or are in large deep ravines that are used as dumps for rock and soil waste from nearby excavations. A large area of artificial fill between Mount Olympus and Golden Gate Park is now largely concealed by urban development. Evidence for the filling was found by studying logs of boreholes and by comparing old and modern topographic maps. The comparison showed that the valley was formerly deeper and its walls were steeper. STRUCTURE The largest conspicuous structural features in and bordering the San Francisco North quadrangle are northwest-trending faults and shear zones (figs. 1, 63) , including the San Andreas fault—one of the most per- sistent structural features in California (fig. 6) (Dick- inson and Grantz, 1968). The San Andreas fault is known to extend more than 600 miles northwestward from the Gulf of California to Point Arena. It is the boundary between the North American and Pacific Ocean crustal plates. The North American crustal plate extends to the east of the fault and contains the Fran- ciscan Formation as the oldest known rock in this area; the Pacific Ocean crustal plate extends westward and contains Cretaceous granitic rocks that intrude older gneisses, schists, marbles, limestone, and dolomite. On San Francisco Peninsula the Pilarcitos fault, which lies west of the San Andreas fault, is the local west bound- ary of the Franciscan Formation. The San Andreas fault intersects the coast north of Mussel Rock, approx- imately 51/2 miles south of the quadrangle. The Pilarcitos fault intersects the coast at the Pedro Valley district of Pacifica, 10 1/2 miles south of the quadrangle. The San Francisco North quadrangle lies near the west edge of the crustal plate made up of the Fran- ciscan Formation; the Golden Gate Bridge is about 61/; miles east of the San Andreas fault. The Hayward fault, another prominent active northwest-trending shear zone, lies 10 miles east of San Francisco and, in the bay area, is nearly parallel to the San Andreas fault. Strong historic earthquakes were centered on the San Andreas fault in 1838, 1865, and 1906 and on the Hay- ward fault in 1836 and 1868. Movement along these active faults is mostly horizontal and right lateral, such that points along the west side of the faults move northwestward relative to points along the east side. Part of the fault movement is vertical, for small vertical displacement of the surface accompanied the 1868 faulting on the Hayward fault (Radbruch, 1967) and the 1906 faulting on the San Andreas fault. Additional evidence for vertical movement on these faults is the associated fault scarps. The San Andreas fault is at the base of a high ridge, Bolinas Ridge, north of Bolinas Lagoon, about 12 miles northwest of the quadrangle. The Hayward fault lies mostly at or near the base of the Berkeley Hills, and small scarps are within the fault zone in the Centerville area. At least some of the verti- cal movement on these faults is post-Pliocene, for the faults disrupt Pliocene sediments. Lawson (1914, p. 14) believed that the San Fran- cisco North quadrangle is part of a structural block in which bedrock was tilted down towards the northeast from San Bruno Mountain, Bolinas Ridge, and Mount Tamalpais to the foot of the Berkeley Hills. The crust between San Bruno Mountain and the Berkeley Hills, however, probably is cut by numerous faults and prob- ably has not acted as a rigid block. The presence of bedrock far above sea level on Angel Island, at El Cer- rito, at Potrero San Pablo in Richmond, at Red Rock STRUCTURE 91 122° 22' 30" | I malsaw Potrero Point 37°45' NMHQIHNEHQ 3N! 7A)/S SAN FRANClSCO CIT‘ AND\) COUNTY SAN MATEO CO—UNTY ‘ w ‘_ < 0 37°42’30" O 2 M|LES |_—*%_J \ l} / Geology in San Francisco QUATERNARY Soulh quadrangle modified after Bonilla (1965} CRETACEOUS AND JURASSICl?) 1L r Great Valley sequence Surficial rocks Sandstone and shale Rocks in the Point Labos area, west ofthe City College shear zone, are tentatively assign- ed to the Great Valley sequence . . ‘ Francnscan Formanon A fl Radiolarian chert, Sandstone Sheared rocks greenstone, and and shale Includes fragments some sandstone ofFranL-iscan and shale Forma/ion, Greal l L ¥ ’ l ' Contact Valley sequence, and serpentine Anticline Syncline FIGURE 63.—Generalized geology of parts of the San Francisco North and South quadrangles showing major shear zones and fold axes. Geology south of latitude 37°45’ by M. G. Bonilla (1965) . 92 GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE Island, and at Coyote Hills near Newark is not com- patible with Lawson’s tilted-block hypothesis, which was based mostly on the presence of bedrock more than 1,000 feet above sea level at San Bruno Mountain and more than 1,000 feet below sea level west of the Berke- ley Hills in the vicinity of Berkeley. Gravity observa- tions by Taylor (1957) in San Francisco Bay south of the San Mateo Bridge are also incompatible with simple northeast tilting of the crust. Cross sections by Taylor (1957, pl. 5—7) showed that the top of bedrock is generally deepest in the middle of San Francisco Bay and that steeply-dipping faults with a strong vertical component of displacement break the bedrock into six or more blocks. Furthermore, epicenters of many small- magnitude earthquakes are located in the area between the San Andreas and Hayward faults. More than 100 were recorded in 1968 by the U.S. Geological Survey (Roller and others, 1968). Tocher (1959, p. 48) wrote that a small shock (magnitude 2.1) in December 1956 had an epicenter near Yerba Buena Island and noted that Townley and Allen (1939) mentioned that many shocks were felt on Russian Hill in San Francisco in April 1905. An epicenter of a magnitude 4 earthquake was determined to be in San Francisco Bay near Red Rock Island, 8 miles north of San Francisco (California Department of Water Resources, 1964). Thus, instead of simply being tilted northeast, the block between San Bruno Mountain and the Hayward fault is struc- turally complex and is cut by many faults, some of which are probably active. The position of the bedrock is related more to local tectonism within the area rather than to simple tilting. The structure of the Franciscan Formation in the San Francisco North quadrangle is poorly known, owing in large part to a lack of exposures and also to its structural complexities. The character of the for- mation further contributes to the difficulties, for dis- tinctive and persistent marker beds are lacking, fossils are almost nonexistent, lateral variations in texture, thickness, and lithology are the rule, and bedding is poorly developed in many places. Repeated crustal movements have left their mark in the intensity to which the Franciscan is sheared and shattered. Almost every exposure has one or more sets of slickensided slip planes and gouge zones. Moreover, a large part of the Franciscan is covered by water, surficial deposits, and manmade structures. The rocks between Fort Point and Potrero Hill (fig. 62) are believed to be part of a major shear zone that extends northwest to the middle of Golden Gate chan- nel and southeast toward Hunters Point about 1 mile southeast of the quadrangle. Airborne magnetometer studies (US. Geological Survey, 1966) suggest that this shear zone extends southeast across San Francisco Bay to a point north of Coyote Hills about 16 miles from Hunters Point. Orientation of shear planes in Potrero Hill and the Presidio suggest that the zone may dip eastward at a low angle. A second shear zone exposed at Lands End is believed to be part of another major northwest-trending shear zone that is largely concealed by dune sand in the San Francisco North quadrangle but that is also found in the Sunset Reser- voir area at the south edge of the quadrangle. It is cor- related with the City College fault mapped by Bonilla (1961) in the San Francisco South quadrangle. In the central highlands of San Francisco, rocks in the block bounded by these shear zones are mostly radiolarian chert and greenstone folded primarily on east-west axes and secondarily on north-south axes, but rocks northeast of the Fort Point—Hunters Point shear zone and southwest of the City College shear zone are mostly graywacke sandstone and shale folded mostly along northwest-trending axes. The potassium feldspar-bearing sandstone at Point Lobos and the Cliff House and the concealed bedrock southwest of the City College shear zone at Lands End are probably part of the San Bruno Mountain block of the Great Valley sequence, which is the same age as the Franciscan Formation. The Great Valley sequence has either been thrust westward over the Franciscan (Irwin, 1964, p. 66; Bailey and others, 1964, p. 123— 141, fig. 72) or the Franciscan Formation has been thrust eastward under the Great Valley sequence (Bailey and others, 1970) . Thus, the City College shear zone may also be the locus of Late Cretaceous thrust- ing that carried miogeosynclinal rocks of the Great Valley sequence over eugeosynclinal rocks of the Fran- ciscan Formation. Later movement along the same zone may have been vertical or horizontal. The rocks between Bakers Beach and the shear zone at Lands End consist mostly of graywacke and minor radiolarian chert, greenstone, and serpentine. The sed- imentary rocks dip northeastward; however, graded bedding and channeling indicate that part or all of the section is overturned. Similar features in the litholog- ically similar sedimentary rocks at the north end of Bakers Beach indicate that the northeast-dipping sedi- mentary rocks are not overturned. These attitudes suggest the presence of a northwest-trending anticline whose axis lies within Bakers Beach. No additional evidence for such a fold was seen to the southeast be- cause bedrock is concealed by beach and dune deposits. A syncline that plunges northwestward along Colum- bus Avenue between Russian Hill and Telegraph Hill in San Francisco is suggested by bedding attitudes and by the disposition of an apparently persistent unit of shale and thin-bedded sandstone of the Franciscan Formation. Orientation of graded bedding and small- SEISMICITY 93 scale channeling suggest that locally the west limb of the syncline is overturned. Opposing flanks of a large anticline are suggested by northeasterly dips on Alca- traz Island and southwesterly dips on Telegraph Hill. Meager structural data indicate one or more northwest- trending large folds between the Columbus Avenue syncline and the belt of sheared rocks between Potrero Hill and Fort Point. On Marin Peninsula the immense section of southwestward-dipping sediments and green- stone apparently forms the limb of a large anticline whose axis lies in Richardson Bay (pl. 1, section A—A’). On Angel Island the principal structure is a broad syn- cline whose axis plunges northwestward to Hospital Cove and beyond along the middle of Tiburon Peninsula. No satisfactory structural correlation exists between Marin and San Francisco Peninsulas. The marked con- trast in stratigraphy, lithology, and local structures between the two suggest that a major fault exists in the Golden Gate channel, but no other supporting evidence was found, with the possible exception of the westward-trending shear zone at Point Diablo. The alinement of the Richardson Bay anticlinal axis with the axis between Alcatraz Island and Telegraph Hill is one possible link. The radiolarian chert and greenstone southwest of the Fort Point—Hunters Point shear zone may be represented by similar rocks in the Point Bonita quadrangle. Rocks in both areas are folded mostly on east-west fold axes (Lawson, 1914, geologic map of Tamalpais 15’ quad.) . Another interpretation of the structural link across the Golden Gate is that the Columbus Avenue syncline is a part of the Angel Island—Tiburon Peninsula syn- cline offset 2 miles westward by a northeast-trending fault south of Angel Island. Under such an interpreta- tion the Fort Point—Hunters Point shear zone could continue in offset as the serpentine of Angel Island and Tiburon. However, fieldwork by the author in the Point Bonita quadrangle indicates that the Fort Point— Hunters Point shear zone may continue northwestward across the Golden Gate to the shear zone of the Bonita Cove area. SEISMICITY The San Francisco North quadrangle lies between the active San Andreas and Hayward faults and is within a region of high seismic activity, in which earth tremors are frequent and unpredictable. The five larg- est earthquakes since 1800, as well as many other strong shocks, originated from movement on or near these two faults. Parts of the Calaveras fault zone, located about 20 miles east of the quadrangle, are also active. Epicenters located by seismographs suggest that other faults in the bay area are also active, but data are too meager to identify them. Surface ground breakage by historic fault movement during earthquakes has occurred on the San Andreas fault (Lawson and others, 1908) and on the Hayward fault (Radbruch, 1967). No field evidence of recent fault movement has been found in the quadrangle. Epicenters for small earthquakes in and near the quadrangle are shown on maps by Byerly (1951, p. 159) for the periods 1930—41 and 1947—48 and by Tocher (1959, p. 46) for the period 1942—57. None of these epicenters are located in the quadrangle, though several are shown in nearby areas. The epicenter of the strongest shock recorded between 1930 and 1957, that of March 22, 1957 , Richter magnitude 5.3, is near Mussel Rock Where the San Andreas fault intersects the Pacific Ocean shore 51/2 miles south of the quad- rangle. Several other epicenters of weaker shocks are also near Mussel Rock and west of Golden Gate, gen- erally along the alinement of the San Andreas fault; one epicenter is about 1 mile south of Twin Peaks; one is on the shore of the bay near Islais Creek; two are near Hunters Point, 2 miles southeast and 3 miles southwest of the point; and three are within 1—4 miles west and northwest of the northwest corner of the quadrangle. Except for the March 22, 1957, earth- quake, all shocks were only strong enough to rattle windows and doors and occurred during the period 1930—41. During 1968 the US. Geological Survey recorded six epicenters of microearthquakes in the quadrangle (Roller and others, 1968), all located near the northwest shore of San Francisco between Fort Mason and Hunters Point. Earthquake history of the San Francisco Bay area is discussed by Tocher (1959, p. 39—48) and Byerly (1951, p. 141—160). The five largest known earth- quakes are those of June 10, 1836, and October 21, 1868, which originated from movements along the Hay- ward fault, and those of June 1838, October 8, 1865, and April 18, 1906, which originated from movements along the San Andreas fault. Serious damage to struc- tures in San Francisco is recorded for the 1865, 1868, and 1906 earthquakes. Almost no information is avail- able on damage to structures from the 1836 and 1838 earthquakes because the population of San Francisco was small and structures were few, though the walls of the Presidio and Mission Dolores were reported to have been seriously damaged by the 1838 earthquake (Louderback, 1947). Effects of the 1906 earthquake have been extensively documented in many publications. Some of the more comprehensive reports are by Gilbert, Humphrey, Sewell, and Soule (1907), by the American Society of Civil Engineers (1907 ), and by Lawson and others 94 GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE (1908). Though structures were damaged in all parts of the city, much of the damage was moderate. Never- theless, in the heavily built-up northeastern part of the city, from North Point to Townsend Street and from the Embarcadero to an irregular line approximately along Van Ness Avenue to the intersection of Dolores and 20th Streets, destruction by the earthquake and the resulting fire was a great catastrophe. About 400 lives were lost in the city, and estimates of total damage range from $350 million to $1 billion. Esti- mates of the loss due to the earthquake exclusive of the fire range from 5 to 20 percent of the total damage. In San Francisco the relationship between damage to manmade structures and geology was investigated by Wood (1933, p. 67—82; 1908, p. 220—245; see also Lawson and others, 1908, atlas map 19). Though Wood postulated a general increase in intensity of shaking towards the fault, he found that damage to structures was influenced mostly by the geology of their founda- tions. The least damage was in hilly areas of the city where buildings were founded on or near Franciscan Formation bedrock. Damage was 5—10 times greater in the areas where structures were founded on artificial fill lying on bay sediments (Duke, 1958, p. 9). Wood found that some of the damage in the filled-in areas was caused by settling of poorly consolidated material. Another pertinent observation on damage to structures in the filled-in areas was made by the American Society of Civil Engineers ( 1932, p. 40) : “No class ‘A’ building founded on well-driven piles or well-placed concrete piers suffered any material injury due to the earth- quake; serious damage or collapse occurred only in poorly constructed buildings erected on soft founda- tions in the ‘filled-in’ areas.” Steinbrugge (1968) dis- cussed relations of building damage to foundation material, earthquake risk zoning, and building code or- dinances requirements for earthquake resistant design. For the Alaska earthquake of March 27, 1964, dam- age was also greatest in areas underlain by thick satu- rated unconsolidated deposits, whereas there was no significant damage to structures founded on indurated bedrock or on bedrock with a thin veneer of unconsoli- dated deposits (Eckel, 1970, p. 29). Damage was related more to local geology than to distance from the epicenter (Eckel, 1970, p. 29). A rough appraisal of earthquake stability of the geologic units in the quadrangle is given in table 11. A cooperative effort of the Earthquake Engineering Research Institute, the National Science Foundation, the National Center for Earthquake Research (NCER) of the US. Geological Survey, and several State of California agencies is being made to place more than 40 instruments in and near the quadrangle to record motion during strong local earthquakes. These instru- ments will obtain data on shaking characteristics of various types of geologic rock units, surficial and bed- rock, in various geologic environments. Measurements made by the National Center for Earthquake Research in and south of this quadrangle of ground motion generated by underground nuclear explosions in Nevada show good correlation of motion amplitude with geologic setting of the recording site. Maximum horizontal ground velocities at sites under- lain by bay mud generally increased with thickness of bay mud and were as much as ten times greater than those recorded on nearby bedrock (Borcherdt, 1970). Additional information on the behavior of founda- tions in the San Francisco Bay area during earthquakes is given by Bonilla (1970), Seed (1970), and Cluff and Bolt (1969). ENGINEERING GEOLOGY The San Francisco North quadrangle, particularly the city of San Francisco, is blanketed by various kinds of unconsolidated surficial deposits through which pro- trude scattered patches of more durable bedrock. The quadrangle is between two major active faults and is therefore subject to frequent earth tremors. Other geologic phenomena peculiar to the quadrangle are two broad zones of sheared rock, many landslides, and large areas of land that have been reclaimed by covering soft mud with artificial fill. All these features are of concern to the engineering geologist and the engineer because of the special problems they pose individually and in combination with each other. The differences in permeability, shearing strength, bearing capacity, and other engineering properties between bedrock and surficial deposits are generally large, and in addition, these properties may vary con- siderably over short distances within a single bedrock or surficial deposit map unit. At some sites on the heavily utilized land areas, heavy structures have been built partly on soft bay mud and partly on hard sand- stone of the Franciscan Formation. Proper use of such sites for foundation purposes demands careful explora- tion and engineering design. Even construction sites entirely on the Franciscan Formation encompass shear zones consisting of soft clayey material studded with hard rocks. These zones are a few inches to many feet wide. Thus, many building sites are founded on rocks with differing bearing capacities and resistances to seismic shocks. Wood (1933; 1908) concluded that the damage to structures in the 1906 earthquake was controlled mostly by the foundation of the building. Buildings founded on bedrock were the least damaged, and build- ings on well-designed and well-placed piles or concrete ENGINEERING DEPOSITS 95 footings or mats in soft saturated poorly consolidated bay or marsh sediments did not suffer material damage. Serious damage occurred only in poorly constructed buildings erected on soft foundations in the filled-in areas. On the other hand, Steinbrugge (1968, p. 30) observed that damage from the July 28, 1957, earth- quake in Mexico City, which is founded on soft wet expansible lake clays, was mostly to tall reinforced concrete buildings, whereas 1—2 story “collapse hazard” buildings performed well. These and other observa- tions made in a worldwide study make it apparent that the natural period of vibration of the foundation and the engineering structure, as well as the nature of the earthquake vibrations and the foundation material, must be considered in design and construction. In addition to the increased earthquake effects to buildings on artificial fill, differential subsidence is a second serious problem. In less than 60 years, poorly made fill has settled more than 7 feet (fig. 64) (Bonilla and Schlocker, 1966, p. 452). To avert possible subsi- dence, construction of modern fills should be preceded by careful study of the properties and ground-water conditions of the fill foundation, and only selected materials should be laid under controlled conditions. Engineering problems of reclaimed land in the north- eastern part of the city are discussed in “Subsidence FIGURE 64.——Diiferential settlement of buildings on artificial fill in former Mission Swamp. Photograph taken shortly after street, sidewalk, and sewerline were raised to former level before subsidence. Near Sixth and Folsom Streets, San Francisco. and the Foundation Problem in San Francisco” (Amer- ican Society of Civil Engineers, 1932, p. 24—36). Less serious engineering problems in areas of artificial fill along the fringes of the bay are raised by the old pilings and sunken ships which are found in pile driving and caisson sinking. Also, 01d garbage dumps present foundation excavation problems because of the varia- tion in strength and bearing capacity of miscellaneous objects in the dump. Because of the great contrast in engineering behavior between bedrock and surficial deposits, the location of the buried bedrock surface (pl. 3) and the generalized engineering characteristics of all the exposed rock units (table 11) are important in planning structures. The contours shown on the bedrock map are based largely on records of borings made for foundation purposes, water wells, and on records of tunnel excavations. Most of them are unpublished. Some of the logs are given as abbreviated notations on the geologic map (pl. 1), and some are published in “Selected Logs of Borings” (Institute of Transportation and Traffic Engineering, 1951) and in “Subsidence and the Foundation Prob- lems in San Francisco” (American Society of Civil Engineers, 1932). As a further aid in evaluating the exposed bedrock surface for possible construction sites, the locations of faults, shear zones, and known land- slides are also shown on plate 3. Engineering properties of San Francisco Bay sedi- ments are discussed briefly in the section “San Fran- cisco Bay Sediments,” where the reader will find refer- ences to other more detailed discussions. Engineering properties of other geologic map units in the quad- rangle are discussed by Trask and Rolston (1951), Lee (1953), California Department of Public Works (1955) , and Schlocker (1969; 1970) . The reader should also consult the sections “Landslide Deposits” and “Seismicity” for additional material on engineering geology. In conclusion, each construction site within the quadrangle presents a unique combination of such conditions as topography, hydrology, and foundation materials. Foundation materials, in turn, vary in degree of consolidation, fracturing, and chemical alteration. The bedrock surface map (pl. 3) and the table sum- marizing the characteristics of rock units (table 11) provide only a general framework on which to base plans for particular site investigations. The information is generalized, covers a large area, and does not sup- plant detailed site investigations—both field and labo- ratory—that are necessary for evaluating specific sites for specific structures. It is hoped that the presentation of these data will serve as a reminder to stimulate the acquisition of precise geologic data for use in the proper location and design of structures. 96 GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE TABLE 11.—Generalized description of [Franciscan Formation extends Permeability Workability Slope stability Name and map symbol Weathering. soil development (pl. 1) Lithology alteration Artificial fill Mostly dune sand but None. (Qaf) . includes silt. clay, rock waste, manmade debris, and organic waste. Landslide Variable. Rock pieces of None or very little. deposits all sizes in sand, silt, (Q1, 010, and clay matrix. 01y). Alluvium Gray silty clayey medium No well-developed soils; top (Qal) . sand; fine to medium 1—2 ft of older alluvium has sand; clayey silt; some abundant plant fragments and pebbles. Grades to slope higher silt content than lower debris and ravine fill. portions. Older alluvium is rich in plant fragments. Bedding obscure. Modern and Gray well-sorted medium None. older beach to coarse sand. A few deposits small gravel beaches. (Qb. (3201))- Generally none, but a slight weathering of top 2—4 ft seen in older material on Marin Peninsula. Light-yellow to reddish- brown unsorted rock fragments, gravel, sand, silt, and clay in various proportions. Clayey parts exhibit moderate swelling and plasticity when wet. Slope debris and ravine fill (12”) . Gray silty organic (lig- None. nitic and diatomaceous) clay with minor amounts of sand; 10- cally, lenses of sand, peat. or shell frag- ments. Soft and plastic near top, moderately stifi' at depth. Shrinks and becomes hard when dried during excava- tion; plastic and swells when rewetted. Bay mud and clay (Qm). Yellowish-brown to light- Slight to none. Most grains coated gray well-sorted fine to with film of iron oxide. Minor medium sand. Quartz, amounts of carbonaceous plant feldspar, and horn- matter disseminated locally in blende are chief top 2—3 ft. minerals. Dune sand Qd) . Moderate to slight soil develop- ment; in places soil identified only by presence of organic matter, increase in silt and clay content, and iron staining. Light-brown to orange fine to medium sand with minor amounts of clay. Evenly spaced horizontal or nearly horizontal bedding and crossbedding. Beds 1—3 in. thick. Cobble-size rubble beds rare. Low swelling and plasticity when wetted. Mostly soft friable sheared rock enclosing hard spheroidal knobs of unsheared serpen- tine (sp). Rarely massive and tough (sph). Various colors but generally greenish- gray, blue, or brown. Colma Formation Qc) . Soil generally absent or less than 1 ft thick. Locally, 5—10-ft-thick mantle of dark-gray clayey material containing much high- swelling montmorillonite; probably derived from sheared and hydrothermally altered serpentine. This material is “adobelike” and develops deep shrinkage cracks on drying. Serpentine (sp, sph) . Generally easy to remove except locally in tangles of ship timbers and other manmade debris. High, except where clayey materials predominate. Variable, but generally high. Variable, but generally easy to excavate and compact. Generally moderate; fairly high in sand. or power equipment, such as bulldozer, front-end loader, backhoe, or scraper. Generally easy to compact, except where silt and organic content is high. High. Easily excavated with hand or power equipment. Com- pacts rapidly; compaction density increased by water flooding and vibration. Owing to high water table, deep excavations require pumping. Variable, but generally moderate to low. Generally easily excavated and compacted with power equipment. Excavated with power equip- ment, such as dragline and clamshell bucket, bulldozer, front-end loader, or back- hoe. Sheet piling generally required for excavation. Highly compressible. difficult to compact. One method of placing artificial fill is to push aside surface layers of soft mud to avoid trapping it below the fill. Another method is to place a thin layer of fill on the undisturbed top layer of mud that has been some- what consolidated by dry- ing for a few years. Very low, except sand lenses. Easily excavated. Compacts rapidly; compaction density increased by water flooding and vibration. High. In most places zone of saturation is deep. Variable from low to Easily excavated by hand or high in adjacent power equipment. Scraper beds. Found above with light ripping or no and below water ripping used on large table. grading jobs. Easily compacted. Low to moderate sheared serpentineexcavated readily with power equip- ment. Massive serpentine may require heavy ripping, blasting, or equivalent. Mixtures of massive and sheared serpentine excavated with light to moderate ripping. Generally low because most fills are uncemented and lie near or below the water table. Slopes cut in landslide deposits generally unstable. Although some undisturbed natural slopes of landslide deposits are stable for many years, sliding may be reacti- vated by changes in stress or strength conditions. Easily excavated with hand Clayey material stands in steep or vertical cuts for several months when dry. Subject to severe gullying. Sandy alluvium unstable in steep cuts. Generally unstable and free running, especially on slopes greater than about 30°. Susceptible to wind and rain erosion. Excavation walls more than 2 or 3 ft in height require support. Stands in steep to vertical cuts for several months when dry. Generally unstable and prone to sliding when wet. Gullying severe. Generally unstable; where above water table, has moderate stability at 1:1 cut slope for foundation excava- tions for several months during dry season. Generally unstable and free running. Slopes steeper than 30° generally unstable. Susceptible to wind and rain erosion. Lagging required to support excavation walls more than 2 or 3 ft high. Fair to good, except for silt‘ and clay-free layers, which are unstable in cut slopes greater than 30°. Excavated vertical faces stand for several weeks to several months when dry. Massive serpentine stable in steep or vertical cuts. Cut slopes in sheared serpentine should not exceed 1:1, and embankment height should be limited. Serpentine at and near other bedrock types generally sheared and altered, and slope stability low. Nodules of hard serpentine tend to fall out of sheared matrix. ENGINEERING DEPOSITS engineering properties of map units from KJss through KJm] 97 Shearing strength; Earthquake stability foundation conditions Unit weight (pounds per cubic foot) 1 Possible or reported use Unified soil classification group symbol2 Generally moderate shearing strength, but exceedineg variable depending on composition, method of placement, age, thickness, underlying material, and history following placement, such as ground-water conditions, loading and so forth. In 1906 earthquake the greatest damage to structures was inflicted in areas of artificial fill overlying bay mud and clay along east shore of city. Generally unsuitable for foundations. Poor to fair. Most movement where thick, poorly compacted, and overlying soft bay mud and clay. Fair where thin, well compacted, and overlying firm materials. Low. Moderate to high in sandy alluvium; low where deposits are predominantly clay and silt or high in plant fragments. To safeguard foundations, clay-filled surface and subsurface channels should be adequately drained. Moderate. Moderate to high shearing strength where confined. Susceptible to wave erosion on beach. Uniform sands have lower shearing strength than well-graded sands. Probably moderate. Moderate. Variable. Deposits with relatively high clay content are weak and plastic when wet; sandy and gravelly deposits have moderate to high strength. Low shearing strength. An older bay clay, lying below Colma Formation, is firm and preconsolidated in most places and has moderate shearing strength. Very low; structures erected on artificial fill overlying bay mud and clay severely damaged in earthquakes of 1865, 1868, and 1906. Probably moderate. Moderate to high shearing strength when confined. Moderate to high shearing strength. Used for pile and caisson support. Probably moderate to high. High. Shearing strength high in massive rock but decreases with increasing proportion of shearing and alteration. Thoroughly sheared and altered serpentine has low shearing strength. Veins of soft altered material in hard serpentine may present special problems. See footnotes at end of table, p. 99. 534-039 0 — 74 - 7 Variable within wide limits. Used extensively for construction material and foundation purposes. See Putnam (1947, p. 271—278) for estimating storm wave heights in San Francisco Bay and for planning building locations on artificial fill on edge of bay. Rockfalls possible local source of . Highly variable. pervnous fill, riprap, and so forth. Surface material possible source of 100—114 topsoil for lawns and gardens. Blenlding sand for concrete aggregate; 105—110 Fill. 104—124. Although of very poor quality, bay mud 43—98 (older bay and clay have been used for fill behind mud). part of Embarcadero seawall. Recent shell deposits and underlying clay and mud dredged from bay for manu- facture of cement. Blended with better quality clay to make structural clay ceramic products. May be suitable for making foundry sand. Fair quality fill. Admixed with clay to make foundry sand. Small tonnages used as blending sand in concrete aggregate. 65—102 at surface; 110 compacted. Good quality fill. 105430. Massive and moderately sheared serpentine widely used for fill; highly sheared and altered rock is unsuitable. 78 (sheared and altered) —158 (massive) (See table 8). Variable within. wide limits. Variable. GC to CH. SM, SC, SW, 0L. SP. CL, SM, SC, rarely GC and GM. CL, CH. SP. Mostly SP; some is Some is CL, CH. 98 GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE TABLE 11.—Generalized description of Name and map symbol Weathering, soil development, (pl. 1) Lithology alteration Permeability Workability Slope stability Gabbro and Gray coarse- to fine- Observed weathering depth is less Low, except where Generally requires heavy Variable; high where fresh; diabase grained equigranular than 3 ft. Moderately altered in fractured. ripping or blasting except moderate to low where (gb) . and diabasic igneous places, mostly to prehnite. where altered. altered. rock that occurs as Weathered rock is speckled segregations in brown and orange and ranges serpentine. from crumbly to moderately hard; generally nonswelling. Sandstone Gray tough nonporous Maximum depths of weathering Generally low; Altered rock excavated Fresh or moderately fresh rock (KJss) . fine- to coarse-grained observed. 60 ft; average 30 ft. moderate to high in readily by ripping. Back- stable in vertical cuts. Blocks thick-bedded sandstone Soils well developed locally. with fractured rock. hoe trenching is generally may fall from vertical faces and minor thin—bedded 1—18-ft-thick B horizon of sandy slow; highly altered rock in jointed sandstone. shale. Conglomerate clay. A and B horizons high in trenched rapidly by back- Moderately sheared and beds rare. Pervasive swelling clay minerals hoe. Fresh to moderately fractured sandstone stable in nearly random fractur- montmorillonite and vermiculite, fresh rock generally cut slopes of 55°. Badly ing to %-2 in. blocks. but sand content keeps swelling excavated by heavy rip- altered, sheared, fractured but locally, blocks low to moderate. Partly altered ping; massive rock usually sandstone tends to slump and between major frac- rock, C horizon, is brown or requires blasting. slide, especially when wet. tures are 1 ft or more orange, friable, and nonswelling. in size. Shale and Dark-gray shale inter- Maximum depth of weathering Low, except where Fresh rock excavated by Steep cut slopes tend to rave], thin-bedded bedded with dark-gray observed, 30 ft. Well-developed fractured. moderate to heavy ripping; but are fairly stable for long sandstone fine-grained sandstone. soils with thick clayey B horizon. ' blasting required in some periods except where exten- (KJsh) . Beds generally 2—5 in. A and B horizons of moderate places. Support required sively sheared, fractured, thick; paper-thin to high swelling and (or) for excavation walls and and altered. Sliding likely laminations locally plasticity when wetted. Altered tunnels. Altered rock on bedding dipping in same common. Sheared rock and sheared shale generally of excavated by ripper, front- direction as cut slope. is slaty or reduced to high swelling and (or) plasticity end loader, bulldozer, or soft, friable. and when wetted. equivalent. mashed material enclosing hard nodules. Radiolarian Reddish-brown alternate Slight; cherts whiten from removal Low, except where Bedded chert generally Generally stable in steep cuts, chert and beds of hard chert, of iron and become more broken fractured. excavated by moderate except for minor raveling. shale 1—5 in. thick, and along joints; shale more broken ripping. Massive chert may Sheared and hydrothermally (KJc) . brittle friable shale as than chert. Hydrothermal require blasting. altered zones may slide in much as % in. thick. alteration to orange and white, steep cuts. Dip slopes should Includes thick irregular but durable. rock; pronounced in be cut at lower angle than masses of unstratified some places. especially in fault dip of beds. Commonly chert chert with brecciated zones, to white plastic clay of lies on badly altered green- structure. Some altered little or low swell when wetted. stone and may slide on slip chert is badly fractured surface developed in green- and splintery. stone. Greenstone Greenish-gray aphanitic Maximum depth of weathering Low, except where Altered greenstone can be Fresh and moderately fresh KJ g) . to medium-grained observed, 40 ft. Soil well fractured. excavated by light to rock stable in steep cuts, but volcanic rocks. Pre- dominantly basalt fiows, agglomerates, and tuffs. Pillow lavas, locally interbedded with radiolarian chert, are common. Most natural exposures are reddish-brown soft crumbly altered rock: hard tough unaltered rock limited to deep excavation. Some rock altered to soft clay that swells when wet. Metamorphic Hard fine- to coarse- rocks grained slate, schist, (KJm) . and granofels. Sheared rocks, Hard rocks of the Fran- undifi‘er- ciscan Formation, as enciated much as hundreds of (Ks). feet in diameter, in a soft and crumbly matrix of sheared shale and serpentine. Only moderate to slight soil developed and reddish brown or grayish orange in color, contain- ing iron-rich swelling clay, though swelling and plasticity low to moderate when wetted. Hydrothermal alteration common to clayey material containing halloysite and of low to moderate swelling and plasticity when wetted. Low, except where development observed in fractured. quadrangle. Partly altered rock on Angel Island contains much vermiculite. Moderate to well-developed soil. Low. Hydrothermal alteration common. Swelling and plasticity on wetting is moderate to high in shale and serpentine matrix as well as in soils and hydro- thermally altered materials. moderate ripping. Fresh massive greenstone requires heavy ripping or blasting. Generally excavated by heavy ripping or blasting. Soft material easily excavated by light to moderate ripping; large, hard inclusions generally require heavy ripping or blasting. lava pillows may fall out of weak matrix. Altered rock stable at 1:1 or gentler slopes depending on degree of alteration and fracturing and such local conditions as ground water and height of slope. Steep cut slopes are stable. Dip slopes should be cut lower than angle of schistosity. Generally low. especially when wet (matrix) ; stable in steep cuts where hard rock inclusions are large enough. ENGINEERING DEPOSITS engineering properties of map units—Continued 99 Shearing strength; Earthquake stability foundation conditions Generally high. Generally moderate to high shearing strength. Shearing strength is high except in badly High. shattered and altered rock. Shearing strength high in fresh rock. Foundations on badly sheared, altered rock may require pile support. High in fresh rock. Probably moderate in thoroughly sheared or altered rock. High. Generally moderate to high shearing strength, except where badly altered. High. Shearing strength high in relatively fresh rock, low in altered clayrich rock. High. High shearing strength except where parallel to cleavage or schistosity. Moderate. Matrix has low to moderate shearing strength. Large. rock fragments found in exploratory bonngs may give false impression of sound foundation conditions. Unified soil classification group symbol2 Unit weight (pounds per cubic foot) 1 Possible or reported use Good quality fill. Possible limited source 180—192 (fresh); of concrete aggregate and large size 160 (altered). riprap. Fresh rock suitable for good quality 128—144. fill, road metal, riprap, concrete aggregate. Moderately altered rock suitable for fill. Highly altered rock has been used" for impervious lining of reservoir. Fresh rock and much moderately 127—142. altered rock suitable for fill. Calcined. expanded shale is source of light- weight aggregate. Shale used to make common bricks. Possible use in manufacturing of cement. 166 (fresh) ; 151 (moderately altered) ; 1 10 (badly altered). Widely and successfully used as fill and road metal. Suitability of chert for concrete aggregate questionable (Goldman and Klein, 1959). 113 (thoroughly altered) to 185 (fresh). Moderately altered greenstone with associated chert and shale, known locally as “redrock.” is used as fill and road metal. Fills of badly altered greenstone are prone to sliding on moderate or steep slopes. Relatively fresh rock is possible source of concrete aggregate and riprap. Good quality fill. road metal, large- 169—195. size riprap, concrete aggregate. 78—110 (matrix); Some is CL. CH. 126—170 (rock fragments) . Used extensively for low quality fill. Hard rock inclusions used for good quality fill and large riprap. 1Pounds of dry material per cubic foot of original material. ”Classification used by US. Bureau of Reclamation (1953) and U.S. Army Corps of Engineers (1953). 100 REFERENCES Allen, J. E., 1946, Geology of the San Juan Bautista quadrangle, California: California Div. Mines Bull. 133, p. 9—75. American Society of Civil Engineers, General and Special Committees, 1907, The effects of the San Francisco earth- quake of April 18, 1906, on engineering constructions: Am. Assoc. Civil Engineers Trans, v. 59, no. 1056, p. 208—329. American Society of Civil Engineers, 1932, Subsidence and the foundation problem in San Francisco: Rept. San Francisco Sec., Subsoil Comm., 107 p. Anderson, F. M., 1938, Lower Cretaceous deposits in California and Oregon: Geol. Soc. America Spec. Paper 16, 339 p. Arden, D. D., Jr., 1961, Sediments from borings along the east side of San Francisco Bay: California Univ., Berkeley, Ph.D. thesis, 106 p. Arnal, R. E. and Conomos, T. J ., 1963, Sedimentary and foram- iniferal aspects of southern San Francisco Bay [abs]: Geol. Soc. America Spec. Paper 73, p. 21—22. Ashley, G. H., 1896, The Neocene stratigraphy of the Santa Cruz Mountains of California: California Acad. Sci. Proc., ser. 2, V. 5, p. 273—367. Association of Bay Area Governments, 1968, Regional geology, supplemental report IS—3—The geology of the San Fran- cisco Bay area and its significance in land use planning: Berkeley, Calif, Assoc. Bay Area Govts., 47 p. Bailey, E. H., Blake, M. C., Jr., and Jones, D. L., 1970, On-land Mesozoic ophiolitic oceanic crust in California, in Geologi- cal Survey research 1970: U.S. Geol. Survey Prof. Paper 700—C, p. 070—081. Bailey, E. H., and Irwin, W. P., 1959, K-feldspar content of Jurassic and Cretacedus graywackes of northern Coast Ranges and Sacramento Valley, California: Am. Assoc. Petroleum Geologists Bull., V. 43, no. 12, p. 2797—2809. Bailey, E. H., Irwin, W. P., and Jones, D. L., 1964, Franciscan and related rocks and their significance in the geology of western California: California Div. Mines and Geology Bull. 183, 177 p. Bailey, T. L., 1921, The petrography and origin of the recent sediments along the east side of San Francisco Peninsula: California Univ., Berkeley, M.A. thesis, 175 p. Barnes, Ivan, and O’Neil, J. R., 1969, The relationship between fluids in some fresh alpine-type ultramafics and possible modern serpentinization, western United States: Geol. Soc. America Bull., v. 80, p. 1947—1960. Bartell, M. J ., 1913, Report on the underground water supply of San Francisco County: City and County of San Fran- cisco, Dept. Public Works, M. M. O’Shaughnessy, City Engineer, 157 p. Beal, M. A., and Shepard, F. P., 1956, A use of roundness to determine depositional environments: Jour. Sed. Petrol- ogy, V. 26, no. 1, p. 49—60. Beechey, F. W., 1831, Narrative of a voyage to the Pacific and Beering’s Strait in the years 1825—28; London, H. Colburn and R. Bentley, 2 vols., 742 p. Blake, M. C., Jr., Irwin, W. P., and Coleman, R. G., 1967, Upside-down metamorphic zonation, Blueschist facies, along a regional thrust in California and Oregon, in Geo- logical Survey research 1967: U.S. Geol. Survey Prof. Paper 575—C, p. C1—C9. 1969, Blueschist-fades metamorphism related to regional thrust faulting: Tectonophysics, v. 8, p. 237—246. Blake, W. P., 1857, Geological report, in Lt. R. S. Williamson’s Report of Explorations in California for railroad routes: GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE Exploration and surveys for a railroad route from the Mississippi River to the Pacific Ocean: U.S. 33d Cong, 2d sess., H. EX. Doc. 91, v. 5, 370 p. Bloxam, T. W., 1956, Jadeite-bearing metagraywackes in Cali- fornia: Am. Mineralogist, v. 41, nos. 5—6, p. 488—496. 1960, J adeite-rocks and glaucophane-schists from Angel Island, San Francisco Bay, California: Am. Jour. Sci., v. 258, no. 8, p. 555—573. Bonilla, M. G., 1959, Geologic observations in the epicentral area of the San Francisco earthquake of March 22, 1957: California Div. Mines Spec. Rept. 57, p. 25—37. 1960, Landslides in the San Francisco South quadrangle, California: U.S. Geol. Survey open-file report, 44 p. 1961, The City College fault, San Francisco, California, in Geological Survey research 1961: U.S. Geol. Survey Prof. Paper 424—C, p. C190—Cl92. 1965, Geologic map of the San Francisco South quad- rangle, California: U.S. Geol. Survey open-file map, scale 1220,000. L .9 / F: " N 5* 3% ~ _, 1970, Surface faulting and related effects, chap. 3, in Earthquake engineering: Englewood Cliffs, N.J., Prentice- Hall, Inc., p. 47—74. Bonilla, M. G. and Schlocker, Julius, 1966, Field Trip, San Francisco peninsula, in Bailey, E. H., ed., Geology of northern California: Calif. Div. Mines and Geology Bull. 190, p. 441—452. Borcherdt, R. D., 1970, Effects of local geology on ground motion near San Francisco Bay: Bull. Seismol. Soc. America, v. 60, p. 29—61. Borg, I. Y., 1956, Glaucophane schists and eclogites near Healdsburg, California: Geol. Soc. America Bull., v. 67, no. 12, p. 1563-1583. Bradley, W. C., 1957, Origin of marine-terrace deposits in the Santa Cruz area, California: Geol. Soc. America Bull., v. 68, no. 4, p. 421—444. Bramlette, M. N., 1946, The Monterey formation of California and the origin of its siliceous rocks: U.S. Geol. Survey Prof. Paper 212, 57 p. Brewer, W. H., 1930, Up and down California in 1860—1864: New Haven, Yale Univ. Press, 601 p. Brothers, R. N., 1954, Glaucophane schists from the North Berkeley Hills, California: Am. Jour. Sci., v. 262, no. 10, p. 614—626. Buckland, Sir William, 1839, Geology, in The zoology of Cap- tain Beechey’s voyage in H.M.S. Blossom: London, H. G. Bohn, p. 157—180. Byerly, Perry, 1951, History of earthquakes in the San Fran- cisco Bay area, in Jenkins, 0. P., ed., Geologic guidebook of the San Francisco Bay Counties: California Div. Mines Bull. 154, p. 151—160. California Department of Public Works, Division of Water Resources, 1955, Geology, Appendix H of Report to the Project Authority of the State of California, on feasibility of construction by the State of barriers in the San Fran- cisco Bay system: 62 p. California Department of Water Resources, 1964, Crustal strain and fault movement investigation—Faults and earthquake epicenters in California: California Dept. Water Resources Bull. 116—2, 96 p. Carlson, P. R., Alpha, T. R., and McCulloch, D. S., 1970, The floor of central San Francisco Bay: Calif. Div. Mines and Geology Mineral Inf. Service, v. 23, no. 5, p. 97—107. Carlson, P. R. and McCulloch, D. S., 1970, Bedrock-surface map of central San Francisco Bay: U.S. Geol. Survey open—file map, scale 1 :15,840. REFERENCES Clarke, F. W., 1924, The data of geochemistry: U.S, Geol. Survey Bull. 770, 841 p. Cluff, L. S., and Bolt, B. A., 1969, Risk from earthquakes in the modern urban environment, with special emphasis on the San Francisco Bay area, in Danehy, E. A., ed., Urban environmental geology in the San Francisco Bay region: San Francisco, Calif., Assoc. Eng. Geologists Spec. Pub., p. 25—64. Coleman, R. G., 1956, Jadeite from San Benito County, Cali- fornia: Gems and gemology, v. 8, no. 11, p. 331—334. 1961, Jadeite deposits of the Clear Creek area, New Idria district, San Benito County, California: J our. Petrol- ogy, v. 2, no. 2, p. 209—247. 1965, Composition of jadeitic pyroxene from the Cali- fornia metagraywackes, in Geological Survey research: U.S, Geol. Survey Prof. Paper 525—C, p. C25—C34. 1971, The petrologic and geophysical nature of serpen- tinites: Geol. Soc. America Bull., v. 82, no. 4, p. 897—918. Coleman, R. G., and Lee, D. E., 1962, Metamorphic aragonite in the glaucophane schists of Cazadero, California: Am. J our. Sci., v. 260, p. 577—595. Conomos, T. J ., 1963, Geologic aspects of the Recent sediments of South San Francisco Bay: San Jose State Coll., San Jose, Calif., M.S. thesis, 118 p. Cooper, W. S., 1958, Coastal sand dunes of Oregon and Wash- ington: Geol. Soc. America Mem. 72, 169 p. 1967, Coastal dunes of California: Geol. Soc. America Mem. 104, 131 p. Crandall, Roderic, 1907, The geology of the San Francisco Peninsula: Am. Philos. Soc. Proc. 46, p. 3—58. Curray, J. R., 1965, Late Quaternary history, continental shelves of the United States, in Wright, H. E., Jr., and Frey, D. G., eds., The Quaternary of the United States: Princeton Univ. Press, p. 723—735, Davis, E. F., 1918a, The Franciscan sandstone: California Univ. Pubs., Dept. Geology Bull., v. 11, no. 1, p. 1-44. 1918b, The radiolarian cherts of the Franciscan group: California Univ. Pubs. Dept, Geology Bull., v. 11, no. 3, p. 235—432. Davis, F. F., 1949, San Francisco County: California Div. Mines Bull. 142, p. 105—107. Davis, W. M., 1933, San Francisco Bay—In Middle California and western Nevada: Internat. Geol. Cong., 16th, Wash- ington, D.C., 1933, Guidebook 16, Excursion C—l, p. 18—21. Day, D. T. and Richards, R. H., 1905, Useful minerals in the black sands of the Pacific slope: U.S. Geol. Survey Mineral Resources, V. 44, p. 1175—1258. DeGroot, Henry, 1890, The San Francisco ocean placer—the auriferous beach sands: California State Mining Bur., 10th Ann. Rept., p. 545-547. DeSitter, L. U., 1956, Structural geology: New York, McGraw- Hill, 552 p. de Roever, W. P., 1955, Genesis of jadeite: Am. Jour. Sci., v. 253, no. 5, p. 283—298. Dickert, P. F., 1966, Tertiary phosphatic facies of the Coast Ranges, in Bailey, E. H., ed., Geology of northern Cali- fornia: California Div. Mines and Geology Bull. 190, p. 289—304. Dickinson, W. R., and Grantz, Arthur, eds, 1968, Proceedings of conference on geologic problems of San Andreas fault system: Stanford Univ. Pubs. Geol. Sci., v. 11, 374 p. Duke, C. M., 1958, Effects of ground on destructiveness of large earthquakes: Am. Soc. Civil Engineers Proc., Jour. Soil Mech. and Found. Div., v. 84, no. SM3, paper 1730, 23 p. 101 Eakle, A. S., 1901, Mineralogical notes, with chemical analysis by W. T. Schaller: California Univ. Pubs, Dept. Geology Bull., v. 2, no. 10, p. 315-325. Eckel, E. B., 1970, The Alaska earthquake March 27, 1964: Lessons and conclusions: U.S. Geol. Survey Prof. Paper 546, 57 p. Einstein, H. A., and Krone, R. B., 1961, Estuarial sediment transport patterns: J our. Hydraulics Div., v. 87, no. HY 2, p. 51—59. Evans, Percy, Hayman, R. J., and Majeed, M. A., 1934, The graphical representation of heavy mineral analyses: Geol. Mining Metall. Soc. India Quart. Jour., v. 6, no. 2, p. 27—47. Faust, G. T., Murata, K. J ., and Fahey, J. J ., 1956, Relation of minor element content of serpentines to their geological origin: Geochim. et Cosmochim. Acta, v. 10, nos. 5—6, p. 316—320. Flint, R. F., 1947, Glacial geology and the Pleistocene epoch: New York, John Wiley & Sons, 589 p. Folk, R. L., and Ward, W. C., 1957, Brazos River bar [Texas] —a study in the significance of grain size parameters: Jour. Sed. Petrology, v. 27, no. 1, p. 3—26. Francis, G. H., 1956, The serpentine mass in Glen Urquhart, Inverness-shire, Scotland: Am. Jour. Sci., v. 254, no. 4, p. 201—226. Gabb, W. M., 1869, Cretaceous and Tertiary fossils: California Geol. Survey Paleontology, v. 2, 299 p. Gibson, W. M., 1951, Sand waves in San Francisco Bay: US. Coast and Geod. Survey J our., v. 4, p. 54—58. Gilbert, G. K., 1917, Hydraulic-mining debris in the Sierra Nevada: US. Geol. Survey Prof. Paper 105, 154 p. Gilbert, G. K., Humphrey, R. L., Sewell, J. S., and Frank Soule, 1907, The San Francisco Earthquake and fire of April 18, 1906: US. Geol. Survey Bull. 324, 170 p. Gluskoter, H. J ., 1962, Geology of a portion of western Marin County, California: Calif. Univ., Berkeley, Ph. D. thesis, 184 p. Goddard, E. N., chm. and others, 1948, Rock-color chart: Wash- ington, Natl. Research Council (repub. by Geol. Soc. America, 1951), 6 p. Goldman, H. B., 1967, Geology of San Francisco Bay: San Francisco Bay Conservation and Devel. Comm., 58 p. 1969a, Geology of San Francisco Bay, in Geologic and engineering aspects of San Francisco Bay fill: California Div. Mines and Geol. Spec. Rept. 97, p. 9—29. 1969b, Salt, sand, and shells—Mineral resources of San Francisco Bay, in Geologic and engineering aspects of San Francisco Bay fill: California Div. Mines and Geol. Spec. Rept. 97, p. 31—40. Goldman, H. B., and Klein, I. E., 1959, Franciscan chert in California concrete aggregates: California Div. Mines Spec. Rept. 55, 28 p. Goldsmith, Richard, 1959, Granofels, a new metamorphic rock name: Jour. Geology, v. 67, no. 1, p. 109—110. Gram, Ralph, 1966, The marine geology of the Recent sedi- ments of central San Francisco Bay: San Jose State Coll., San Jose, Calif., M.S. thesis, 132 p. Hall, Timothy, 1965, Late Cenozoic stratigraphy between Mus- sel Rock and Fleischhacker Zoo, San Francisco Peninsula; in INQUA, Internat. Assoc. for Quaternary Research, 7th Cong., United States 1965, Guidebook for Field Conference 1, Northern Great Basin and California; p. 151—158. Harding, R. C., 1969, Landslides—A continuing problem for Bay Area development, in Danehy, E. A., ed., Urban environmental geology in the San Francisco Bay region: 102 San Francisco, Calif. Assoc. Eng. Geologists Spec. Pub., p. 65-74. Hart, E. W., 1966, Shell deposits of southern San Francisco Bay: California Div. Mines and Geology Mineral Inf. Service, v. 19, no. 3, p. 41—45. Hay, O. P., 1927, The Pleistocene of the western region of North America and its vertebrated animals: Carnegie Inst. Wash- ington Pub. 322B, 346 p. Heezen, B. C., and Ewing, W. M., 1955, Orleansville earthquake and turbidity currents: Am. Assoc. Petroleum Geologists Bull., v. 39, no. 12, p. 2505—2514. Heizer, R. F., 1951, Indians of the San Francisco Bay area: California Div. Mines Bull. 154, p. 39—56. Hem, J. D., and Skougstad, M. W., 1960, Coprecipitation effects in solutions containing ferrous, ferric, and cupric ions: U.S. Geol. Survey Water-Supply Paper 1459—E, p. 95—110. Hertlein, L. G., 1956, Cretaceous ammonite of Franciscan group, Marin County, California: Am. Assoc. Petroleum Geologists Bull., v. 40, no. 8, p. 1985—1988. Hess, H. H., 1955, Serpentines, orogeny, and epeirogeny, in Poldervaart, Arie, ed., Crust of the earth—a symposium: Geol. Soc. America Spec. Paper 62, p. 391—407. Hibbard, C. W., Ray, D. E., Savage, D. E., Taylor, D. W., and .Guilday, J. E., 1965, Quaternary mammals of North America, in Wright, H. E., J r., and Frey, D. G., eds., The Quaternary of the United States: Princeton, N.J., Prince- ton Univ. Press, p. 509—525. Hinde, G. J ., 1894, A note on the radiolarian chert from Angel Island and from Buri-Buri Ridge, San Mateo County, California: California Univ. Pubs, Dept. Geology Bull., v. 1, p. 235—240. Hopkins, D. M., 1959, Cenozoic history of the Bering land bridge: Science, v. 129, no. 3362, p. 1519—1528. Hopkins, D. M., MacNeil, F. S. and Leopold, E. B., 1960, The coastal plain at Nome, Alaska—a late Cenozoic type sec- tion for the Bering Strait region, in Chronology and clima- tology of the Quaternary: Internat. Geol. Cong., 21st, Copenhagen, 1960, Proc. pt. 4, p. 46—57. Hoover-Young San Francisco Bay Bridge Commission, 1930, San Francisco Bay Bridge—Report of the Commission to the President of the United States and the Governor of the State of California: Sacramento, California State Printing Office, 234 p. Howard, A. D., 1951, Development of the landscape of the San Francisco Bay Counties, in Jenkins, 0. P., ed., Geologic guidebook of the San Francisco Bay Counties: California Div. Mines Bull. 154, p. 95—106. Huber, N. K., 1958, The environmental control of sedimentary iron minerals: Econ. Geology, v. 53, no. 2, p. 123—140. Hutton, C. 0., 1940, Metamorphism in the Lake Wakatipu region, western Otago, New Zealand: New Zealand Dept. Sci. and Indus. Research, Geol. Mem. 5, 84 p. 1950, Studies of heavy detrital minerals: Geol. Soc. America Bull., v. 61, p. 635—716. 1959, Mineralogy of beach sands between Half Moon and Monterey Bays, California: California Div. Mines Spec. Rept. 59, 32 p. Inman, D. L., 1952, Measures for describing the size distribu- tion of sediments: Jour. Sed. Petrology, v. 22, no. 3, p. 125—145. Inman, D. L., and Chamberlain, T. K., 1955, Particle-size dis- tribution in nearshore sediments, in Finding ancient shore- lines: Soc. Econ. Paleontologists and Mineralogists Spec. Pub. no. 3, p. 106—129. GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE Institute of Transportation and Traffic Engineering, University of California, and the US. Geological Survey, 1951, Selected logs of borings in the City and County of San Francisco, California: The Inst. of Transp. and Traffic Eng., California Univ. Inf. Circ. 11, 240 p. Irwin, W. P., 1957, Franciscan group in Coast Ranges and its equivalents in Sacramento Valley, California: Am. Assoc. Petroleum Geologists Bull., v. 41, no. 10, p. 2284—2297. 1964, Late Mesozoic orogenies in the ultramafic belts of northwestern California and southwestern Oregon: US. Geol. Survey Prof. Paper 501—C, p. C1—C9. Johannsen, Albert, 1932, A descriptive petrography of the igneous rocks; v. 2, The quartz-bearing rocks: Chicago, Univ. Chicago Press, 428 p. Keller, W. D., 1956, Clay minerals as influenced by environ- ments of their formation: Am. Assoc. Petroleum Geolo- gists Bull., v. 40, no. 11, p. 2689—2710. Knopf, Adolph, 1960, Analysis of some recent geosynclinal theory (Bradley volume): Am, Jour. Sci., v. 258—A, p. 126—136. Kopstein, F. P. H. W., 1954, Graded bedding of the Harlech Dome: Rijksuniversiteit te Groningen, Ph. D. dissertation, 97 p. Krauskopf, K. B., 1955, Sedimentary deposits of rare metals: Econ. Geology 50th ann. v., pt. 1, p. 411—463. 1956, Dissolution and precipitation of silica at low temperatures: Geochim. et Cosmochim. Acta, v. 10, p. 1—26. 1967, Introduction to geochemistry: New York, McGraw- Hill Book Co., 721 p. Kroeber, A. L., 1911, Shellmounds at San Francisco and San Mateo: Records of the past, v. 10, p. 227—228. Krone, R. B., 1962, Flume studies of the transport of sediment in estuarial shoaling processes—final report: Hydraulic Eng. Lab. and Sanitary Eng. Research Lab., California Univ., Berkeley, 110 p. 1963, A study of rheologic properties of estuarial sedi- ments—final report: Hydraulic Eng. Lab. and Sanitary Eng. Research Lab., California Univ., Berkeley, SERL Rept. 63—8, 91 p. Kuenen, P. H., 1950, Marine geology: New York, John Wiley & Sons, 568 p. 1951, Properties of turbidity currents of high density, in Turbidity currents and the transportation of coarse sedi- ments to deep water—a symposium: Soc. Econ. Paleontolo- gists and Mineralogists Spec. Pub., no. 2, p. 14-33. 1955, Sea level and crustal warping, in Poldervaart, Arie, ed., Crust of the earth—a symposium: Geol. Soc. America Spec. Paper 62, p. 193—203. Kuno, Hisashi, 1954, Study of orthopyroxenes from volcanic rocks: Am. Mineralogist, v. 39, nos. 1—2, p. 30—46. Kuno, Hisashi, Yamasaki, Kazuo, Iida, Chuzo, and Nagashima, Kozo, 1957, Differentiation of Hawaiian magmas: Japa- nese J our. Geology and Geography, v. 28, no. 4, p. 179—218. Kiipper, Klaus, 1956, Upper Cretaceous pelagic Foraminifera from the “Antelope shale,” Glenn and Colusa Counties, California: Cushman Found. Foram. Research Contr., v. 7, pt. 2, p. 40—47. Kvenvolden, K. A., 1962, Normal paraffin hydrocarbons in sediments from San Francisco Bay, California: Am. Assoc. Petroleum Geologists Bull., v. 46, no. 9, p. 1643—1652. Langston, R. B., Trask, P. D., and Pask, J. A., 1958, Effect of mineral composition on strength of central California sedi- ments: California Jour. Mines and Geology, v. 54, no. 2, p. 215—235. REFERENCES Lawson, A. C., 1895, Sketch of the geology of the San Francisco Peninsula: U.S. Geol. Survey 15th Ann. Rept., p. 399-476. 1914, San Francisco Folio: U.S. Geol. Survey Atlas Folio 193, 24 p. Lawson, A. C. and others, 1908, The California earthquake of April 18, 1906, Report of the State Earthquake Investiga- tion Commission: Carnegie Inst. Washington Pub. 87, v. 1, pt. 1, 254 p.; pt. 2, p. 255—451; atlas, 25 maps and seismo- grams. Lee, C. H., 1953, Building foundations in San Francisco: Am. Soc. Civil Eng. Proc., v. 79, separate 326, 32 p. Lee, C. H., and Praszker, Michael, 1969, Bay mud develop- ments and related structural foundations, in Geologic and engineering aspects of San Francisco Bay fill: California Div. Mines and Geol. Spec. Rept. 97, p. 41—85. Lincoln, A., 1969, The Beechey expedition visits San Fran- cisco: Pacific Discovery, v. 22, no. 1, p. 1—8. Louderback, G. D., 1939, San Francisco Bay sediments: Pacific Sci. Cong., 6th, Berkeley, Calif. 1939, Proc., v. 2, p. 783—793. 1947, Central California earthquakes of the 1830’s: Seismol. Soc. America Bull., v. 37, no. 1, p. 33-74. 1951, Geologic history of San Francisco Bay: California Div. Mines Bull. 154, p. 75—94. McCulloch, D. S., Taylor, D. W., and Rubin, Meyer, 1965, Stratigraphy, non-marine mollusks, and radiometric dates from Quaternary deposits in the Kotzebue Sound area, western Alaska: J our. Geol., v. 73, no. 3, p. 442—453. MacDonald, G. A., and Katsura, Takashi, 1961, Lava of the 1959 eruption in Kilauea Iki: Pacific Sci., v. 15, no. 3, p. 358—369. McKee, E. D., 1953, Report on studies of stratification in mod— ern sediments and in laboratory experiments: [U.S.] Office of Naval Research Proj. Nonr 164(00), NR 081 123, 61 p. [1953?]. Martin, Bruce, 1916, The Pliocene of middle and northern California: California Univ. Pubs. Dept. Geology Bull., v.9, p. 215—259. Meade, R. H., 1967, Petrology of sediments underlying areas of land subsidence in central California: US. Prof. Paper 497—0, 83 p. Means, K. D., 1965, Sediments and foraminifera of Richardson Bay, California: Univ. Southern California, Los Angeles, M.A. thesis, 80 p. Miller, R. C., Ramage, W. D., and Lazier, E. L., 1928, A study of physical and chemical conditions in San Francisco Bay especially in relation to the tides: California Univ. Pubs. Zoology, v. 31, no. 11, p. 201—267. Milliman, J. D., and Emery, K. 0., 1968, Sea levels during the past 35,000 years: Science, v. 162, p. 1121—1123. Mitchell, J. K., 1963, Engineering properties and problems of the San Francisco Bay mud: California Div. Mines and Geology Spec, Rept. 82, p. 25—32. Moore, D. B., 1965, Recent coastal sediments, Double Point to Point San Pedro, California: California Univ., Hydraulic Eng. Lab. Tech. Report HEL 2—14, 86 p. Moore, D. G., and Shumway, George, 1959, Sediment thickness and physical properties: Pigeon Point shelf, California: J our. Geophys. Research, v. 64, no. 3, p. 367—374. Mumpton, F. A., 1965, Coalingite, a new mineral from the New Idria serpentinite, Fresno and San Benito Counties, Cali- fornia: Am. Mineralogist, v. 50, no. 11—12, p. 1893—1913. Murdoch, Joseph, and Webb, R. W., 1956, Minerals of Cali- fornia: California Div. Mines Bull. 173, 452 p. 103 Nelson, N. C., 1909, Shellmounds of the San Francisco Bay region: California Univ. Pubs. Am. Archaeology and Ethnology, v. 7, no. 4, p. 310—348. Packard, E. L., 1918, Molluscan fauna from San Francisco Bay: California Univ. Pubs. Zoology, v. 14, no. 2, p. 199—452. Page, N. J, 1968, Serpentinization in a sheared serpentinite lens, Tiburon Peninsula, Calif., in Geological Survey research 1968: US. Geol. Survey Prof. Paper 600—B, p. B21—B28. Page, N. J, and Coleman, R. G., 1967, Serpentine-mineral analyses and physical properties, in Geological Survey research 1967: US. Geol. Survey Prof. Paper 575—B, p. B103—B107. Palache, Charles, 1894, The lherzolite serpentine and associated rocks of the Potrero, San Francisco: California Univ. Pubs, Dept. Geology Dept., v. 1, no. 5, p. 161—179. Palache, Charles, Berman, Harry, and Frondel, Clifford, 1951, The system of mineralogy of James Dwight Dana and Edward Salisbury Dana, Yale University, 1837—1892; v. 2, Halides, nitrates, borates, carbonates, sulfates, phosphates, arsenates, tungstates, molybdates, etc. [7th ed.]: New York, John Wiley & Sons, 1124 p. Patton, C. L., 1956, Climatology of summer fogs in the San Francisco Bay area: California Univ. Pubs. Geography, v. 10, no. 3, p. 113—200. Peabody, F. E., 1945, An occurrence of late Pleistocene mam- mals in the San Francisco Peninsula: J our. Paleontology, v. 19, no. 1, p. 60—63. Pessagno, E. A., Jr., 1970, The biostratigraphic potential of Radiolaria in the Upper Cretaceous portion of the Great Valley sequence, California Coast Ranges [abs.]: Geol. Soc. America Abstracts with program, 66th Ann. Mtg., Cordilleran Sec., Hayward, Calif., v. 2, no. 2, p. 130. Pettijohn, F. J ., 1949, Sedimentary rocks: New York, Harper and Brothers, 526 p. Phoenix, John (pseudonym of George H. Derby), 1869, Phoe- nixiana; or, sketches and burlesques [12th ed.]: New York, D. Appleton and Co., 274 p. Poldervaart, Arie, 1955, Chemistry of the earth’s crust, in Poldervaart, Arie, ed., Crust of the earth—a symposium: Geol. Soc. America Spec. Paper 62, p. 119-144. Powers, M. C., 1953, A new roundness scale for sedimentary particles: Jour. Sed. Petrology, v. 23, no. 2, p. 117—119. Putnam, J. A., 1947 , Estimating storm-wave conditions in San Francisco Bay: Trans. Am. Geophys. Union, v. 28, no. 2, p. 271—278. Quaide, W. L., 1958, Clay minerals from salt concentration ponds [California]: Am. Jour. Sci., v. 256, no. 6, p. 431—437. Quinterno, P. J ., 1968, Distribution of Recent Foraminifera in central and south San Francisco Bay: San Jose State Coll., San Jose, Calif., M.S. thesis, 83 p. Radbruch, D. H., 1957, Areal and engineering geology of the Oakland West quadrangle, California: US. Geol. Survey Misc. Geol. Inv. Map L239. 1967, Approximate location of fault traces and historic surface ruptures within the Hayward fault zone between San Pablo and Warm Springs, California: US. Geol. Sur- vey Misc. Geol. Inv. Map I—522, scale 1:62,500. Radbruch, D. H., and Schlocker, Julius, 1958, Engineering geology of Islais Creek Basin, San Francisco, California: US. Geol. Survey Misc. Geol. Inv. Map I—264. Ransome, F. L., 1893, The eruptive rocks of Point Bonita. Cali- fornia Univ. Pubs., Dept. Geology Bull., v. 1, no. 3, p. 71—114. 104 1894, The geology of Angel Island: California Univ. Pubs, Dept. Geology Bull., v. 1, no. 7, p. 193—234. Reese, J. I., 1965, Sedimentation of the Suisun Bay complex: San Jose State Coll., San Jose, Calif., M.S. thesis, 109 p. Reidel, W. R., and Schlocker, Julius, 1956, Radiolaria from the Francisco group, Belmont, California: Micropaleontology, v. 2, no. 4, p. 357—360. Richards, H. G., and Thurber, D. L., 1966, Pleistocene age determinations from California and Oregon: Science, v. 152, no. 3725, p. 1091—1092. Roberts, D. V. and Darragh, R. D., 1963, Areal fill settlements and building foundation behavior at the San Francisco airport: Amer. Soc. for Testing Materials Spec. Technical Pub. 322, p. 211—230. Robinson, G. D., 1956, Geology of the Hayward quadrangle, California: US. Geol. Survey Geol. Quad. Map GQ—88. Roller, J. C., Eaton, J. P., and Lee, W. H. K., 1968, Summary of microearthquakes along the San Andreas fault system in central California: US. Geol. Survey open-file reports, Jan. l—Mar. 31, 1968, 24 p.; April 1—June 30, 1968, 25 p.; July l—Sept. 30, 1968, 22 p.; Oct. 1—Dec. 31, 1968, 20 p. Rose, R. L., 1958, Pre-Tertiary stratigraphy near Petaluma, California [abs]: Geol. Soc. America Bull., v. 69, no. 12, pt. 2, p. 1703—1704. San Francisco Bay Conservation and Development Commis- sion, 1967, Fill—three reports on aspects of fill in San Francisco Bay: Bay mud developments and related struc- tural foundations, by C. H. Lee and M. Praszker, 38 p.; Seismic problems in the use of fills in San Francisco Bay, by H. B. Seed, 43 p.; Seismic risk to buildings and struc- tures on filled lands in San Francisco Bay, by K. V. Stein- brugge, 44 p. Sanders, J. E., and Swinchatt, J. P., 1957, Deep-sea origin of radiolarian cherts [abs.]: Geol. Soc. America Bull., v. 68, no. 12, pt. 2, p. 1791. Savage, D. E., 1951, Late Cenozoic vertebrates of the San Francisco Bay region: California Univ. Pubs, Dept. Geol. Sci. Bull., v. 28, no. 10, p. 215—314. Schlocker, Julius, 1960, Rodingite from Angel Island, San Francisco Bay, California, in Short papers in the geological sciences: U.S. Geol. Survey Prof. Paper 400—B, p. 311—312. 1969, The geology of the San Francisco Bay area and its significance in land use planning, in Danehy, E. H., ed, Urban environmental geology in the San Francisco Bay region: San Francisco, Calif, Assoc. Eng. Geologists Spec. Pub., p. 7-24. 1970, Generalized geologic map of the San Francisco Bay area, California: US. Geol. Survey open-file report, scale 1:500,000. Schlocker, Julius, Bonilla, M. G., and Imlay, R. W., 1954, Am- monite indicates Cretaceous age for part of Franciscan group in San Francisco Bay area, California: Am. Assoc. Petroleum Geologists Bull., v. 38, no. 11, p. 2372—2381. Schlocker, Julius, Bonilla, M. G., and Radbruch, D. H., 1958, Geology of the San Francisco North quadrangle, Cali- fornia: U.S. Geol. Survey Misc. Geol. Inv. Map I—272, scale 1:24.000. Scott, Mel, 1959, The San Francisco Bay area—A metropolis in perspective: California Univ. Press, Berkeley and Los Angeles, Calif., 333 p. Seed, H. B., 1969, Seismic problems in the use of fills in San Francisco Bay, in Geologic and engineering aspects of San Francisco Bay fill: California Div. Mines and Geology Spec. Rept. 97, p. 87—99. GEOLOGY OF THE SAN FRANCISCO NORTH QUADRANGLE 1970, Soil problems and soil behavior, in Earthquake engineering: Englewood Cliffs, NJ, Prentice-Hall, Inc., p. 227—251. Selleck, R. E., Pearson, E. A., Glenne, Bard, and Storrs, P. N., 1966, Physical and hydrological characteristics of San Francisco Bay, v. 4 of A comprehensive study of San Fran- cisco Bay: Sanitary Eng. Research Lab., Coll. Eng. and School Public Health, California Univ., Berkeley, SERL Rept. 65—10, 99 p. Shepard, F. P., 1951, Transportation of sand into deep water, in Turbidity currents and the transportation of coarse sediments to deep water—a symposium: Soc. Econ. Pale- ontologists and Mineralogists Spec. Pub. 21, p. 53—65. 1952, Revised nomenclature for depositional coastal fea- tures: Am. Assoc. Petroleum Geologists Bull., v. 36, no. 10, p. 1902—1912. Slater, R. A., 1965, Sedimentary environments in Suisun Bay, California: Univ. Southern California, Los Angeles, M.A. thesis, 104 p. Smith, B. J., 1963, Sedimentation in the San Francisco Bay system: Federal Interagency Sedimentation Conf., Sub- committee Sedimentation, ICWR, Jackson, Miss, 58 p. 1966, Sedimentation aspects of San Francisco Bay: San Francisco Bay Conserv. and Devel. Comm., 48 p. Smith, D. D., 1960, The geomorphology of part of the San Francisco Peninsula, California: Stanford Univ., Stanford, Calif., Ph. D, thesis, 359 p. Sriramadas, Aluru, 1957, Diagrams for the correlation of unit cell edges and refractive indices with the chemical compo- sition of garnets: Am. Mineralogist, v. 72, no. 3 and 4, p. 294—298. Steinbrugge, K. V., 1968, Earthquake hazard in the San Fran- cisco Bay Area; A continuing problem in public policy: California Univ., Berkeley, Inst. Governmental Studies, 80 p. 1969, Seismic risk to buildings and structures on filled lands in San Francisco Bay, in Geologic and engineering aspects of San Francisco Bay fill: California Div. Mines and Geology Spec. Rept. 97, p. 101—115. Stewart, H. B., Jr., 1958, Sedimentary reflections of deposi- tional environment in San Miguel Lagoon, Baja California, Mexico: Am. Assoc. Petroleum Geologists Bull., v. 42, no. ' 11, p. 2567—2618. Stewart, R. B., 1930, Gabb’s California Cretaceous and Tertiary type lamellibranchs: Philadelphia Acad. Nat. Sci. Spec. Pub. 3, 314 p. Stock, Chester, 1925, Cenozoic gravigrade edentates of western North America, with special reference to the Pleistocene Megalonychinae and Mylodontidae of Rancho La Brea [California]: Carnegie Inst. Washington Pub. 331, 206 p. Storrs, P. N., Selleck, R. E., and Pearson, E. A., 1965, North, central, and lower San Francisco Bay areas—Appendixes to a comprehensive study of San Francisco Bay 1963—1964: Sanitary Eng. Research Lab., Coll. Eng, and School Public Health, California Univ., Berkeley, SERL Rept. 65—2, 319 p. Story, J. A., Wessels, V. E., Wolfe, J. A., 1966, Radiocarbon dating of Recent sediments in San Francisco Bay: Cali- fornia Div. Mines and Geology Mineral Inf. Service, v. 19, no. 3, p. 47—50. Sumner, F. B., Louderback, G. D., Schmitt, W. L., and Johns- ton, E. C., 1914, A report upon the physical conditions in San Francisco Bay: California Univ. Pubs. Zoology, v. 14, no. 1, p. 1—198. REFERENCES Sverdrup, H. U., Johnson, M. W., and Fleming, R. H., 1946, The oceans, their physics, chemistry, and general biology: New York, Prentice-Hall, Inc., 1087 p. Taliaferro, N. L., 1933, The relation of volcanism to diatoma- ceous and associated siliceous sediments: California Univ. Pubs., Dept. Geol. Sci., Bull., v. 23, no. 1, p. 1—56. 1934, Contraction phenomena in cherts: Geol. America Bull., v. 45, no. 2, p. 189~232. 1943, Franciscan-Knoxville problem: Am. Assoc. Petrol- eum Geologists Bull., v. 27, no. 2, p. 109—219. Taliaferro, N. L., and Hudson, F. S., 1943, Genesis of the man- ganese deposits of the Coast Ranges of California: Cali- fornia Div. Mines Bull. 125, p. 217—275. Taylor, S. G., J r., 1957, Gravity investigations of the southern San Francisco Bay area, California: Stanford Univ., Stan- ford, Calif, Ph. D. thesis, 105 p. Thompson, W. 0., 1937, Original structures of beaches, bars, and dunes: Geol. Soc. America Bull., v. 48, no. 6, p. 723—751. Tocher, Don, 1959, Seismic history of the San Francisco region: California Div. Mines Spec. Rept. 57, p. 39—48, Townley, S. D. and Allen, M. W., 1939, Descriptive catalogue of earthquakes of the Pacific Coast of the United States, 1769 to 1928: Seismological Soc. America Bull., v. 29, p. 1—297. Trask, P. D., 1956, Geology of some American estuarine har- bors: Am. Soc. Civil Engineers Proc., Jour. Hydraulics Div., v. 82, no. HY 2, Proc. Paper 956, 18 p. 1959, Beaches near San Francisco, California: Beach Erosion Board, Corps of Engineers, US. Army, Tech. Memo. 110, 89 p. Trask, P. D., and Rolston, J. W., 1951, Engineering geology of San Francisco Bay, California: Geol. Soc, America Bull., v. 62, no. 9, p. 1079—1109. Treasher, R. C., 1959, Engineering geologic conditions, San Francisco Bay studies [abs.:] California Assoc. Eng. Geol- ogists, 2d ann. mtg, p. 13. 1963, Geology of the sedimentary deposits in San Fran- cisco Bay, California: California Div. Mines and Geology Spec. Rept. 82, p. 11—24. Turner, F. J ., and Verhoogen, Jean, 1951, Igneous and meta- morphic petrology [1st ed.]: New York, McGraw-Hill, 602 p. 1960, Igneous and metamorphic petrology [2d ed.] : New York, McGraw-Hill, 694 p. US. Army Corps of Engineers, 1953, The unified soil classifi- cation system: Waterways Expt. Sta. Tech. Memo. 3—357, v. 1, 30 p., 9 pls. and v.2, 11 p.,1 pl. US. Army Engineer District, San Francisco, Corps of Engi- neers, 1963, Technical report of barriers—a part of the comprehensive survey of San Francisco Bay and tribu- taries, California; Appendix E, Barrier plans—geology, soils and construction materials: San Francisco, Calif, Soc. 144 p. US. Bureau of Reclamation, 1953, Unified soil classification system: 26 p. US. Coast and Geodetic Survey, 1941, Nautical chart, United States, West Coast, California, Point Sur to San Francisco [7th ed.] : Chart 5402. 1943, United States Coast Pilot, Pacific Coast, [6th ed.]: Serial No. 649, 415 p. 105 -1959, Nautical chart, United States, West Coast, Cali- fornia, San Francisco Bay, Candlestick Point to Angel Island [18th ed.] : Chart 5535. US. Geological Survey, 1966, Aeromagnetic map of the San Francisco Bay area, California: US. Geol. Survey open- file report, scale 1: 50,000. US. Office of Area Development, 1959, Future development of the San Francisco Bay area 1960—2020: Washington, US. Govt. Printing Office, 94 p. VanderHoof, V. L., 1951, History of geologic investigation in the bay region, in Jenkins, 0. P., ed., Geologic guidebook of the San Francisco Bay Counties: California Div. Mines Bull. 154, p. 109—116. Varnes, D. J ., 1958, Landslide types and processes, chap. 3, in Eckel, E. B., ed., Landslides and engineering practice: Natl. Research Council, Highway Research Board Spec. Rept. 29, p. 20—47. Weaver, C. E., 1949a, Geology of the Coast Ranges immediately north of the San Francisco Bay region, California: Geol. Soc. America Mem. 35, 242 p. 1949b, Geology and mineral deposits of an area north of San Francisco Bay, California: California Div. Mines Bull. 149, 135 p. Weinstein, R. A., 1967, Gold rush daguerreotypes: The Am. West, v. 4, no. 3, p. 33—39, 71—72. White, D. E., Brannock, W. W., and Murata, K. J., 1956, Silica in hot-spring waters: Geochim. et Cosmochim. Acta, v. 10, p. 27—59. Whitney, J. D., 1865, Geology, volume 1, Report of progress and synopsis of the field work from 1860 to 1864: California Geol. Survey, 498 p. Whittaker, E. J. W., and Zussman, J ., 1956, The characteriza- tion of serpentine minerals by X-ray diffraction: Min- eralog. Mag, v. 31, no. 233, p. 107—126. Wigginton, W. B., 1969, Bay mud—why does it have such a bad reputation?, in Danehy, E. A., and Harding, R. C.. eds, Urban environmental geology in the San Francisco Bay region: Sacramento, Calif, Assoc. Eng. Geologists, p. 75—81. Williams, Howel, Turner, F. J ., and Gilbert, C. M., 1954, Petrography: San Francisco, Freeman and Company, 406 p. Winchell, A. N., and Winchell, Horace, 1951, Elements of opti- cal mineralogy—an introduction to microscopic petrog- raphy [4th ed.] : New York, John Wiley & Sons, Inc., 551 p. Wood, H. 0., 1908, Distribution of apparent intensity in San Francisco, in the California earthquake of April 18, 1906: Rept. State Earthquake Inv. Comm., Carnegie Inst. Wash- ington Pub. 87, v. 1, pt. 1, p. 220—245. 1933, “Apparent” intensity and surface geology: Nat. Research Council Bull. 90, . 67—82. Woodring, W. P., and Bramlette, M. N., 1950, Geology and paleontology of the Santa Maria district, California: US. Geol. Survey Prof. Paper 222, 185 p. Yancey, T. E., and Lee, J. W., 1972, Major heavy mineral assemblages and heavy mineral provinces of the central California Coast region: Geol. Soc. America Bull., v. 83, no. 7, p. 2099—2104. Yoder, H. S., 1950, The jadeite problem: Am. Jour. Sci., v. 248, no. 4, p. 225—248. A Page Age, beach deposits ............................................ 75 chert, radiolarian ................ 48 elastic sedimentary rocks 27 clay ........................... 84 Colma Formation 7.9 dune sand .............. 80 Franciscan Formation 9 gabbro 66 mud ....... 34 ravine fill 81 serpentine .4 65 shale, radiolarian 48 slope debris ....... 81 surficial deposits .. 66 Agglomerate .......................... Alameda Creek, sediments .. Alameda Formation, obsolete name . 72 Alamo Square 3 serpentine .. 56 Alcatraz Island, elastic sedimentary rocks .................. 26 fossil locality .. 27 metamorphosed sedimentary rocks . 21 sand bars .................................................... 82 sandstone detrital grains 15 sandstone matrix .............. 14 sediment deposition area 82 structure ............. 93 Allen, J. E., quoted . 74 Alluvium Alteration, hydrothermal, greenstone .. 29. 38, 36 hydrothermal, radiolarian chert and shale sandstone serpentine . shale Angel Island .. beach deposits chert, radiolarian ............ elastic sedimentary rocks Colma Formation .. conglomerate .. dune sand ....... Franciscan Formation . gabbro ........... .. greenstone 29, 30, 32. 33, .95 metamorphic rocks .. 50, 52, 53, 54, 55 metamorphosed sedimentary rocks... 20, 21, 65, 66 22 sandstone detrital grains 15 sandstone matrix ........... 14 sediment deposition area 82 serpentine ................ 51. 56, 57, 58, 60. 64, 65 antigoritic ..... 59 shale, radiolarian .69 structure ................. 93 Arkose ...................... 14 Aromas Red Sands . 74 Artificial fill ..... 66, 76, 81, 88, 94 Bakers Beach ............ 3 beach deposits 76 elastic sedimentary rocks 24 INDEX [Italic page numbers indicate major references] Page Bakers Beach—Continued dune sand .......................... Franciscan Formation metamorphic rocks sandstone detrital grains .. serpentine .. structure Barlow, I. H.. analyst . Basalt ............................ .. 92 . 19, 22, 35, 43 Bastite .............. 58 Beach deposits 74, 76 Bedding .............. 39, 67. 92 sandstone 11 Bedrock .. Belvedere Island ...... .. 3 beach deposits _. 76 chert, radiolarian 49 elastic sedimentary rocks ., 26 Colma Formation ................ 70 conglomerate . ............... 20 greenstone ..... . 29, 30, 35, 36 metamorphic rocks 50, 53, 54, 55 metamorphosed sedimentary rocks ...... 20 sediment deposition area ...................... 82 serpentine .. . 56, 61 antigoritlc 59 shale, radiolarian 49 Berkeley Hills, metamorphic rocks .. 52 structure ., 90 Black Paint . 3 Bloxam, T. W., quoted . 22 Blunt Point, artificial fill .. 90 elastic sedimentary rocks 26 greenstone .................. 29 metamorphic rocks ., 53, 54 metamorphosed sedimentary rocks .. 21 Bolinas Ridge, structure ................................ 90 Bonita Sandstone, obsolete name .................. 10 Botts, S. D., analyst ........ Bramlette, M. N., quoted . Breccia .......................... Brewer, W. H., quoted .. 19, 22. 35, 43 48 Broadway Tunnel area, Colma Formation .. 67 shale ............................................................ 20 Buena Vista Park, elastic sedimentary rocks 24 dune sand ........... 78 greenstone ....................................... 34 Buri-Buri Ridge, radiolarian chert 48 C Cahil Sandstone, obsolete name 10. 25, 28 Calaveras fault zone 93 California Current ..... 5 Campbell Point, elastic sedimentary rocks 26 greenstone . 32, 36 metamorphosed sedimentary rocks 22 sediment deposition area ....................... 82 Carbonaceous matter .................. 15, 19, 45, 69, 83 Carquinez Strait, sediments 82 Chert ..................... .. 27, 86 radiolarian .. 9, 24, 26, 33, 86, 4.9, 76, 81, 87, 92. 93 origin ................ M' Chert grains. sandstone 15 China Basin, artificial fill 88. 90 Page Chloe, Gillison, analyst .. . 19, 22, 35, 43 City College fault ........ . 24, 56, 92 City College shear zone . 50, 55, 64 Clay 66, 82, 83 Clifi‘ House area, elastic sedimentary rocks ........ 24 Colma Formation .. 70 Franciscan Formation .. 9 structure . 92 Climate ........... 5 Coast Miwok Indians ......... 6 Coast Range thrust fault . 9 Coast Ranges ........................ 3, 26, 28, 56 Colluvium .............. 66, 90 Colma Formation . 66, ‘77, 81, 86 Conglomerate 20. 26 Corinthian Island 3 elastic sedimentary rocks .. .. 26 metamorphosed sedimentary rocks ...... 21 Corona Heights, radiolarian chert and shale .............. sandstone detrital grains Correlation, clay .......................... Colma Formation . mud .................... Costanoan Indians Crocidolite ................ D, E Debris flows ....... 87 Deformation, shale .. 20 Depositional environment, elastic sedimentary rocks .. Colma Formation ........ Detrital grains, sandstone Detrital source, elastic sedimentary rocks ............................................ 27 Colma Formation Dikes ......... Dune sand 3, 66, '71, 75. 78, 90 Earthfiows ...... 87 , Earthquakes Elmore, Paul, analyst .. 19. 22, 35, 43 Engineering geology 94 Engineering properties, mud and clay 84 Exploration ................................................ 6' F Faults .................................................................. 90 Ferry Building area, Colma Formation ...... 70, 72, 74 Fieldwork .. 7, 8 Flows .. 29, 36 Folds 39 Foliation .................................................... 21, 51, 52 Fort Baker, elastic sedimentary rocks 26 greenstone pillows .4 30 radiolarian chert 39 Fort Mason, artificial fill . 90 Colma Formation 70 seismic activity . 93 Fort Point .............. 3 beach deposits . 76 elastic sedimentary rocks . . 95 dune sand .................................................... 80 107 108 Page Fort Point—Continued Franciscan Formation .......... 9 radiolarian chert and shale .. .. 49 serpentine . 56, 62, 63, 65 structure .. ' 92, 93 Fort Pointhunters Point shear zone ,. 50, 55, 64, 92 Fort Point Rock, beach deposits .................. 76 Fort Scott, serpentine ..................................... 56 Fossils: Bison 74 Bryozoa 83 chert ................................... 44 elastic sedimentary rocks 27 clay ......................... 84 Colma Formation . ., 73 Diatoms .......... 47, 73, 83, 85 meilleiceras _ . 12, 27, 49, 65 sandstone detrital grains ................ 15 echinoids 77 Foraminifera .1 83 Franciscan Formation 9 Inoccramus ellioti 28 Juniperus califmica 73 Lucina alcatrazia ....... 28 Mantelliceras ........................... 27 Marthasterites tribrachiatus . 28 mollusks . . 26, 74, 77, 83, 85 mud ..... 84 Mylodon ‘73 Paladmete perforate . 27 plant fragments ......... 83 Radiolaria ................ 15, 20, 26,28, 43, 44, 45, 48, 83 shale 45 sloth fragment . 73 sponge spicules . 43, 45, 48, 73, 83 Fractures ....................................... 30. 40 Franciscan Formation . 9, 55, 88, 90, 94 structure 92 Franciscan Group, obsolete name . _ 10, 28 G Gabbro ................................................................ 65 Glen Ellen Formation ...................................... 74 Gold 77 Golden Gate ............................... 3 elastic sedimentary rocks .1 25 landslide deposits ............ 86 sandstone detrital grains 15 seismic activity . 93 serpentine ..... 56 Golden Gate channel beach deposits formation ....... greenstone sandstone detrital grains . 15 sediment deposition area . 82 sediments .......................... 81 structure ....... 92, 93 Golden Gate Park . 3 artificial fill 90 dune sand 80 radiolarian chert and shale _ 49 Grand View Park, dune sand ..... 78 Granitic rock grains, sandstone 15 Granitic rocks .................................................... 27 Granodiorite, as Franciscan Formation source 27 Granofels .................. . 50 Graywacke ..... . 9, 14, 15, 92 Great Valley 9, 72 Great Valley sequence . 92 Greenstono .................. 9, 26, 29, 81, 86, 88, 92, 93 H Hayes Valley . 3 dune sand 78 Hayward fault ................................................ 90, 93 INDEX Page Heavy-mineral analyses, sandstone . 15 Hinde, G. J., quoted ‘ 48 Horseshoe Bay, alluvium . 85 greenstone 29, 30, 32, 33, 34 radiolarian chert and shale .................... 49 Hospital Cove, elastic sedimentary rocks ........ 26 Colma Formation ,. 70 conglomerate .............................................. 20 metamorphosed sedimentary rocks 21, 22 radiolarian chert and shale 49 structure ................ _ 93 Huichica Formation . 74 Hunters Point ........... 93 Franciscan Formation 9 seismic activity 93 serpentine ........ 62 structure .................................................... 92 Hunters Point—Fort Point shear zone .................... 50, 55, 64, 92 I, J, K Ingleside Chert, obsolete name 10 Introduction ........................ 2 Iron oxidation, greenstone 30 Islais Creek, seismic activity . 93 Jadeitization ............................ . 22, 36 James D. Phelan Beach, elastic sedimentary rocks .................... 25 fossil locality .......................... 27 radiolarian chert and shale 49 sandstone 12 serpentine . . 56 Joints ..................... 12. 18, 30 Kahn Playground, Colma Formation .......... 70 Klamath Mountains, Franciscan Formation border .. Knox Point, greenstone ........... metamorphic rocks .......................... 55 metamorphosed sedimentary rocks 21, 22 L Lafayette Square ................................... 3 Lake Mereed area, Colma Formation .. _ 67 Laguna Honda, elastic sedimentary rocks .. 24 fossil locality ............................................. 28 radiolarian chert and shale 50 shale 20 Lands End, beach deposits 76 elastic sedimentary rocks greenstone landslide deposits . .. 87 metamorphic rocks .............. 50, 54 radiolarian chert and shale 49 serpentine 56 structure 92 Lands EndrPoint Lobos 3 sheared rocks . 56 Landslides 11, 56. 66. 77, 86, 94 Lime Point, elastic sedimentary rocks ........ 25 greenstone ........................................ 29, 31, 34 landslides ............................... 87 radiolarian chert and shale 49 sandstone matrix ....... 14 Lincoln Park, serpentine . 56 Location ............................... .9 Lone Mountain, greenstone 30, 33 sheared rocks .. ..... 56 M Mack, M. S., analyst ........................................ 22 Manganese 37, 40 Manganese oxidation, greenstone . 30 Mapping, methods 8 Marin Peninsula .. 3 alluvium ................. ._ 85 chert, radiolarian 37. 39, 49 elastic sedimentary rocks 25 .Merced Formation .. Marin Peninsula—Continued Colma Formation ..................................... fossil locality . greenstone ....... landslides ........................................... metamorphosed sedimentary rocks . ravine fill ........................ sandstone detrital grains sandstone matrix shale, radiolarian slope debris . structure ..... Marin Sandstone, obsolete name Marina Park, artificial fill ............ Masonic Temple area, sandstone Matrix, Colma Formation .. sandstone ........................ serpentine ......... Page 70 Merritt Sand .............................................. 73, '74, 84 Metamorphic rock grains, sandstone .......... Metamorphic rocks 25, 27, 50, 55 Metamorphism ...... 21, 29, 36 Millerton Formation ...... 74 Mission Dolores, elastic sedimentary rocks 24 seismic activity ....................................... 93 Montezuma Formation ........ 74 Mount Caroline Livermore 5 elastic sedimentary rocks 26 greenstone ......................... 35 Mount Olympus, artificial fill .. 90 elastic sedimentary rocks 24 radiolarian chert and shale .. 49 Mount Sutro ................................ 3 elastic sedimentary rocks 24 Colma Formation 70 dune sand .............. 78 Franciscan Formation ...... 9 radiolarian chert and shale .. . 49, 50 sheared rocks ..... 56 Mount Tamalpais 5 structure ...................................... 90 Mountain Lake, Colma Formation ‘ 70 Mud, bay , . 66. 82, 8.7, 94 Mussel Rock, seismic activ1ty 93 N Napa River, sediments ...................... 81 Nob Hill, elastic sedimentary rocks 23 Colma Formation 70 dune sand ............. . 78, 80 sandstone detrital grains ..... 15 North Point, Colma Formation . 70 0 Ocean Beach ......... 3 beach deposits 76 dune sand 80 sand source 77 Olivine 58 Origin, chert, radiolarian £7 elastic sedimentary rocks 26 clay ....................... 85 Colma Formation 71 dune sand 80 gabbro ....... 66 greenstone .. .96 landslides 86 metamorphic rock . 51 mud .......................... 85 San Francisco Bay sediments 81 serpentine .................................. 58, 6‘4 shale, radiolarian . 48 Orthopy roxene .............. 58 P Peninsula Point, sediment deposition area 82 Phosphate nodules, shale ................................ 20 Phyllite Pilarcitos fault .......................... 90 Pillows 26, 29, 30, 32, 36, 76 Plugs . 36, 65 Point Bonita, Franciscan Formation 9 Point Diablo, greenstone 29 sheared rocks ............ 56 structure 93 Point lone, sandstone detrital grains 15 Point Lobos ..... 3 dunc sand . 80 structure .. 92 Point Lobos—Lands End 3 sheared rocks ......... 56 Posey Formation ............................................. 84 Potassium feldspar, relation to age, clastic sedimentary rocks . 28 Potrero Hill ............................................... 3 elastic sedimentary rocks 26 gabbro ................................... .. 65, 66 serpentine ...... . 56, 60, 63, 65 sheared rocks . 55, 56 structure . 92, 93 Precipitation 5 Presidio 6 beach sand elastic sedimentary rocks . Colma Formation . gahbro ...................... greenstone ........................... 34 sandstone detrital grains . 15 seismic activity 93 serpentine ..... 56, 65 structure .. 90 Previous work 7 Pyroxenite, in serpentine . 65 Q Quarry Point, dune sand ................................ 78 metamorphic rocks ................................. 50 metamorphosed sedimentary rocks 21, 22 sandstone detrital grains . 15 sandstone matrix ............... l4 Quartzite ........................................... 27 R Raccoon Strait .................................................. 5 sand bars 82 Rainfall 5 Ransome, F. L., quoted ..... 20, 54 Ravine fill ...................... Richards, H. G., quoted . 74 Richardson Bay ............ 3 Franciscan Formation 9 structure ...... 93 Richmond District .. 3 Rincon Hill 3 clay 84 dune sand . 80 greenstone .. 33, 34 Rockslides 87 Rodeo Lagoon 5 Russian Hill ........................... 3 elastic sedimentary rocks , Colma Formation greenstone ....... sandstone matrix 14 structure 92 S Sacramento River 3 ancient sediments ...... San Andreas fault ., __ San Antonio Formation .............................. 73, 84 San Bruno Mountain, clastic sedimentary rocks 24 structure .................................. 90 INDEX Page San Francisco .................................................... 60 clay ...................................... elastic sedimentary rocks chert, radiolarian ...... 39,49 Colma Formation , . 67, 69. 70 dune sand .. 80 greenstone 34 metamorphic rocks 53 ravine fill .. .. 81 serpentine . 56, 58, 62 shale, radiolarian .. 39, 49 slope debris ........ 81 structure San Francisco Bay sediments .................. San Francisco Peninsula . 3 metamorphosed sedimentary rocks _ .. 21 structure ..................... 93 San Joaquin River 3 ancient ...... 71 sediments .. 81 San Pablo Bay, sediments 82 Sand bars ............................. 82 Sand drift .. 78 Sandstone .. 9, 11, 23, 24,25, 26, 27, 35, 81, 92 Sausalito Chert, obsolete name ......... 10, 28 Sausalito-Tiburon area, artificial fill .. 88 Schistosity 51. 53 Sedimentary rock grains, sandstone l5 Sedimentary rocks, elastic 11 metamorphosed 20 structure . 92 Sedimentation 8, 67 Sediments, bay 81 Seismicity 93 Serpentine . 9, 27, 56, 76, 88, 92 antigoritic .................................. 59 related to metamorphic rocks sheared .......... 55, 56, 86 Serpentinization . 64 Settlement s Shale ........ 9, 18, 23, 26, 27, 81 radiolarian .96, L5, 48 sheared 12, 20, 55 Shear zones ......... 90 Sheared rocks . Sheared serpentine Silica, origin, radiolarian chert and shale .. 47 Sills ............................................................. 36, 65, 66 Simpton Point, conglomerate 20 greenstone .......................... 32, 35, 36 metamorphosed sedimentary rocks 22 Slate ..................................................... 50 Slope debris .................... 80, 86 Soils, sedimentary rocks _ 11 Source, dune sand 80 sand .................. 77 South Bay, fossil locality . 27 sandstone detrital grains 15 South Twin Peak ................... 3 Stratigraphy ................ 9 sedimentary rocks 11 Structure ...................... 90 sedimentary rocks 11 Stuart Point, conglomerate 20 greenstone ..................... _ 29, 32, 35 radiolarian chert and shale .. 49 sediment deposition area 82 Suisun Bay, sediments ........... . 81, 82 Sunset Heights 3 dune sand 78 Franciscan Formation 9 greenstone ...................... 32 radiolarian chert and shale 39, 45, 49, 50 Sunset Reservoir, structure 92 Surficial deposits ............................ . 66‘, 94 Sutro Reservoir, Colma Formation .. 69 T Telegraph Hill .................................................. 3 U. 5. GOVERNMENT PRINTING OFFlCE: 1974 O - 534—039 Page Telegraph Hill—Continued beach deposits ................ 76 clastic sedimentary rocks . 23 clay 84 sandstone matrix 14 structure . 92 Temesca] Formation . 73 Terrace Formation, obsolete name . ‘72 Thickness, artificial fill beach deposits ......... chert, radiolarian .............. elastic sedimentary rocks clay ....................... Colma Formation conglomerate . dune sand ....... Franciscan Formation gabbro ...... . 66 greenstone 33, 34, 35 mud ....... . 82 sandstone 11 Thurber, D. L., quoted 74 Tiburon Peninsula 3, 5 artificial fill ...... . 88 beach deposits A. 76 chert, radiolarian ............... 49 elastic sedimentary rocks 26 Colma Formation 70 conglomerate ....... 20 Franciscan Formation greenstone ............... metamorphic rocks . metamorphosed sedimentary rock serpentine .................................................. 56 shale, radiolarian structure Tiburon Peninsula syncline Tidal currents ........................ Topography Treasure Island, artificial fill . 90 Tufi' ......................... Turbidity currents Twin Peaks . alluvium chert, radiolarian Colma Formation _______ Franciscan Formation greenstone ........................... sandstone detrital grains . seismic activity shale, radiolarian .. sheared rocks ............................................. 56 U, V, W, Y U.S. Mint area, elastic sedimentary rocks. 25 serpentine .................................................. 60 sheared rocks ....... 56 Varnes, D. J., quoted 86 Veins in serpentine ....................... 60 Volcanic rock grains, sandstone _ 15 Volcanic rocks 25. 27 Volcanism , 4'7 Waldo Tunnel, fractured greenstone .......... 30 Water depths ................................... 3 Weathering, chert, radiolarian . .46 Colma Formation . 6.9 greenstone .............. 98 metamorphosed sedimentary rocks 23 sandstone ............................................. 18 serpentine . shale ........... radiolarian Weaver, C. E., quoted .. Whitney, J. D., quoted _ Worthing, H. W., analyst Yellow Blufl’, greenstone ..... Yerba Buena Island, metamorphosed sedimentary rocks ................... 21 sandstone matrix ...................................... 14 UNITED STATES DEPARTMENT OF THE INTERIOR PROFESSIONAL PAPER 782 GEOLOGICAL SURVEY MILL VALLEY 5.5 MI.\63D PLATE 1 1 22030, ‘ 4.1 MI. To us. :0! __g 55 [559 IV NW v A” ”052,323,, 5451mm. R. 6 W R. 5 w. 27’30” 549 (SAN QUENTIN) 1 450 000 FEET 554 122 ‘2 37215230” ~ .. - ' 3 - EXPLANATION MOUNT TAMALPAJS / ‘\“ [I \ REFUGE “ _/\ X‘ SURFICIAL DEPOSITS . “"XQ / flflflflflflflfl , T” r S T evkv \‘ ‘i P X 0" 5° § Artificial fill Modern beach deposits 5* g Predominantly dune sand but in— Predominantly well sorted medium i . .3 < eludes silt, clay, rock waste from to coarse gray sand; coarse gravel “gloom N‘ '11 u E excavations, manmade debris, and and large cobbles in some places. i 91 organic waste. Probable fill ma- Maximum thickness approximately ‘5 i terial indicated in parentheses. tofeet 2 I . . . k 0., i‘ Maximum thickness approxi- Qly N S > mately 60feet QI Qal Qd s: q f 0'0 E I; “1 <( . . . . . . SE 0‘ r Landsllde dep0s1ts Alluv1um Slope debrls and ravme flll Dune sand L 35?, 3 Generally unstratified mixtures of Predominantly clayey silty sand and Angular rock fragments in sand, Clean well-sorted fine to medium >_ § bedrock, sand, silt, and clay in clayey silt, some pebbles; locally silt, and clay matrix; generally sand; yellowish brown to light 0: 8 varying proportions. clean medium sand, generally gray light yellow to reddish brown. gray. Maximum thickness ap- > Qm Qu <11 *3 Qly, younger landslide deposits to brown. Maximum thickness Maximum thickness approxi— proximately 150feet Z 500 GOO S Qlo, older landslide deposits approximately 25 feet; observed mately 80feet . . . . _ > [I iigg FEET E thickness 15 feet Bay mud and clay Surflclal depos1ts, und1v1ded E “96 Plastic gray silty clay; some lenses < of sand, peat, and shell fragments; D L fluid to soft upper layers; mod- 0 erately stiff clay at depth Con- cealed except for exposures north of Corinthian Island and north of Sausalito Point. Maximum thickness approximately 140feet w Older beach deposits Colma Formation I g Predominantly well sorted medium Unconsolidated fine to medium sand ............. ’ § to coarse gray sand. Maximum with small to moderate amounts . .3 < thickness approximately 30 feet of silt and clay; in places includes ,,/ / x‘ , . . bi: clay beds, 6 inches to 5ft thick; ' " ; ,r y “ “89 cobble-size rubble rare; commonly ’ light brown to gray. Observed » thickness 75feet; probably max— 1, imum thickness approximately J L 300 feet 2 w BEDROCK ,,,,,,,,,,,,, \ w a; Ks IU) 3 D 8 . . O 4188 § Sheared rocks, undifferentiated m E Coherent blocks and pieces of hard rock as much as several > U D hundred feet in diameter, in a matrix of intensely sheared < E shale and serpentine. Clasts predominantly sandstone, shale, E g and serpentine, but also include all other rock types known 0: b in Franciscan Formation. Matrix gray to greenish gray, U moderately firm to soft and clayey; generally expansive and plastic when wet J 50’ 7 .\ gb C q o . '5‘ ,- , ~ .1 1 . . _ - Serpentlne Gabbro 3 g -. ' g k‘ . . ._.- “87 Mostly soft sheared rock containing hard knobs of un- Fi ne- to coarse-grained gabbro; E ’5 --------- sheared serpentine, rodingite, and rocks of the includes diabase where texture is E 3‘ U) Franciscan Formation. Derived from peridotites subophitic. Occurs as inclusions (75 g D (mostly harzburgite, some dunite). or segregations in serpentine E m 8 sph, hard serpentine, slightly sheared. Shown sepa- ,5 2 U rately on Potrero Hill only. Variously colored, but g < generally greenish gray, blue, or brown. Includes Cu I“ sheared rocks (Ks) where that unit is mostly serpena “J “a tine 6 a“ \ D i z , . - < /“ S a u ‘ ‘ o u u I: A ’ § § Clast1c sedlmentary rocks Rad1olar1an chert and shale Greenstone Metamorphic rocks g g; .3 § KJss, sandstone; thick-bedded and massive graywacke sandstone inter- Reddish-brown; alternate beds of Greenish-gray aphanitic to medium- Fim- to coarse-grained slate, schist, '2'; U S g bedded with thin layers of shale and fine-grained sandstone; hard brittle chert, 1 to 5 inches grained altered volcanic rocks; and granofels of the blueschist E a g $ some thick conglomerate lenses. thick, and firm brittle shale, 1%; to predominantly basalt. Includes facies E: u) a g < KJsh, shale and thin-bedded sandstone; predominantly interbedded and 1/2 inch thick. Locally includes flows, agglomerate, and tuff; > a < I A ’1 4' ’2/ / . ‘g S. laminated shale and fine-grained graywacke sandstone; beds generally bodies of massive chert pillow lavas common; less com- 8 g I G , , .. " :5 . ' , fi ’ » , a DR 2 to 5 inches thick. monly massive; interbedded with .5 _) I " 1/ .. """"""" . , . . z .. . ‘L _. \ S '8 KJs, sandstone and shale, undifferentiated; consists of units KJss or radiolarian chert and sandstone x: a , ,r/ ‘ «, ‘ _ ' ~~ ,. ‘-‘ if) g : KJsh so poorly exposed that predominant lithology could not be S q 85 water ”i \ ~ _ 5 ,,,,,,,,,,,, - "T" la a determined. B4 3., m. . . \ ////// m . . . . . \ “ ‘iflpent'ne ' , , ,, / 4/ I \ E u) Sandstone and shale at Point Lobos and vicinity west of City College \é’ég ‘ \\\\\ / /60 ’//./ ll / 3» fault zone is probably Great Valley sequence of Bailey,Irwin, and j ‘\ \y\:~-»-«. ._——-—"’ /”‘ / ’ / ~. t \ . ‘* m Jones (1964 p. 123) t m is —‘ 30 M/, I , \\ // x X . <{ in I 9 3 ____________________________ ‘ LT“ North Paint (x 90, waté{ ‘. \ ‘Q ,2 9: \ x, . - . » _ _ X » \ ....... FORT MASON ,/ f \~~.\ I \ w I. \ Q \ G r\i i " ' MILITARY 71, Q“ 25' sang 1 \\ i J I, . _ , , _ . . 45' “'33”- O ‘x ‘9 \\ IFort Point . I ‘‘‘‘ ‘ .4 , , , ‘ . , - 49' water h‘ KJsh (Shale)'\\ \‘ lo\o.\ \\ iI' k] _ . " ». s - ' 112, Qm I. “w, \ 110 water \\ l 5 Qu (clay) \I *——84, water’\_ """ R _O 33, Qm \\ \\ o i, O 114' Qm #21104 35’ sand ————— —————— __+_ ______ __I__”’ . —————— ‘ \ 30/ sand w / Contact Approx1mate boundary of transverse sand bars in \ , y \ \ . . . . . . . . \\ 58 Q“ {:3 9:3Vr3'3V91.\_/')\/ K15“ (shale) Long dashed where approximately located; short dashed Antlclme Syncllne San Franc1sco Bay (Modlfled from Glbson, 1951, p. 6) <9o __c 8’ ‘, where indefinite; dotted where concealed Folds \46' water KJsh (shale) \ _ , . _ §8’\Qm_ \\ Showing axis and direction of plunge. 6b 26’ saindx‘bl 50 Dashed where approximately locat - 073' Qu{43’ sandllday _I___U__'_; ..... ed; dotted where concealed. Quarry 9’ clay D 35 '1 . o o ‘J— }KIsh 31, Qaf. \.\ Tow/495v , ate, , Fault, showmg dip and relative movement Strike and dip of be(s 47' Qd 86’ Qm ‘2 525/? W E "3‘ Long dashed where approximately located; short dashed 100, Qu (clay, sand, sand, clay 4%,?69 ‘ ‘0 where inferred; dotted where concealed. Query 60 _d— I ) --------- clay ”:7“; 100' Q“ fxficfifgv 3; indicates greater uncertainty as to existence of fault Strike and dip of foliaaion . 58" c ay , 3’ 7.__5/wa‘e’ Y 3% than does dashed line. U, upthrown side; D, down- AbbreVIated columnar sectlon /fl"‘_ 9; thrown side X Stratigraphic sequence and approximate thickness in MG {38’ sand \ ‘i «x . . feet of geologic map unit based on logs of boreholes ‘ , x 2 30 . . 1 ’Qu 2,6 if"? sand, 50 ‘2 go $3: $63: “ F0581] locallty and excavation. Sand-clay may include one or both Light \ . t I 74' t‘ w of the followin .’ sand silt and cla in various 10x Sand' 30 wa er ' ' - . ’ WW 9 y . y y K Jg , 56’ Qmi é Showmg average dISph ApprOXImately horlzontal I T f I d I'd \ proportions; alternating thin beds of sand and silty Mgie Rgé‘y’» .101' On (clam: E} * ear zone op 0 an s I e scarp clay * it} i-Eghihofiifi , \thtle we Rock sangvifi'a‘f" * ' . ,, H san «cay , ,, J7 30 BAKE/RS ........ m5 ; 47 3O ,. 44444 ' I “'"KJ “7471:- _ 30 ”X” / incon Roint l U stack Head/Raking; amass BHEIIAN \_ . ; (60/ I /, . { STAIE/P ._ \y_\ i y 2 ,7 > '- x I S ...... J l /,- ; / , \ ...... ft \\./(° ‘If/ :3: I ixxl 600 A KJg KJc “3° Angel bland A, 600 —————— ’ 41 I .. I 'x 68’ watero 4’ 32 400' " er _ ms' 3 KJC KJg 400' \‘MQG’Qm /: 200, «.1 , 77 SAN FRANCISCO BAY 200' , / / is m / ; SEA LE2VOEOL ' SEA LEVEL 4 / r‘ L ”/1/ I I ’ / /W’F‘3k " l 400' / h \ \ \ 388T Point? Ldbos O ,1 600' Franciscan Formation 600i ,3 0 j g 800' / / \ \ \ \ 800’ i . x 1 000' ‘ 1000' K455?" 0'3 (Seal! L S Rocwsflg m z i I “94%, E E I .v' Ll Is i z / ~ ‘0 <0 8 = a ,r . «I B Qu g I Nob Hill E Te'eg’aph B / 400' S E er S H“‘ 400’ 200' '3 O L) 200, ission Rock SEA LEglgé—I SEA, LEVEL erminal 400' 3188’ f ...... "”1 \/ ‘..,./ .. ~’-> Z. ;. " : / '_ ; :- . ‘ ‘. . ' :. > / . ' . " " . 1 . . . r - ' ‘ » .» .' C 15 14 13 12 n 10 9 8 7 6 s 4 3 2 1 C . . 4} 50/ _ QC If 50' 80 , Qaf Qd SEA LEVEL " _ L— ._ SEA LEVEL : - ' 0m 3 3 ,: , a 4" : 50' ~ Qu I - ‘1: T Q?“ T. : — 50' 100’ ~ : v” Z : _ I; E — : 2 ~ 100’ Qu if: : ' - 150' A :52 Elf — i ’7 I: — 150' 4179mm,” 4. : '— : — Q ~— _ _ _ u _ LOG SYMBOLS : — “T 2-.'. ...... 200' a _ ._. -0 _ 200' Sand Gravel and sand ‘ ;~ 0; :U ' Franciscan Formation ‘1: o o 250 7 "n - o. i 250' ,. Clay and silt Bedrock 300' — _ 300, . \ fl .\ Sand and clay , \ . . er , 37645, 350' 350' Wgs'r‘gugirg as m}, 1630 000 FEET R. 6 w, 27'30” F'LEf'lSHHACKER) zoo in M! (SAN FRAN SCO SOUT ) OALY C Tv a M 5540mm 5 122 22,30” INTERIORAGEOLOGICAL SURVEY, WASHINGTON, D.C.7I974~G72297 . ,. /3-">..1 r - 1, r. H EISHHACKEF? ) H Mi, HALF .«sccw was A: M 25s9 11/ NW SAN ammo ’0 W . ‘ Base by U.S. Geological Survey, 1956 SCALE 1:24 000 Geology mapped In 1948—61 by Julius Polyconic projection. 1927 North American datum 1 1,42 0 1 MILE Schlocker, M G- BonIlla, D. H- Radbruch, . 1 . , . . 4. . C. A. Kaye, and W. I. Konikoff 10,000-foot grid based on California coordinate system, zone 3 1000-meter Universal Transverse Mercator grid ticks, zone 10 Include revision from aerial photograph taken 1968. 1 .5 o 1 KILOMETER I—I I——-I l—l I—I I—I CONTOUR INTERVAL 25 FEET APPROXIMATE MEAN orcuwmwm DASHED LINES REPRESENT 5-FOOT CONTOURS QUADRANGLE LOCAmN DATUM IS MEAN SEA LEVEL DEPTH CURVES IN FEET—DATUM IS MEAN LOWER LOW WATER SHORELINE SHOWN REPRESENTS THE APPROXIMATE LINE OF MEAN HIGH WATER THE MEAN RANGE OF TIDE IS APPROXIMATELY 4 FEET TRUE NORYH GEOLOGIC MAP OF THE SAN FRANCISCO NORTH QUADRANGLE, SAN FRANCISCO AND MARIN COUNTIES, CALIFORNIA UNITED STATES DEPARTMENT OF THE INTERIOR PROFESSIONAL PAPER 782 GEOLOGICAL SURVEY PLATE 2 0 12°30 12222310", .. 1,11 1— 37 573°” ' 37 52 3° SAND SILT Th 6 4, l uron ‘2} 1 , . . . 9 [(33, Belvedere 54 Very coarse Coarse MedIum FIne Very fIne Coarse MedIum fine .1— 190 08‘ 100 ___’ ' «PO/1’ 000 g—/ 3 +1 <9 . Sausalito 7+ ‘33» Angel Island 1 ______________________________________________________ 05 T +4 ”a“ ___________ ' _ 45 90 — ca\““fi9‘—Tflfl ______________ E‘I‘ I —————— 13 "T F0 aUg“: , r' T ' + \m6 , / 80— 2,}0’ 3 5+ 12+ 30 ‘I' 2 /( ///’ E A18 A193 104! .29 4. co / CO‘N‘ /’/ a ‘14 27112;1r "- I34 ' 111,338 {1540 70 / g 0 — 1687‘ 17 ‘1685 42} :2336 4002 .25 45' ’ é) 0 BA )2 — § 158“‘ A “105411036 3: ’33 /4e$“ C15 C E E A160 16A 1904. 20423 24B 0 22 $ _ in 61" AN Lu 5: 163 ‘ MARIN F 8 E 50 — E A PENINSULA 1‘1 g} 419/1 A Alcatraz Island < Lime Point 8 % E 5 _ 0.5 —- E Treasure Island 2 0 2:; GATE L2” 4N r; I— _ OOLDP’ 1; g 40 a 2 Fort Point 8 1'0 1 I J 30 — PHI MEDIAN DIAMETER. 1 2 3 MILLIMETERS .5 .25 .125 Telegraph Hill I’RESIDIO MILITARY 900°C: RESERVATION 33. 4016‘“ 20 — 03:??? ‘3' 220 0" 2.0 ~— C) 10 "‘ I 4 Q __________ // + 1...". 0 .e’” I I I I I I I D I I L°b°s 2 0.1 .05 .01 E c—v GRAIN SIZE, IN MILLIMETERS Mechanical analyses by GM. Schmidt, 1958 J .3. E. 3” >1 I 1.5 — , "15 _____________________________ t-——""”M'.‘ SAN FRANCISCO 3 B. REPRESENTATIVE CUMULATIVE CURVES OF GRAIN SIZES +1 3 3‘11; """"""" 3“ '1 . E K GOLDEN GATE PARK 35. I 13 8 E o ,6 __________________________ _ -. 1 47 m V _____________________________ + 36 Mt 0| mpus S I S 4 9 e V 3 q CLAY EXPLANATION .2 e ’3 Potrero ”J ‘5. M Su ru : . 52 E _ 17 I); O 37 t t 61 TWin Peaks g l 43 Pom! I (101 E 1.0 18‘ 16c _1__3 41. 40. m 1062 054 53 E 49. Number in parentheses is percent 2 A 168 ‘IGA ‘13 42 ”gum ”0,1114% ++q Q 63 of gravel included in sample S A193 A 10 37"45' '8 1 I “ 55"56 I 37.. , '5 + 122°22'30‘15 Gravel’ > 2 E 685 5 +2 I) I ,2 MILES Sand, 2—0.0625 “L ”A ‘1687 ”A ‘1 'I’ 8 7 siIt,o.062540.0039 1511‘ 13A 1% H 4+ A. LOCATION OF SAMPLES 0930-0039 3 +3. .3 E Sizes, in millimeters 0-5 — . 21 o..40 33 XPLANATION 2. ‘. '32 25 Symbols used on A C and D IQDALMZV? 3.137.823 :22 Numbers are map localities 41 +8 I 55 46 Map locality numbers are not shown l(70) for all dune sand samples Colma Formation Slope debris and 50' 0 I I I I4 ravine fill 59 PHI MEDIAN DIAMETER, 1 2 A ' . .25 .125 Beach sand 062 55 5.3 33:10 .47 MILLIMETERS 5 . CLAYEY <5 1601 T CLAYEY D. SIZE DISTRIBUTION DIAGRAM AI£I Alluv1um SAND I CLAY SILT Raised beach sand 0 58 O 63 26 X 49 I57 _ 51 0 Beach sands of (2). Dune sand Trask(195 9) 44 SILTY 1.401 SANDY SAND I 50 n 48 SILT SAND o 62 ”3’ F. COMPOSITION OF SAND GRAINS 11 2 10 8 I SILT C. SAND, SILT, AND CLAY COMPOSITION, BY WEIGHT '52:) 0 a Map 111111 Colma Formation Beach sand Dune sand 3 '5 CL 5-4 E. GRAIN SIZE DISTRIBUTION MEASURES % g N Median diameter 84th percentile 16th percentile Sorting, . _ Sorting, Skewness, One - *4: deviation Median diameter, Skewness 75th 25th Trask Trask per- Map locality number 2 9 3 13 6 1 1A 15 (*) 29 22 ( ) 57A Map Sample mm phi mm phi mm phi measure Phi phl measure,ph1 percentile percentile coefficient coefficient centile No. No. Approximate frequency*** Md Md¢ p84 4’84 p16 (1’16 M¢—Md¢ mm mm Q1 Q1 Q3 mm a¢=l/2(¢84_¢15) M¢=1/2(¢84+q>16) a =— So: — Sk: — Heav minerals (w» 03 Q. Q3 Md2 C Y . . . (spec1f1c grav1ty >282): Actinolite—tremolite - 2 - - l - - - 2 1 - I Colma Formation Allanite - - - - - — - - - - 1x — Andalusite - - - - - - - - 1x - - 1 SF1304 0.206 2.28 0.043 4.54 0.315 1.67 1.44 3.12 0.58 0.094 0.282 1.73 0.625 0.6 Apatite _ 1 1 1 1 1x 1 _ 1 _ _ _ 2 1306 .183 2.45 .104 3.27 .279 1.84 .72 2.56 .15 .133 .235 1.33 .933 .54 Aragonite - - _ _ _ - _ 1x - - _ _ 3 1169 .258 1.95 .140 2.84 .308 1.70 .56 2.27 .57 .193 .290 1.23 .841 .48 Augite 4 6- 5 6+ 6- 6- 5 4 5 5 5 2 4 335 .236 2.08 .03 5.06 .42 1.25 1.91 3.16 .52 .107 .367 1.85 .71 .8 Biome _ 1 _ 1 - . 1x _ _ _ _ 1 5 1844 .270 1.89 .156 2.68 .389 1.37 .66 2.03 .21 .186 .358 1.39 .914 .58 , Brookite _ _ _ _ _ 1x 1x 1 1,. _ 6 2068 .220 2.18 .086 3.54 .328 1.61 .97 2.58 .41 .147 .287 1.39 .872 .62 Cassiterite _ _ _ _ _ _ _ _ _ 1 _ _ 7 2069 .204 2.29 .120 3.06 .288 1.80 .63 2.43 .22 .160 .265 1.28 1.019 .50 (311101119 1x - 1x 1 _ - 1x - - - _ _ 8 431 .208 2.27 .135 2.89 .317 1.66 .62 2.28 .02 .152 .280 1.36 .983 .57 Chromite 3 3 3 3 2 1x 5 7. 4 5 6 3 9 1460 .220 2.18 .097 3.37 .34 1.56 .91 2.31 .76 .155 .297 1.38 .951 .54 Chrysotfle - - - - 1x - 1x - - - - _ 10 414 .24 2.06 .130 2.94 .383 1.38 .78 2.16 .13 .171 .328 1.39 .973 .8 Clinozoisite—epidote 6- 3 5 5 6 5 4 4 5 6— 5 6**** 11 775 .247 2-02 163 2.62 349 1.52 1451451 2-07 121% ~13; 35121 1?; ~32; '22 Composite, fine-grained,unidentified 4 2 3 1 4 2 2 4 4 2 2 4 12 1770 .243 2.04 .17 2.56 .31 1.69 . 2.13 . . . . . . Enstatite - - _ - 2 - _ - - - _ - 13 1357 .200 2.32 .123 3.02 .254 1.98 .52 2.50 .35 .155 .230 1.22 .892 .38 Garnet, brown, green, or pink 2 - _ 3 1x 1x 1 1x 1 2 1x 2 Garnet, colorless — 1 - 1x 1 x 1x 2 - 1 - 1x 3 Beach sand Glaucophane 2 - 1 3 £11 ‘21 1 1X (11" 21 1x 2 Hornblende, brown 5 5 4 5 4 4 5 14 SF1691 0.44 1.18 0.270 1.88 0.675 0.57 0.66 1.23 0-08 0.31 0.60 1.39 0.96 1.5 T Hornblende, green 7+ 6 7- 7- 6+ 6 5 6+ 6+ 6 5 6+ 15A 16905 .340 1.56 .225 2.15 .54 .89 .63 1.52 --06 .255 .46 1.34 1.04 1.1 Hornblende, oxy 1 4 1 2 4 4 2 1 4 4 2 5 15B 1690B .355 1.49 .240 2.06 .55 .86 .60 1.46 -.05 .269 .48 1.34 1.02 1.1 Hypersthene 5 /6- 4 5 5 6 5 2 5 5 5 2 16A 1688S .369 1.44 .235 2.09 .78 .36 .87 1.23 -.24 .264 .60 1.51 1.16 2.3 Idocrase 1 - - 1 - - - - - - _ _ 16B 1688B .384 1.38 .252 1.99 .82 .29 .85 1.14 -.28 .282 .623 1.49 1.19 2.6 Ilmenite 3 5 5 4 4 5 6+ _ 5 5 6 5 16C 1689 .508 .98 .30 1.74 1.09 .12 .93 .81 -.18 .349 .840 1.55 1.14 2.6 Jadeite - - 1x - - - - - 1x - _ _ 17 2129 .350 1.52 .229 2.13 .558 .84 .65 1.49 —.05 .255 .495 1.39 1.03 1.5 Kyanite - _ _ _ _ _ _ _ _ _ _ _ 18 2130 .56 .84 .262 1.93 .98 .03 .95 .98 .15 .345 .84 1.56 .92 1.5 Leucoxene - - - . - - - 2 2 2 _ * 1685 .318 1.65 .203 2.30 .52 .94 .68 1.62 -.04 .226 .52 1.52 1.16 .85 Lawsonite - - - _ 1x - - - _ - _ _ M 1687 .420 1.25 .265 1.92 .68 .56 .68 1.24 —.01 .306 .70 1.51 1.21 2.4 Magnetite (+200 mesh) 4 3 2 4 5 6- 4 — 5 5 5 2 Monazite 3 4 2 3 2 2 2 — 1 3 3 3 Raised beach sand Nephrite _ 1 _ ' ’ ' ’ ‘ ' 2 ‘ ' ' ' ‘ *** Opaque, undifferentiated - - - - - - - — - — — - 19A SF1432 0.248 2.01 0.197 2.34 0.33 1.60 0.37 1.87 -0.38 0.212 0.295 1.18 1.02 0.54 Pumpellyite 1x 1x - 1 1x - — 1x 1 - — 2 19B 1434 .490 1.03 .247 2.02 .80 .32 .85 1.17 .16 .299 .71 1.54 .884 1.15 Pyrite 1 1 - 1 - - - - - — - - 19C 1435 .260 1.94 .185 2.43 .421 1.25 .59 1.84 —.17 .202 .351 1.34 1.05 .64 Rutile - 1 1 - - 1 1x 1x 1 1 — - 19D 1436 .265 1.92 .213 2.23 .35 1.52 .36 1.87 -.12 .228 .317 1.22 1.03 .53 Sphene 4 4 4 3 3 5 3 4 3 4 4 3 _ Spinel - - - 1 - - — - - - 1x — D e and Staurolite _ - - - - — - - 1 - - — un S Stilpnomelane - - - — 1 - - - - — - - 20 SF1653 0.256 1.97 0.210 2.25 0.338 1.57 0.34 1.91 ‘0.18 0.219 0.305 1.18 1.02 0.40 Thorite (.uranoan) ' ' ' ‘ ' ’ ' 1" ' ' ' ' 21 1126 .240 2.06 .168 2.57 .332 1.59 .49 2.08 .04 .189 .295 1.25 .968 .41 10111111311116 - - - 1 - - x - 1" ' 1 - - 22 2001 .187 2.42 .153 2.71 .256 1.97 .37 2.34 -.16 .162 .232 1.20 1.08 .35 Xenotlme - - - - - 1 - - 1" - 1* - 23 2127 .205 2.29 .16 2.64 .271 1.89 .38 2.27 —.05 .172 .25 1.21 1.02 .38 £11,691“ 1 1 1 2 2 1: 1 ? 3 2 24A 1249A .215 2.22 .163 2.62 .286 1.81 .41 2.22 0 .175 .265 1.23 1 .39 01516 . ' ' ' ”*1“: ' x ‘ ‘ ' ' 24B 124913 .208 2.27 .165 2.60 .275 1.86 .37 2.23 _.11 .170 .256 1.23 101 .39 Percent of heavy minerals (+200 mesh) 10.3 23 14.5 12 10.4 25.7 8.3 70 17.7 28 15.3 - 25 2004 .187 2.42 .147 2.77 .251 1.99 .39 2.38 —.10 .159 .232 1.21 1.05 .28 26 1902 .240 2.06 .190 2.40 .318 1.65 .33 2.03 -.09 .20 .288 1.20 1 .40 Approximate percentage 27 2002 .258 1.95 .187 2.42 .355 1.49 .47 1.96 .02 .213 .321 1.23 1.03 .52 28 880 .259 1.95 .190 2.40 .35 1.51 .45 1.96 .02 .208 .32 1.24 .992 .55 Light components 29 1253 .223 2.16 .161 2.63 .286 1.81 .41 2.22 .15 .175 .267 1.24 .939 .39 (specific gravity <2.82): 30 1094 .223 2.10 .169 2.57 .295 1.76 .41 2.17 .17 .183 .275 1.23 .927 .46 Rock fragments 40 65 60 25 50 6O 30 - 40 30 10 - 31 1581 .217 2.20 .159 2.65 .280 1.84 .41 2.25 .12 .173 .262 1.23 .963 .37 Quartz 40 15 35 70 30 10 50 - 25 30 30 - 32 1353 .20 2.32 .150 2.74 .272 1.88 .43 2.31 -.02 .160 .249 1.25 .996 .37 Feldspar(p1agioclase: potassium 33 1354 .20 2.32 .152 2.72 .278 1.85 .44 2.29 -.07 .164 .252 1.24 1.03 .40 feldspar ratio is 10:1 or larger) 20 20 5 5 20 3O 20 - 35 40 60 - 34 1580 .215 2.22 .160 2.64 .277 1.85 .40 2.25 .08 .173 .263 1.23 .985 .41 35 121;; 2(3): 2.28 .12: 2.65 .275 1.86 .213 5.22 0 08 1;?) 3;: 13% 1 955 2313 1: Black sand, Princeton Beach, 21 miles south of Point Lobos. -115+ 250 mesh only, modified from Hutton (1959, p. 10) 36 . . . 2.62 .316 .66 - - - ‘ .275 ' ' '44 ** Sample from 8.7 miles south of Point Lobos, 700 feet east of Pacific Ocean 3; 1:253; .222 2.1313 IE: 2.2: .23: 1.223 3‘21 332 ~83 1;: .283 1355; 13:5 '40 *** The Evans, Hayman, and Majeed scale given by Hutton (1950, p. 650), as shown below, is used: 39 1355 :226 2:15 :164 2:61 :3 1:74 .44 2.18 .07 .182 .275 1.23 .980 .41 Frequency Approximate percentage Frequency Approximate percentage 40 1575 .228 2.13 .165 2.60 .315 1.67 .47 2.14 .02 .180 .282 1.25 .977 .52 8+ 90-100 6 18-22 41 1295 .245 2.03 .207 2.27 .299 1.74 -27 2-01 707 218 .282 1.14 1.024 .52 8 75-89 6- 14-17 42A 1293B .252 1.99 .188 2.41 .34 1.56 .43 1.98 ”-01 203 311 1.24 .994 .58 8- 60-74 ' 5 7-13 428 12938 .243 2.04 .183 2.45 .35 1.51 .47 198 -~13 -20 295 1-22 -99 -6 7+ 45—59 4 4-6 42C 1294 .244 2.04 .193 2.37 .308 1.70 .34 2~04 0 207 287 1-18 998 -48 7 35-44 3 2-3 43 811 .244 2.04 .188 2.41 .325 1.62 .40 2‘02 <05 20 -3 1-22 1-01 ~56 7- 28-34 2 1-2 6+ 23-27 1 1/2-1 * Sample from beach 1 mile south of San Francisco North quadrangle 1 X 1 grain only ** Sample from beach 8.7 miles south of San Francisco North quadrangle 44* Elevations of samples from map locality 19 are 123, 130, 175, **** Mostly common epidote and 100 feet, respectively ***** -50+70 mesh COMPOSITION AND GRAIN SIZE OF SURFICIAL DEPOSITS, SAN FRANCISCO NORTH QUADRANGLE, SAN FRANCISCO AND MARIN COUNTIES, CALIFORNIA 534-039 0 —(1nside back cover) PROFESSIONAL PAPER 782 UNITED STATES DEPARTMENT OF THE INTERIOR GEOLOGICAL SURVEY M155 $34,533; it", N339 PLATE, 5 122 C 30' 545000m E. _ 549 (SidilfigOILE/EANVTIN) 1450 000 FEET 554 122 D 2230” a ,. H , as w R. 5 w, 27'30” ., . o 37 52 do ., . 37 52 30 LANDSLIDES IN THE SAN FRANCISCO NORTH QUADRANGLE MOUNT TAMALPAIS ,1 \ AM E RE FU G E Approximate . slope EXPLANATION Iii?) b83135 gigging Type Of movement Probable main cause 51%;: ‘2‘: ,,,,,, \\ “910001“ N \‘. Exposed bedrock /; . \ Includes Franciscan Formation and serpentine , /’ Mg} Angel Island so; 0 E: __ 50 — — 1 Qaf/ KJss Rockfall, debris flow Removal of support by wave cutting 65 Eli: x ’00 . 2 KJ ss Debris slide Water, removal of support by wave cutting 35 3 E Contours on buried bedrock 3 KJss do. Removal of support by wave cutting 40 ff 2 ._ .. Contour interval above mean sea level is .100 feet; _be- 4 er/KJSS Debris flow do. 30 it’d \ low med" sea level, 50 feet' 50'f00t mmrmedm? / 5 er Earthflow, debris flow Water, removal of support by wave cutting 20 contours are dashed. Contours not shown on Marin / 6 er d0 d0 25 and Tiburon Peninsulas and Angel Island where f . '. ' 20 bedrock surface is generally less than 20 feet below / 7 er Debris Shde Water , n. \ \ the topographic surface. Reliability of contours is / 500 000 8 KJss Debrls flow Water, removal of support by wave cuttmg 30 “90 U\ ‘\ \\._ , / , x greatest near surface exposures and near boreholes /’ FEET 9 KJSS do. do. 30 ,_ saysaufibquin‘tlt‘” /’/ / / / \\ that reached bedrock. Offshore contours where bore— “90 10 KJss do. do. 30 \ IJAV :\ \\ // 5/ ,r" (‘20— ; I _ \ hole data are lacking are based on US. Coast and . / 11 KJSS do. do. 30 l b, ‘3"\\\ \wy’ .,__i/ \/ / \\ Geodetic. Survey soundings, bottom sampling, tidal // 12 KJss do do. 15 \ current data,_and subjective geologic criteria and are x 13 KJ E thfl ‘ do. 25 only approximate. Bedrock-surface contours for / 55 ar _ 0W . 10 most of the Golden Gate channel and San Francisco / 14 KJg Debrls sllde Removal Of support by wave CUttlng Bay in this quadrangle, obtained by subbottom / 15 KJSS do. do- 15 acoustical profiling, are given in US. Geological . 16 KJss do_ Water 31 1! Survey open file report, 1970, entitled “Bedrock- Va. 17 K J55 do. Removal of support by wave cutting 45 ,1” surface map of central San Francisco Bay, Cali— ‘\ 18 KJSS d0. d0. 30 fornia by Paul R. Carlson and David S. Mc Culloch 19 KJg do. Removal of support 30 “89 0—70 20 er/KJss Debris flow Remméal of support by wave cutting 35) Borehole “89 21 KJss Debris sllde 0- . . . 22 er/ KJg Debris flow, earthflow Removal of support by road out 30 Number indicates elevation of top of bedrock 23 K J m Debris slide Prolonged heavy rains 25 0-50 24 KJss Earthflow Water, removal of support by wave cutting 35 / Borehole that did not reach bedrock 7 Number indicates elevation of bottom of hole _-_ W Tiburon Peninsula and Belvedere Island Top of landslide scarp 25 K5 Rotational slump, debris flow Removal of support by manmade cut other tharroadcut 30 .51 26 er/ KJss Debris slide Removal of support by wave cutting 40 4188 / Landslide 4! 27 er/ KJss do. do. 20 / . 88 28 er/KJss do. do. 25 Number refers to accompanying table 29 er /K J S 8 do. do. 30 30 ' er/KJss do. do. 45 31 er/KJss do. do. 40 C) Fagbag/Rpcks \l '\—\3\‘L’\‘< " . . Marin Peninsula 50' ,, 50' 32 er/ KJss ' Earthflow, debris flow Water 15 “87 \ \ /= r (/ \,~ K ,5 / 1A “87 g: gsrjfijss 30' 30‘ if) \ g \ I\ .90 {/3/7“ ‘ ‘- sr ss 0. o. . \ (J t ,4 assuage; \\ /.-..;://"'ng330 35 er/KJss do. do. 20 x l I ,I \. i3 \\\ .1 v 3%: 33/7 ' 36 er/KJss do. do. 15 ,i I / x. ‘ \J // """" “\\“_:, / .oo/ 37 er/ KJg Earthflow do. 20 l ’ ’ ,/g 38 KJg Block glide, debris slide Removal of support by manmade cut other than roadcut 35 39 er/ KJg Earthflow, debris flow Water, removal of support by manmade out other than roadcut 40 ._ 40 er/KJss Debris avalanche Prolonged heavy rain 35 ..... .Mx/“m‘ ' ,i l 5 \ , r ' 41 er/ KJg Rotational slump, earthflow Water removal of support by manmade cut other than roadcut 35 ' ' ’ . .~_ " -- . ’ ,1 I, \‘~ ._ I’ , . I 42 er/KJg Earthflow Removal of support by loading of slope with Qalf and wave cutting 25 , 43 er ( K Jg)/ K J g Earthflow, debris flow Prolonged heavy rain 25 44 er (KJg )/ KJg do. (10- ' 25 4186 45 er/ KJg do. Removal of‘ support by wave cutting 45 46 er/ KJg Earthflow Removal of support by roadcut and wave cutting 30 47 er/ KJg do. Removal of support by wave cutting 30 , 48 Qlo (er)/KJss *33 --------- t" \ Qly/Qlo Debris flow, rotational slump Water, removal of support by roadcut **40—50 ‘ ’ ‘ 49 er (KJg)/KJg Earthflow Prolonged heavy rain 35 50 KJss Debris slide, debris flow Water 30 51 KJSh/ KJC Rotational slump, debris flow Fault surface, removal of support by roadcut *:21 ................. ~ 45 Q ,,,,,,,,,,,,,,, 3 52 Qlo (er )/ KJss *32 \ t ,,,,,,,,,,,,,, g g Qly/Qlo do. Water, removal of support by roadcut **40—50 \\‘,;§ 0 ' _______ 3 53 er(KJc)/KJss Earthflow, debris flow do. 25 $2 c_ — Locatelli Formation of ’—‘ Nodular olive-gray to pale-yellowishbrown micaceous silt- : n. Cummings and others (1962) 900 stone; masswe arkosm sandstone locally at base FIGURE 2.—Composite stratigraphic section of Tertiary rocks of central Santa Cruz Mountains southwest of San Andreas fault. GEOLOGIC SECTION SOUTHWEST OF SAN ANDREAS FAULT 7 Surface exposures of the Zayante Sandstone extend from 2 miles east of the town of Boulder Creek east- ward for a distance of about 6 miles (pl. 1). To the north the formation is in depositional contact with, or locally faulted against, the Vaqueros Sandstone; to the south it is in fault contact with the Butano Sandstone, except along Newell Creek and the adjacent canyon to the west, where it rests conformably upon the San Lorenzo Formation. This area of approximately 5 square miles is designated as the type area of the Zayante Sandstone. The formation at the type section is about 1,600 ft thick. About 1,800 ft of the Zayante Sandstone is poorly exposed in the vicinity of Lompico Creek between the contact to the southwest with the sub- jacent San Lorenzo Formation and the contact to the northeast with the superjacent Vaqueros Sandstone. The Zayante Sandstone is one of the most distinct lithologic units in the central Santa Cruz Mountains (fig. 3). It is composed predominantly of thick to very thick beds of moderately to poorly sorted pebbly medium- to coarse-grained biotite-bearing arkosic sandstone that is bluish gray where fresh but yellowish orange where weathered. Thick conglomerate inter- beds, lenses, and pods are composed of well-rounded varicolored porphyritic volcanic and quartzite pebbles and cobbles, and of more angular granitic cobbles and boulders. Thin interbeds of grayish-olive poorly sorted slightly granular fine sandy chloritic siltstone are com- mon; they locally weather to various hues of red and appear mottled. This formation most closely resembles the lower part of the Butano Sandstone but may be distinguished by the presence of greenish siltstone FIGURE 3.—Interbedded conglomerate and sandstone of the Zayante Sandstone, 500 ft north of Zayante fault along Newell Creek; larger clasts are granitic. interbeds, greater abundance of biotite, poorer sorting, and the orange weathering color of the sandstone. A nonmarine deposition for the Zayante Sandstone is suggested by the large-scale heterogeneity, poor sorting of individual beds, local channeling at the base of conglomerate beds both within the formation and at contacts, greenish and reddish mottling, and a com- plete absence of marine fossils. Because this formation is lithologically distinct from the Vaqueros Sandstone and is inferred to be nonmarine, it is separated from the Vaqueros, which is marine. The contact between the Zayante Sandstone and the subjacent San Lorenzo Formation is placed at the base of the lowest interbedded sequence of poorly sorted medium- to coarse-grained sandstone, conglom- erate, and greenish siltstone. The Zayante Sandstone is conformably overlain by, and locally intertongues with, the Vaqueros Sandstone. The contact between the Zayante and Vaqueros Formations is mapped at the base of the stratigraphically lowest thick to mas- sive light-colored moderately to well sorted fine- to medium-grained sandstone bed. The conformable position of the Zayante Sandstone above the San Lorenzo Formation of Eocene and Oligo- cene age (Narizian, Refugian, and Zemorrian Stages) and below and partly intertonguing with the Vaqueros Sandstone of Oligocene and early Miocene age (Zemor- rian and Saucesian Stages) dates the Zayante Sand- stone as Oligocene (Zemorrian). The local development of the terrestrial Zayante Sandstone within the Eocene to lower Miocene sequence represents a break in marine sedimentation in the Santa Cruz Mountains resulting from tectonism initiated in Oligocene (Zemorrian) time. OLIGOCENE OROGENY The Butano Sandstone and the lower part of the San Lorenzo Formation were probably laid down at bathyal depths throughout the central Santa Cruz Mountains. This depositional environment is inferred from the nature of the benthonic Foraminifera and from the paucity of mollusks throughout the coarse elastic part of this section (Cummings and others, 1962, p. 186— 188; Brabb, 1964, p. 674; Clark, 1966a, p. 68—69). Brabb (1964, p. 676) suggests that bathyal conditions continued during deposition of the upper part of the San Lorenzo Formation in the Vicinity of the San Lorenzo River and Kings Creek, 2—3 miles north of the town of Boulder Creek. While deepwater conditions persisted to the north, shallowing from bathyal to neritic conditions occurred east of Boulder Creek and north of the Zayante fault during deposition of the San Lorenzo Formation. Along 8 OLIGOCENE STRATIGRAPHY, TECTONICS, AND PALEOGEOGRAPHY, CALIFORNIA COAST RANGES Newell Creek, the lower part of the San Lorenzo For— mation yields deepwater Foraminifera, including cos- tate buliminids and costate uvigerinids. The upper part of this formation is coarser grained and locally contains in abundance such relatively shallow water mollusks as Pitar, Panopea, Solen, and M odiolus. Local shallow- ing continued during Oligocene (Zemorrian) time, and adjacent to the Zayante fault marine conditions gave way to the terrestrial conditions under which the superjacent Zayante Sandstone was deposited. The limited areal extent and coarseness of the Zayante Sandstone and the total absence of the San Lorenzo and Vaqueros Formations south of the Zayante fault suggest a regional uplift and emergence of the terrane south of this fault during Zemorrian time. Initial uplift along the Zayante fault probably occurred during deposition of the upper part of the San Lorenzo Formation. The rising block south of the fault furnished sediment to the adjacent basin to the north, resulting in the coarsening of the upper part of the San Lorenzo Formation and in shallower marine conditions. Continued uplift produced greater topographic relief, and coarser debris was shed northward, with the Zayante Sandstone deposited as an alluvial fan on the south margin of the adjacent marine basin. The granitic cobbles and boulders in the Zayante Sand- stone, locally as much as 4 ft in diameter, record this uplift and indicate that erosion had exposed the crys- talline basement to the south. The Zayante Sandstone interfingers northward with the Vaqueros Sandstone, which in turn changes facies rapidly to the north. About 500 ft north of the Zayante- Vaqueros contact along Zayante canyon, the massive Vaqueros Sandstone includes molluscan bioherms, composed largely of the relatively shallow-water pelecy- pods Dosinia and Crassatella. One to 2 miles farther north along Zayante Canyon where the Vaqueros Sandstone is exposed on the north limb of a syncline, the sandstone is finer grained and includes numerous interbeds of mudstone. Mollusks appear to be absent there, but diverse foraminiferal faunas of Zemorrian age have been collected from the mudstone interbeds by McCollom (1959) and by the senior author. Throughout this Vaqueros section costate uvigerinids and costate siphogenerinids, globose Gyroidinas, and large Cyclamminas are common to abundant, indicating bathyal depths. While alluvial fans of the Zayante Formation were deposited adjacent to the Zayante fault, 1 to 2 miles north, bathyal marine sedimentation continued without interruption during deposition of the San Lorenzo and Vaqueros Formations (geologic section A—A’, pl. 1). McCollom (1959, p. 36—37) has shown that within the central Santa Cruz Mountains the Vaqueros Sand- stone thickens and becomes coarser grained toward the southwest; he postulates that this formation “seems to have had its source area somewhere to the southwest, perhaps in the vicinity of the present day Ben Lomond Mountain.” Brabb (1960, p. 101—116), on the other hand, believes that the source area was more westerly, from what is now the continental shelf. We believe that the Zayante and Vaqueros Sandstones represent orogenic deposits derived from an extensive emergent crystalline terrane south of, and in close proximity to, the Zayante fault. The southwestern extent of this emergent terrane is uncertain. It included crystalline rocks of the Ben Lomond Mountain area but most probably did not include the tectonic block to the west of the San Gregorio fault (fig. 1), because mudstone beds that crop out along the present coast west of this fault and 12 miles west of the mapped area have recently yielded deep water foraminifers diagnostic of late Zemorrian age. The Lambert Shale is conformable upon the Vaqueros Sandstone and records a decrease in the influx of coarse clastic detritus into the basin. This decrease suggests that the terrane uplifted south of the Zayante fault during Zemorrian time had been reduced to a lowland by early Miocene (Saucesian) time. TERTIARY STRATIGRAPHY OF NORTHERN GABILAN RANGE Sedimentary strata exposed in the northern Gabilan Range southwest of the San Andreas fault are of Ter- tiary and Quaternary age. The Tertiary section ranges in age from Eocene to Pliocene and consists of marine and nonmarine clastic sedimentary strata with inter- bedded volcanic flows and agglomerate. This section is more than 10,000 ft thick and is composed of an Eocene to lower Miocene sequence and the Pliocene Purisima Formation (fig. 4). The Eocene to lower Miocene sequence is reported by Allen (1946, p. 27) to rest upon the crystalline base- ment northwest of the town of San Juan Bautista. It includes the San Juan Bautista Formation (siltstone and sandstone) of Kerr and Schenck (1925) , the Pine- cate Formation (massive sandstone) of Kerr and Schenck (1925), the red beds of Kerr and Schenck (1925) , and extrusive andesitic volcanic rocks. Foraminifera from siltstone beds in the lower part of the San Juan Bautista Formation indicate that these beds were laid down at bathyal depths (Castro, 1967; Waters, 1968), whereas the rest of the sequence was deposited under shallow-marine and terrestrial condi- tions. The conglomeratic red beds (fig. 5) were inter- preted by Allen (1946, p. 29) “to be due to sea-cliff erosion along a shoreline, or possibly fanglomerate accumulation resulting from uplift of the Gabilan GEOLOGIC SECTION SOUTHWEST OF SAN ANDREAS FAULT _ 9 .I < 8 SE 3 z 5 u: THICK- E g E g FORMATION LITHOLOGY NESS DESCRIPTION ””4 d E '- (feet) u.| < <0 ‘0 n: O u. e "' ''''' m . . . . . 8 Purisima Formation 2000+ Masswe yellownsh to light-gray sandstone With sult- _ stone, conglomerate, and coquma mterbeds a’ Not in surface contact “l c E .g _ 1000- Dacitic and andesitic flows and agglomerate with light- 8 8 Volcanic 'OCkS 1400 brown arkosic sandstone interbeds 3 s a r Red beds of Kerr and 0—1200 Red pebble and boulder breccia and conglomerate with C Schenck (1925) interbedded red and yellow arkosic sandstone g .E Ll.| ° “ 0 g E Pinecate Formation of Kerr 650— Massive yellow arkosic sandstone with a few interbeds 8 .9 ,3 and Schenck (1925) 1100+ of pebble and boulder conglomerate g s - U a 3 2 .2 o ‘5’ “ “‘5 3 - n: 8 _J 8 c San Juan Bautista 1800— Poorly bedded buff sandstone and interbedded gray to Lu 2 Formation of Kerr dark-brown siltstone; lower part chiefly siltstone N 5000+ 5 'g and Schenck (1925) 0 Z O _ _ _ m 1: .‘E .52 a L“ I) FIGURE 4.—Composite stratigraphic section of Tertiary rocks of northern Gabilan Range southwest of San Andreas fault (Kerr and Schenck, 1925; Allen, 1946; McCroden, 1949; Castro, 1967; Waters, 1968; Turner, 1968; T. W. Dibblee, J r., written commun., 1969). Range, probably by initial movements along the Vergeles fault.” FIGURE 5.——Red beds of Kerr and Schenck (1925), dipping south along San Juan Grade, 11/2 miles northeast of Vergeles fault. The Purisima Formation, of Pliocene age, rests upon the crystalline basement near Logan but is not in sur- face contact with the Eocene to lower Miocene sequence. Here, it is more than 2,000 ft thick and con- tains abundant mollusks, indicating shallow-marine deposition. CORRELATION WITH CENTRAL SANTA CRUZ MOUNTAINS Although the Eocene to lower Miocene sequence of the northern Gabilan Range was probably deposited within the same basin as the correlative section in the central Santa Cruz Mountains, the lack of continuity of surface exposures and lateral facies changes between these two areas have resulted in separate rock-strati- graphic nomenclatures. The fine-grained lower part of the San Juan Bautista Formation is correlative with part of the Butano Sandstone to the northwest (Waters, 1968). The coarser grained upper part of the San Juan Bautista Formation and the overlying Pine- cate Formation are partly correlative with the San 10 OLIGOCENE STRATIGRAPHY, TECTONICS, AND PALEOGEOGRAPHY, CALIFORNIA COAST RANGES Lorenzo and Vaqueros Formations of the Santa Cruz Mountains. The red beds appear to correlate with the Zayante Sandstone to the northwest. Although Kleinpell (1938, p. 114) postulated a Saucesian age for these red beds, a potassium-argon date of 21.6:0.7 my. on dacite from near the base of the volcanic section suggests a Saucesian age for the volcanic rocks and a Zemorrian to earliest Saucesian age for the underlying red beds (Turner, 1968, p. 55). Because of the probable genetic relation of the red beds to movement along the Vergeles fault, the radiometric dating is significant, for it sug- gests that uplift along the Vergeles and Zayante faults was approximately contemporaneous. ZAYANTE FAULT The Zayante fault was originally mapped by Bran- , ner, Newsom, and Arnold (1909) from the vicinity of Boulder Creek slightly south of east to the eastern edge of the Santa Cruz quadrangle. This fault continues 11 miles farther southeast to the vicinity of Corralitos, where it is covered by Quaternary sediments (pl. 1). Westward from Boulder Creek, the Zayante fault has been mapped to its juncture with the Ben Lomond fault, about 2 miles west of the Santa Cruz—San Juan Bautista area (Brabb, 1964, fig. 1). Because of the dense vegetation and deep weather- ing in the area, the Zayante fault is poorly exposed; its trace is determined largely from structural and strati- graphic discordances of adjacent beds. Where the fault is exposed along Bear Creek Road about half a mile east of Boulder Creek, the Butano Sandstone and the Vaqueros Sandstone are separated by a 30-ft wide, near-vertical shear zone. One to 2 miles farther east, several minor fractures branch off from the Zayante fault at acute angles. In the vicinity of Zayante canyon, the Zayante Sandstone and the lower part of the Butano Sandstone are juxtaposed by the fault, indi- cating about 4,000 ft of dip separation with the north side relatively downthrown. Farther east, the fault brings the Purisima Formation into contact with the Butano Sandstone for about 31/2 miles, and its con- tinuation to the vicinity of Corralitos is marked by anomalies in the structural attitudes of the Purisima Formation. Only locally is the fault expressed topographically (fig. 6). East of Zayante canyon, a tributary to Zayante Creek flows southward until it reaches the fault, then turns abruptly westward to follow the fault to Zayante Creek. The relatively straight course of the Zayante fault across the canyons and ridges of this area indicates that the fault is nearly vertical. The Ben Lomond fault, as mapped by Branner, New- som, and Arnold (1909) and locally by Brabb (1964), FIGURE 6.——Westward-trending valley along Zayante fault east of Zayante canyon, looking southwest toward Ben Lomond Mountain at skyline. Tb, Butano Sandstone; sz, Zayante and Purisma Formations; Tm, Monterey Formation; g'rnr, granitic and metasedimentary rocks, is an arcuate fracture that runs east and north of Ben Lomond Mountain for a distance of about 15 miles. The mapping and stratigraphy of the Felton area indicate that south of Boulder Creek this fault is a relatively minor structure, with a dip separation of less than 500 ft (geologic section A—A’, pl. 1). In contrast, gravity measurements and stratigraphic work west of Boulder Creek suggest dip separations of 3,000—10,000 ft along the northwestern part of the Ben Lomond fault (Cum- mings and others, 1962, p. 215). Because the suggested amount and sense of displacement on the north flank of Ben Lomond Mountain are of the same order of magnitude as that along the Zayante fault to the east, we believe that the northwestern part of the Ben Lomond fault of earlier mappers is the westward con- tinuation of the Zayante fault. In this report, the name “Ben Lomond” is restricted to the relatively minor dislocation that trends southeastward from near Boulder Creek, through the town of Ben Lomond, to the vicinity of Felton (pl. 1) . VERGELES FAULT The Vergeles fault separates the crystalline rocks of the northern Gabilan Range from the volcanic and sedimentary rocks of the Eocene to lower Miocene sequence. South of San Juan Bautista, this fault curves eastward and trends into the San Andreas fault about 2 miles east of the mapped area. To the northwest the Vergeles fault is covered by Quaternary sediments. Where exposed along San Juan Grade, this fault is reported by Allen (1946, p. 57) to strike N. 60—800 W. and to dip 7 0—800 S. with vertical striations and grooves in rhyolite. Allen (1946, p. 57) states, “***Minimum vertical displacement on the fault is at least 3500 ft, the total being undoubtedly much more.” The similarities in the geometry and the middle BOUGUER GRAVITY FIELD 11 Tertiary history of the Vergeles and Zayante faults suggested to the writers that the Vergeles fault was a southeastward continuation of the Zayante fault with the continuity now obscured by Quaternary deposits for about 17 miles. The gravity data of this report con- firm the continuity of these two major faults. BOUGUER GRAVITY FIELD The Bouguer gravity field (pl. 1) is composed of a series of northwest-southeast-trending highs and lows superimposed upon a regional gravity gradient that decreases eastward at approximately 1 mgal per mile. Within the mapped area, Bouguer gravity values range from a maximum of more than +36 mgal at Ben Lomond Mountain to a minimum of less than —40 mgal east of San Juan Bautista. The Bouguer anomalies correlate well with the known geology and are generally alined with the regional structural trend. Positive anomalies are asso- ciated with exposures of the relatively dense rocks of the crystalline basement and the Franciscan Forma- tion. Negative anomalies correspond with thick sec- tions of Tertiary sedimentary rocks. Steep gravity gradients are associated with many of the known faults where rocks of contrasting densities have been juxta- posed at shallow depths. The prominent positive anomaly that is partly defined in the western part of the survey area is asso- ciated with crystalline rocks of the Ben Lomond Moun- tain area and extends approximately 15 miles to the northwest beyond the mapped area (Chapman and Bishop, 1968). On the northeast flank of Ben Lomond Mountain, a steep gravity gradient is associated with the Ben Lomond fault from near Boulder Creek south- eastward to Felton, where surface expression of the fault disappears. Here, the gravity contours trend northeastward, extending the gravity high in that direction. Scattered outcrops of granitic rocks south and east of Scotts Valley also confirm that the crystal- line basement is shallow in this area. In the vicinity of Boulder Creek, the northwest- trending gravity gradient that is related to the density contrast between the crystalline complex of Ben Lomond Mountain and the middle Miocene sequence to the east is oblique to the trend of the Zayante fault. In this area and to the east for 4 miles, the Zayante fault is not reflected in the gravity field, and a north- west-striking bench extends across the Zayante fault. Although the Tertiary section south of the fault is thinner than that to the north, the southern section has a greater density contrast with the crystalline basement that partially offsets the gravitational effects of the thickness difference. To the south, the section includes nearly 3,000 ft of relatively low-density mud- stones of the Monterey Formation, whereas north of the fault, the thicker sedimentary section is composed largely of higher-density Vaqueros, Zayante, and Butano Sandstones. Farther southeast where the low- density Miocene a'nd Pliocene sections are thinner and granitic rocks are exposed locally south of the Zayante fault, the course of the fault is approximated by a steep gravity gradient. An elongate northwest-trending gravity maximum is associated with the crystalline rocks of the Gabilan Range (Bishop and Chapman, 1967). This anomaly extends into the southern part of the survey area, where it forms two distinct highs. The saddle between these two maximums coincides with the northwestern part of the exposed trace of the Vergeles fault. South and west of this fault, the positive Bouguer anomaly is associated with the outcrops and subsurface extension of the crystalline complex. The eastern anomaly between Pinecate Peak and Logan is associated with the distinctive gabbroic rocks that are exposed along the southwestern side of the San Andreas fault. The gravity anomaly suggests that in the subsurface this body extends southwest of the known exposures. The eastern part of the Vergeles fault appears to lack any obvious gravity expression. This may result in part from limited gravity control in this area of steep gravity gradients. East of San Juan Bautista the gravity field is domi- nated by a steep gradient of approximately 25 mgal per mile associated with the San Andreas fault. Here, crys- talline rocks southwest of the fault have been juxta- posed against approximately 7,000 ft of Pliocene and Pleistocene sedimentary rocks to the northeast (geo- logic section C—C’, pl. 1). West of the gravity high associated with the north- westward extension of the crystalline rocks of the Gabilan Range, the gravity field decreases in response to a deepening of the basement beneath the Salinas trough. A gravity low extending eastward from Moss Landing and represented as a flexure in the contour lines has been interpreted by Starke and Howard ( 1968) as reflecting a buried eastward extension of the Monterey submarine canyon. The Bayside Develop- ment Vierra 1 well (pl. 1, N0. 17), drilled within this low to a total depth of 7,916 ft, failed to reach the granitic basement. Between the Ben Lomond Mountain and Gabilan Range basement outcrops, a broad saddle in the Bou- guer anomaly field south of Corralitos corresponds to a depression in the crystalline basement. The general form of this depression as inferred from exploratory well data (pl. 1, Nos. 6, 7, 8, 12, 13, 14, and 15) gen- erally follows the trend of the Bouguer anomaly contours. 12 OLIGOCENE STRATIGRAPHY, TECTONICS, AND PALEOGEOGRAPHY, CALIFORNIA COAST RANGES One of the more prominent features of the gravity field is the northwest-trending gravity minimum that extends from just west of Pinecate Peak to the northern limit of the mapped area. This anomaly reaches a minimum value of less than — 22 mgal about 2 miles east of Corralitos. The surface geology of the central Santa Cruz Mountains suggests that the anomaly is produced by a thick section of Tertiary sedimentary rocks. The northeastern boundary of this elongate low corresponds with the surface trace of the San Andreas fault. To the northwest along this fault, the Tertiary section is in contact with Franciscan base- ment and Upper Cretaceous rocks. In the central part of the area, a flattening of the gravity anomaly north- east of the San Andreas fault is produced by a 7,000-ft section of lower Tertiary sedimentary rocks. Farther east the gravity field rises where the Franciscan base- ment is exposed east of the Sargent fault. In the southern part of the area, the southwestern boundary of this prominent gravity minimum corre- sponds with the surface trace of the Vergeles fault. A gravity gradient of approximately 6 mgal per mile parallels the postulated northwestern extension of the Vergeles fault. This gradient is continuous with the gradient that marks the southeastern trace of the mapped Zayante fault and is interpreted as probable confirmation of the subsurface continuity of the Zayante and Vergeles faults. The location of the con- cealed part of the Zayante-Vergeles fault, as shown on plate 1, is based upon a relatively straight-line connec- tion bowed slightly southwest around the Corralitos gravity low between the mapped portions of the Zayante and Vergeles faults. This elongate negative anomaly is interpreted as being produced by a thick Tertiary sedimentary section that is bounded on the northeast by the San Andreas fault and Franciscan basement and on the southwest by the continuous Zayante—Vergeles fault and rela- tively shallow Salinian basement. Additional confir- mation of this interpretation is provided by surface geology, well data, and quantitative models of the gravity field. More than 14,000 ft of Tertiary section has been mapped in the central Santa Cruz Mountains. The Union Oil Teresa Hihn 1, the Texas Blake 1, and the Occidental Petroleum Bingaman 1 wells (pl. 1, Nos. 3, 5, 16), located on the north (downdropped) side of the Zayante-Vergeles fault, were all drilled to total depth greater than 7,000 ft and failed to reach base- ment. Other wells (pl. 1, Nos. 6, 7, 8, 12, 13, 14, 15, 18, 19) southwest of the Zayante-Vergeles fault pene- trated the Salinian basement at relatively shallow depths. The Texas Blake 1 well is reported (A. J. Mac- millan, written commun., 1969) to have bottomed in “Lower Miocene sands” at a total depth of 7,522 ft. If the subsurface thickness of the pre-lower Miocene sec- tion east of Corallitos approximates the 10,000 ft exposed in the central Santa Cruz Mountains, then the Tertiary section in this area could exceed 17,000 ft. MODEL STUDIES OF GRAVITY FIELD Three density models, corresponding to geologic cross sections A—A’, B—B’, and 0—0 (pl. 1), have been constructed to determine if the geology, as presented, can adequately account for the observed Bouguer grav- ity. These density models support the interpretation that the Zayante and Vergeles faults are continuous and also give approximate depths to the basement. The models are necessarily limited by incomplete infor- mation on the regional gravity field and on the distri- bution and variation of rock densities. A regional gravity field was constructed in order to eliminate from the models the gravitational effects of deep crustal structure, large-scale density variations within the basement, and density variations outside the model area. The regional field was derived from gravity data from this survey, published data from the Santa Cruz and San Francisco sheets of the Bouguer gravity map of California (Bishop and Chapman, 1967; Chapman and Bishop, 1968), and a few regional sta- tions obtained by the writers. Regional profiles were drawn on north-south and east-west lines at 5-mile intervals across the survey area. These profiles were required to be without fluctuation and always above the observed Bouguer field. The regional gravity field (fig. 7) was constructed by contouring the regional pro- files within the survey area. The regional field, as constructed, generally de- creases eastward at approximately 1 mgal per mile. The most prominent feature is a high centered on Ben Lomond Mountain. This high is primarily a reflection of the relative magnitude of the Bouguer anomaly associated with this crystalline complex as compared with anomalies associated with other Salinian com- plexes of similar dimensions. This suggests that the crystalline complex of Ben Lomond Mountain is denser, particularly at depth. The residual Bouguer anomaly, which results from subtracting the regional gravity from the observed Bouguer gravity, represents the gravitational effects of density variations within the sedimentary section and between the sedimentary and basement rocks. The residual anomaly is used for comparison with the grav- ity computed from the models (pl. 1) . The models used to calculate gravitational effects are essentially density contrast models. Structural and stratigraphic variations are converted into density con- trasts with the basement rocks and the gravitational effects calculated by computer. The computer com- BOUGUER GRAVITY FIELD 13 7' a 122° 00' Santa Cruz 0 2 4 6 8 10 MILES I I I I | | CONTOUR INTERVAL 1 MILLIGAL 36° 45' 37° 00' ' % \ 0 Cl San’ Juan Bautista . / °/ 121°45' 121°30' FIGURE 7.—Regional Bouguer gravity field of the Santa Cruz—San Juan Bautista area. Sections are shown on plate 1. pared calculated and residual gravity values and, within certain constraints imposed by the known sur- face geology, well data, and assumed rock densities, altered the models to give a desired fit. In order to achieve a fit within the survey area, the geology was extended approximately 8 miles beyond the ends of the cross sections. In this modeling scheme, any vari- ance in the true value of rock densities from that assumed in the models will result in an error in estima- tion of basement depth. Samples for density determinations were collected from sedimentary and basement rocks throughout the mapped area. Wherever possible, wet densities were measured. Where not measured, wet density was calcu- lated using a method described by Byerly (1966). In addition to measured densities, we used values reported by Brabb (1960) and Byerly (1966) and in several unpublished student reports at the University of Cali- fornia at Santa Barbara. A density of 2.67 g/cm3 (grams per cubic centimeter) was accepted as an “aver- age” for Franciscan basement rocks (Clement, 1965; Thompson and Talwani, 1964). The measured densities and the values we used for modeling are listed in table 1. In addition, a density of 2.17 g/cm3 was assumed for for the thick Pliocene section east of the San Andreas fault near San Juan Bautista in order to allow for density increase with depth because of compaction and cementation. 14 OLIGOCENE STRATIGRAPHY, TECTONICS, AND PALEOGEOGRAPHY, CALIFORNIA COAST RANGES TABLE 1.—Measured rock densities and model densities in grams per cubic centimeter Number of Measured density Model Rock type Samples Mean Range density Miocene and Pliocene mudstones ................ 2.05 1.60—2.25 1.97 Middle Tertiary sandstones and siltstones .................. 27 2.35 1.90—2.40 2.37 Eocene and Creta- ceous sandstones, well cemented __________ 15 2.48 2.35—2.62 2.47 Granitic basement ...... 18 2.70 2.65—2.74 2.67 Franciscan basement .. ...................... 2.67 To simplify the computations, a model density of 2.67 g/cm" was assigned to both the Franciscan and granitic basement rocks, and model densities for other rocks were selected to produce a density contrast with the basement rocks of an even multiple of 0.1. Three density models, corresponding to geologic cross sections A—A’, B—B’, and C—C’, are shown on plate 1 with the associated computed Bouguer anoma- lies and residual Bouguer anomalies. The program used to calculate the gravitational effects of the density models is similar to the one described by Talwani, Wor- zel, and Landisman (1959), and has been modified to take in account and effects where a strict two-dimen- sional model is not applicable (Nettleton, 1940, p. 117). Minor modifications in basement depths and sub- surface fault locations were made to the original den- sity models to improve the fit between the computed and residual anomalies. Comparison of the density models with the geologic cross sections shows the models to be in reasonable agreement with the extrapolated geology. These geo- logic sections were constructed prior to the analysis of the gravity data and have not been altered by the model studies. Where the density models and the geol- ogy are in agreement, we believe that the indicated depths to basement are reasonable. Elsewhere, the differences may result either from density variations not accounted for in the models or from a lack of sub- surface geologic control. Although the Zayante fault is not associated with a steep gravity gradient along section A—A’, the model indicates that the granitic basement is displaced approximately 6,000 ft by this fault. These gravity data also suggest that the Tertiary section between the Zayante and San Andreas faults is several thousand feet thinner and the basement correspondingly shal- lower than shown on the geologic cross section. Along section B—B’, the geology and model indicate that the granitic basement is offset approximately 8,500 ft and 9,000 ft, respectively, by the subsurface continuation of the Zayante and Vergeles faults. The gravity data along this section support the interpreta- tion that the granitic basement rises abruptly south- west of the San Andreas fault. Subsurface geologic control is not sufficient to determine whether this gra- nitic uplift results from folding or faulting. The vertical offset of the granitic basement by the Vergeles fault (section C—C’, pl. 1) is estimated by the geology to be 9,000 ft and by the gravity to be approx- imately 8,500 ft. The similarity of these estimates results largely from the subsurface control provided by the Occidental Bingaman well, which constrained the modeling program. The gravity data suggest that the granitic basement rises more rapidly east of this well than is shown on the geologic cross section. ZAYANTE-VERGELES FAULTING AND MIDDLE TERTIARY PALEOGEOGRAPHY The Paleocene and Eocene to lower Miocene sequences of the central Santa Cruz Mountains are strikingly similar to the correlative sections in the northern Santa Lucia Range, 60 miles to the south (Clark, 1968, p. 172—174). Dickinson (1956, p. 132) has suggested that the Church Creek Formation in the northern Santa Lucia Range and the San Lorenzo For- mation of the Santa Cruz Mountains may have been deposited within the same basin. The San Juan Bau- tista Formation of Kerr and Schenck (1925) also would have been deposited within this basin. According to Addicott’s (1968, p. 147—148) paleogeographic recon- struction, during Refugian time this basin extended southeast of San Juan Bautista with an eastern strand- line that trended southward through the Santa Lucia Range. Although the proposed continuity of these deposi- tional basins is largely inferred from limited outcrop data, the gravity analysis of this report confirms the continuity of a thick Tertiary sedimentary section between the central Santa Cruz Mountains and the northern Gabilan Range. Thus the analysis supports the interpretation that the Santa Cruz basin extended southeastward to the San Juan Bautista area during early Tertiary time. Had the basin continued south- ward to the Santa Lucia area, as suggested by Dickin- son and Addicott, the Zemorrian uplift along the Zayante-Vergeles fault would have separated the two areas. This faulting strongly influenced the Zemorrian paleogeography of the Santa Cruz Mountains and northern Gabilan Range. The shallowing recorded by the upper part of the San Lorenzo Formation in the central Santa Cruz Moun- tains probably reflects initial movement along the Zayante-Vergeles fault. Continued uplift along this fault resulted in emergence of the crystalline basement to the south, deposition of terrestrial beds—the Zayante Sandstone and the red beds of Kerr and Schenck (1925)—along the fault, and restriction of marine con- ZAYANTE-VERGELES FAULTING AND MIDDLE TERTIARY PALEOGEOGRAPHY 15 ditions to an embayment to the north, the Santa Cruz Bay of Loel and Corey (1932, map 1). Thus uplift produced a northwest-southeast-trending, northeast- facing Zemorrian shoreline Whose position was in close proximity to the fault. The positive block south of the fault, which included both the Ben Lomond and Gabi- lan Mountain blocks of Clark (1930) and is here refer- red to as the Ben Lomond—Gabilan block, separated the Santa Cruz basin from the Santa Lucia basin to the south. Starke and Howard (1968) have recently postulated a terrestrial origin for the ancestral Monterey canyon. If their hypothesis is correct, this canyon may have been eroded subaerially during Zemorrian time by streams draining the southern slope of the emergent Ben Lomond—Gabilan block. The Zayante-Vergeles fault ceased to be an impor- tant structure by early Miocene (Saucesian) time. The fine clastic deposition of the Lambert Shale north of the Zayante fault suggests that the Ben Lomond— Gabilan block had been reduced to a lowland, which was subsequently transgressed by the middle Miocene seas with deposition of the Monterey Formation. East of Ben Lomond Mountain, the middle Miocene sequence coarsens toward the west, and on both slopes of the mountain, the Monterey Formation contains thick sandstone interbeds. These relations and marked differences in the middle Miocene faunas on opposite sides of the mountain indicate that the crystalline com- plex of Ben Lomond Mountain was locally high. Whether this topography represented a residual high or resulted from local upwarping of the basement dur- ing middle Miocene time is uncertain. Later vertical movements occurred along the Zayante-Vergeles fault. Post-early Miocene displace- ment is indicated by grooving in volcanic rocks in the northern Gabilan Range and post-middle Miocene displacement by tectonic slivers of the Monterey For- mation along the fault in the central Santa Cruz Moun- tains. The later displacement may have occurred in post-early Pliocene time, for the Purisima Formation is offset along this fracture. The western segment of the Zayante fault (the Ben Lomond fault of earlier workers) appears to have been inactive in post-late Miocene time, as 8 miles west of the mapped area, the Santa Cruz Mudstone of Clark (1966b) is unaffected by the fault. Although several recent earthquake epicenters have been plotted between Watsonville and San Juan Bau- tista near the trace of the Zayante-Vergeles fault (California Department of Water Resources, 1964; Brabb, 1967), Quaternary deposits in the area do not appear to have been offset by the fault, and these earthquakes, including the magnitude 5.2 shock of September 14, 1963, are liklier to be related to move- ment along the San Andreas fault. RELATION TO SAN ANDREAS FAULTING Recent right-lateral slip has been documented along the mapped segment of the San Andreas fault. Surface rupture along the fault during the San Francisco earth- quake of April 18, 1906 extended southward through the area to one-half mile northwest of San Juan Bau- tista (Lawson and others, 1908, p. 38). Right-lateral movement along the fault during the 1906 earthquake produced horizontal displacements of 4 ft about 3 miles northwest of San Juan Bautista, 31/2 ft between rail- road bridge abutments at Pajaro Gap, and 5 ft in a railroad tunnel at Wrights Station about 1 mile west of Lake Elsman (Lawson and others, 1908, p. 38, 111— 113). Active deformation with right-lateral offset of roads and fences has been recorded along the trace of the San Andreas fault at San Juan Bautista (Nason and Rogers, 1967). The history of this segment of the San Andreas fault has remained controversial. The marked similarity of the “lower Miocene volcanics, red beds, and marine lower Miocene and Oligocene strata” of the northern Gabilan Range west of the fault with the correlative section of the San Emigdio Mountains at the southern end of the San Joaquin Valley east of the fault sug- gested to Hill and Dibblee (1953, p. 448—449) approxi- mately 175 miles of right-lateral offset along the San Andreas since early Miocene time. Other workers (Cummings and others, 1962; Oakeshott, 1966) , noting the similarity in the stratigraphic sequences on oppo- site sides of the San Andreas fault in the northern Santa Cruz Mountains, have considered such large- scale lateral displacement along this segment of the fault to be improbable. Recent stratigraphic work by Clark (1968) and radiometric dating by Turner (1969) have revealed that the Tertiary stratigraphy of the Santa Cruz Mountains does not preclude large cumulative slip along this segment of the fault. Oligocene and early Miocene zoogeographic and paleogeographic analysis by Addicott (1968) and detailed comparison and radiometric dating of the lower Miocene volcanic rocks of the northern Gabilan Range and of the San Emigdio Mountains by Bazeley (1961) and Turner (1969) have strengthened substantially Hill and Dibblee’s hypothe- sis for a 175-mile displacement. The time of initiation of right-lateral slip on the San Andreas fault system is uncertain. Page (1970, p. 685) believes that there is little geologic evidence to indicate that the San Andreas fault system existed prior to Oli- gocene time and even less evidence that it existed prior to the Eocene. Crowell (1968, p. 327), working in the 16 OLIGOCENE STRATIGRAPHY, TECTONICS, AND PALEOGEOGRAPHY, CALIFORNIA COAST RANGES central Transverse Ranges of southern California, has suggested that San Andreas faulting originated in early Miocene or possibly during Oligocene time. This dating is supported by the analysis of magnetic anomalies in the northeastern Pacific basin by McKenzie and Mor- gan (1969, p. 130), who state, “***It is difficult to understand how the right lateral motion on the San Andreas and related faults can have begun on a large scale before the Oligocene** *.” An Oligocene or later origin for the right-lateral slip on the San Andreas fault system and the postulated 175-mile post-early Miocene displacement are difficult to reconcile with an apparent 350-mile offset of the Franciscan basement—Sierran basement contact (Hill and Dibblee, 1953; Hamilton, 1969, p. 2415). A pos- sible resolution of this apparent conflict has recently been suggested by Atwater (1970, p. 3516ff.). On the basis of her study of Pacific Ocean magnetic anomaly patterns and plate theory, Atwater postulates two epi- sodes of right-lateral slip on the San Andreas, a late Mesozoic and early Cenozoic episode and a more recent episode that she believes started not earlier than 30 my. (Oligocene). If Atwater’s later episode of San Andreas motion commenced during the Oligocene, then the initiation of this displacement on the San Andreas fault system and the initial movement of the Zayante-Vergeles fault were approximately contemporaneous. The Oligocene stratigraphy of the central Santa Cruz Mountains and of the northern Gabilan Range does not appear to have been influenced by topography alined with the San Andreas fault but rather was controlled by uplift along the Zayante-Vergeles fault. The stratigraphic record suggests that Zayante-Vergeles faulting may have pre- ceded the later Cenozoic movement on the San Andreas fault, or alternatively that Oligocene displacement along this segment of the San Andreas lacked a signifi- cant component of dip slip. Although large-scale post-early Miocene displace- ment along the San Andreas fault now appears prob- able, several lines of evidence suggest that lateral move- ment has not been transferred from the San Andreas to the northwest-trending Zayante-Vergeles fault. Metasedimentary clasts in the red beds of the San Juan Bautista area were clearly derived from metamorphic pendants of the Gabilan Range to the south directly across the Vergeles fault. An areally restricted bio- strome composed of the tests of the irregular echinoid Astrodapsis spatiosus Kew in the upper part of the Santa Margarita Sandstone in the central Santa Cruz Mountains has a nearly contiguous distributional pat- tern across the Zayante fault.1 Verticle grooves in the 1A critical northern exposure of Santa Margarita Sandstone along Zayante canyon is not shown on plate 1 because of scale. volcanic rocks of the northern Gabilan Range suggest that the latest movement along the Vergeles fault was vertical rather than lateral. Strike-slip displacement does not appear to have been superposed upon the Cenozoic high-angle movements of the Zayante-Ver- geles fault. REFERENCES CITED Addicott, W. O., 1968, Mid—Tertiary zoogeographic and paleo- geographic discontinuities across the San Andreas fault, California, 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. 144—165. Allen, J. E., 1946, Geology of the San Juan Bautista quad- rangle, California: California Div. Mines Bull. 133, p. 9—75. American Association of Petroleum Geologists, Committee for Cross Sections (Payne, M. B., chm), 1964, San Andreas fault cross section parallel along the east and west sides of the fault from Bielwaski Mountain to Hollister, Cali- fornia: Am. Assoc. Petroleum Geologists, Pacific Sec., Cross Section CS 17. 1967, San Andreas fault cross sections, east and west longitudinal cross sections parallel to the San Andreas fault and east to west cross sections across the San Andreas fault, California: Am. Assoc_ Petroleum Geologists, Pacific Sec. Atwater, Tanya, 1970, Implications of plate tectonics for the Cenozoic tectonic evolution of western North America: Geol. Soc. America Bull., v. 81, no. 12, p. 3513—3536. Bailey, E. H., and Everhart, D, L., 1964, Geology and quick- silver deposits of the New Almaden district, Santa Clara County, California: US. Geol. Survey Prof. Paper 360, 206 p. Bailey, E. H., Irwin, W. P., and Jones, D. L., 1964, Franciscan and related rocks, and their significance in the geology of western California: California Div. Mines and Geology Bull. 183, 177 p. Bazeley, W. J. M., 1961, 175 miles of lateral movement along the San Andreas fault since lower Miocene [abs]: Pacific Petroleum Geologist Newsletter, v. 15, no. 5, p. 2—3. Becker, G. F., 1888, Geology of the quicksilver deposits of the Pacific slope, with atlas: U.S. Geol. Survey Mon. 13, 486 p. Bishop, C. C., and Chapman, R. H., 1967, Bouguer gravity map of California, Santa Cruz sheet: California Div. Mines and Geology, scale 1: 250,000. Blacut, Gustavo, and Kleinpell, R. M., 1969, A stratigraphic sequence of benthonic smaller foraminifera from the La Boca Formation, Panama Canal Zone: Cushman Found. Foram. Research Contr., v. 20, pt. 1, p. 1—22. Bowen, 0. E., Jr., and Gray, C. H., Jr., 1959, Geology and economic possibilities of the limestone and dolomite depos- its of the northern Gabilan Range, California: California Div. Mines Spec. Rept. 56, 40 p. Brabb, E. E., 1960, Geology of the Big Basin area, Santa Cruz Mountains, California: Stanford Univ., Stanford, Calif, Ph.D. thesis, 192 p. 1964, Subdivision of San Lorenzo Formation (Eocene- Oligocene) west-central California: Am. Assoc. Petroleum Geologists Bull., V. 48, no. 5, p. 670—679. 1967, Chittenden, California, earthquake of September 14, 1963: California Div. Mines and Geology Spec. Rept. 91, p. 45—53. REFERENCES CITED 17 Bramlette, M. N ., and Wilcoxon, J. A., 1967, Middle Tertiary calcareous nannoplankton of the Cipero section, Trinidad, W. I.: Tulane Studies Geology, v. 5, no. 3, p, 93—132. Branner, J. C., Newsom, J. F., and Arnold, Ralph, 1909, Description of the Santa Cruz quadrangle, California: US. Geol. Survey Geol. Atlas, Folio 163. Burchfiel, B. C., 1964, Stratigraphic reassignment of four spe- cies in the lower Miocene rocks of the Bear Creek area, Santa Cruz County, California: J our. Paleontology, V. 38, no. 2, p. 401—405. Burford, R. 0., 1961, Geology of the Glenwood Basin area, Santa Cruz Mountains, California: Stanford Univ., Stan- ford, Calif, graduate report. Byerly, P. E., 1966, Interpretations of gravity data from the Central Coast Ranges and San Joaquin Valley, California: Geol. Soc. America Bull, V. 77, no. 1, p, 83—94. California Department of Water Resources, 1964, Earthquake epicenter and fault map of California: California Dept. Water Resources Bull. 116—2, pt. 3. California Division of Mines and Geology, 1965, Potassium- argon age dates for some California localities: California Div. Mines and Geology Mineral Inf, Service, v. 18, no. 1, p. 16. Castro, M. J ., 1967, Geology and oil potential of the area west- erly of San Juan Bautista, California, in Guidebook Gabilan Range and adjacent San Andreas fault: Am. Assoc. Petroleum Geologists and Soc. Econ. Paleontolo- gists, Pacific Sec., Joint Ann. Field Trip, 1967, p. 81—86. Chapman, R. H., 1966, The California Division of Mines and Geology gravity base station network: California Div. Mines and Geology Spec. Rept. 90, 49 p. Chapman, R. H., and Bishop, C. C., 1968, Bouguer gravity map of California, San Francisco sheet: California Div. Mines and Geology, scale 1 : 250,000. Christensen, R. S., 1960, Middle Tertiary Foraminifera from Soquel Creek, Santa Cruz County, California: California Univ. (Berkeley) M. A. thesis. Clark, B. L., 1930, Tectonics of the Coast Ranges of middle California: Geol. Soc. America Bull., v. 41, p. 747—828. Clark, J. C., 1966a, Tertiary stratigraphy of the Felton-Santa Cruz area, Santa Cruz Mountains, California: Stanford Univ., Stanford, Calif, Ph.D, thesis, 184 p. 1966b, Tertiary stratigraphy of the Felton-Santa Cruz area, Santa Cruz Mountains, California [abs]: Dissert. Abs., v. 27, no. 4, p. 1184—B. 1968, Correlation of the Santa Cruz Mountains Tertiary —implications for San Andreas history, 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. 166—180. Clement, W. G., 1965, Complete Bouguer gravity map of the northern part of the San Francisco Bay area and its geo- logic interpretation: U.S. Geol. Survey Geophys. Inv. Map GP—468, scale 1: 125,000. Compton, R. R., 1966, Granitic and metamorphic rocks of the Salinian block, California Coast Ranges, in Bailey, E. H., ed., Geology of northern California: California Div. Mines and Geology Bull. 190, p. 277—287. Crowell, J . C., 1968, Movement histories of faults in the Trans- verse Ranges and speculations on the tectonic history of California, 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. 323—341. Cummings, J. C., Touring, R. M., and Brabb, E. E., 1962. Geol- ogy of the northern Santa Cruz Mountains, California: California Div. Mines and Geology Bull. 181, p. 179—220. Dickinson, W. R., 1956, Tertiary stratigraphy and structure west of the Arroyo Seco, Monterey County, California: Stanford Univ., Stanford, Calif, M.S. thesis, 160 p. 1965, Tertiary stratigraphy of the Church Creek area, Monterey County, California: California Div. Mines and Geology Spec. Rept. 86, p. 25—44. Hall, C. A., Jr., Jones, D. L., and Brooks, S. A., 1959, Pigeon Point formation of Late Cretaceous age, San Mateo County, California: Am. Assoc. Petroleum Geologists Bull, V. 43, no. 12, p. 2855—2859. Hamilton, Warren, 1969, Mesozoic California and the underflow of Pacific mantle: Geol. Soc. America Bull, v. 80, no. 12, p. 2409—2430. Hill, M. L., and Dibblee, T. W., Jr., 1953, San Andreas, Gar- lock, and Big Pine faults, California—a study of the char-. acter, history, and tectonic significance of their diSplace- ments: Geol. Soc. America Bull, v. 64, no. 4, p. 443—458. Ingram, R. L., 1954, Terminology for the thickness of stratifi- cation and parting units in sedimentary rocks: Geol. Soc. America Bull, v. 65, no. 9, p. 937—938. Jennings, C. W., and Burnett, J. L., 1961, Geologic map of California, Olaf P. Jenkins edition, San Francisco sheet: California Div. Mines, scale 1:250,000. Jennings, C. W., and Hart, E. W., 1956, Exploratory wells drilled outside of oil and gas fields in California to Decem- ber 31, 1953: California Div. Mines Spec. Rept. 45. Jennings, C. W., and Strand, R. G., 1958. Geologic map of California, Olaf P. Jenkins editionfi Santa Cruz sheet: California Div. Mines, scale 1: 250,000. Kerr, P. F., and Schenck, H. G., 1925, Active thrust-faults in San Benito County, California: Geol. Soc. America Bull., v. 36, no. 3, p. 465—494. Kleinpell, R. M., 1938, Miocene stratigraphy of California: Tulsa, Okla, Am. Assoc. Petroleum GeOIOgists, 450 p. Lawson, A. C., 1893, The post-Pliocene diastrophism of the coast of southern California: California Univ. Pub. Dept. Geol. Bull, v. 1, no. 4, p. 115—160. 1914, Description of the San Francisco district: U.S. Geol. Survey Geol. Atlas, Folio 193. Lawson, A. C., and others, 1908, The California earthquake of April 18, 1906. Report of the State Earthquake Investiga- tion Commission: Carnegie Inst, Washington Pub. 87, 3 v., 1 atlas. Leo, G. W., 1961, The plutonic and metamorphic rcoks of Ben Lomond Mountain, Santa Cruz County, California: Stan- ford Univ., Stanford, Calif, Ph.D. thesis, 194 p. 1967, The Plutonic and metamorphic rocks of the Ben Lomond Mountain area, Santa Cruz County, California: California Div. Mines and Geology Spec. Rept. 91, p. 27—43. Loel, Wayne, and Corey, W. H., 1932, The Vaqueros formation, Lower Miocene of California. I, Paleontology: California Univ. Pubs. Dept. Geol. Sci, Bull, v, 22, no. 3, p. 31—410. McCollom, R. L., J r., 1959, Lithofacies study of the Vaqueros formation, Santa Cruz Mountains, California: Stanford Univ., Stanford, Calif, M.S. thesis, 48 p. McCroden, T. J., 1949, Geology of a portion of the Gabilan Range, California: Stanford Univ., Stanford, Calif, M.S. thesis. McKee, E. D., and Weir, G. W., 1953, Terminology for strati- fication and cross-stratification in sedimentary rocks: Geol. Soc. America Bull, v. 64, no, 4, p. 381—390. 18 OLIGOCENE STRATIGRAPHY, TECTONICS, AND PALEOGEOGRAPHY, CALIFORNIA COAST RANGES McKenzie, D. P., and Morgan, W. J ., 1969, Evolution of triple junctions: Nature, v. 224, p. 125—133. Mallory, V. S., 1959, Lower Tertiary biostratigraphy of the California Coast Ranges: Tulsa, Okla., Am. Assoc. Petro- leum Geologists, 416 p. Martin, B. D., and Emery, K. 0., 1967, Geology of Monterey Canyon, California: Am. Assoc. Petroleum Gologists Bull., v. 51, no. 11, p. 2281—2304. Nason, R. D., and Rogers, T. H., 1967, Self-guiding map to active faulting in the San Juan Bautista quadrangle, con- ference on geologic problems of the San Andreas fault system: Stanford, Calif, Stanford Univ., scale 1:24,000. Nettleton, L. L., 1940, Geophysical prospecting for oil: New York, McGraw-Hill, 444 p. Oakeshott, G. B., 1966, San Andreas fault in the California Coast Ranges province, in Bailey, E. H., ed., Geology of northern California: California Div. Mines and Geology Bull. 190, p. 357—373. Page, B. M., 1970, Sur-Nacimiento fault zone of California: Continental margin tectonics: Geol. Soc. America Bull., v. 81, no. 3, p. 667—690. Rogers, T. H., 1966, Geologic map of California, Olaf P. Jenkins edition, San Jose sheet: California Div. Mines and Geol- ogy, scale 1:250,000. Ross, D. C., 1970, Quartz gabbro and anorthositic gabbro: markers of offset along the San Andreas fault in the Cali- fornia Coast Ranges: Geol. Soc. America Bull., V. 81, no. 12 p. 3647—3662. Schenck, H. G., and Kleinpell, R. M., 1936, Refugian Stage of Pacific Coast Tertiary: Am. Assoc. Petroleum Geologists Bull., V. 20, no. 2, p. 215—225. Starke, G. W., and Howard, A. D., 1968, Polygenetic origin of Monterey Submarine Canyon: Geol. Soc. America Bull, v. 79, p. 813—826. Swick, C. H., 1942, Pendulum gravity measurements and iso- static reductions: U.S. Coast and Geodetic Survey Spec. Pub., 232, 82 p. Talwani, Manik, Worzel, J. L., and Landisman, Mark, 1959, Rapid gravity computations for two-dimensional bodies with application to the Mendocino submarine fracture zone: Jour. Geophys. Research, v. 64, no. 1, p. 49—59. Thompson, G. A., and Talwani, Manik, 1964, Crustal struc- ture from Pacific Basin to central Nevada: J our. Geophys. Research, v. 69, no. 22, p. 4813—4837. Trask, P. D., 1926, Geology of the Point Sur quadrangle, Cali- fornia: California Univ. Pub. Dept. Geol. Sci. Bull., v. 16, no. 6.1). 119—186. Turner, D. L., 1968, Potassium-argon dates concerning the Tertiary foraminiferal time scale and San Andreas fault displacement: California Univ. (Berkeley) Ph.D. thesis, 99 p. 1969, K-Ar ages of California Coast Range volcanics—- implications for San Andreas fault displacement [abs]: Geol. Soc. America, Abstracts with Programs, 1969, pt. 3, p. 70. Waters, J. N., 1968, Eocene faunule from the basal San Juan Bautista formation of California: Cushman Found. Foram. Research Contr. v. 19, pt. 1, p. 18—20. Wilkinson, E, R., 1967, Hollister field, in Guidebook Gabilan Range and adjacent San Andreas fault: Am. Assoc, Petroleum Geologists and Soc. Econ. Paleontologists and Mineralogists, Pacific Sec—Joint Ann. Field Trip, 1967, p. 95—98. U.S. GOVERNMENT PRINTING OFFICE: 1973 0—506-591 UNITED STATES DEPARTMENT OF THE INTERIOR GEOLOGICAL SURVEY Base from U.S. Geological Survey 75’ edition sheets, 1969 2000’ SEA L EV E L 2000’ 4000' 6000’ 8000' 10,000’ 12,000' 14,000’ 16,000' 10 20 30 40 50 60 70 2000' S EA LEV E L 2000' 4000' 6000' 8000’ 10,000' 12,000' 14,000' 16,000' «Rt L La 7 < 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. BEN LOMOND FAULT INDEX TO GEOLOGIC MAPPING 1. Clark (1966a) 2. Brabb, E.E. Burchfiel (1964) Clark, J.C. Dibblee, T.W., Jr. McColIom (1959) 3. Burford (1961) 4. Dibblee, T.W., Jr., modified locally by: Brabb, E.E. Christensen (1960) Vlark, LC. Dibblee, T.W., Jr. Dibblee, T.W., Jr. Allen (1946) Dibblee, T.W., Jr, modified locally by: Nason and Rogers (1967) Clark, J.C. >190- tSequel Cove EXPLORATORY WELLS SHOWN ON MAP Union Oil, Lorna Prieta 1; projected sec. 19, T. 10 8., R. 1 E.;T.D. 1942 feet. Union Oil, Loma Prieta 2; projected sec. 20, T. 10 8., R. 1 E.;T.D. 4570 feet. Union Oil, Teresa Hihn 1; projected sec. 34, T. 10 8., R. 1 E.;T.D. 7747feet. Union Oil, Teresa Hihn 2; projected sec. 34. T. 10 8., R. 1 E.;T.D. 3569 feet. Texas,J.H. Blake 1; projected sec. 12. T. 11 8., R. 1 E.;T.D. 7522 feet. Texas, M. Light 1; projected sec. 14, T. 11 8., R. l E.; T.D. 3694 feet. Texas,Pierce 1; projected sec. 15,T. 11 8., R. 1 E.;T.D. 2670 feet. Texas,Carpenter 1; projected sec. 35,T. 11 8., R. 1 E.; T.D. 3100 feet Fullerton Oil, Peterson 1; projected sec. 27, T. 11 8., R.2E.;T.D.3197 feet. _ Monterey Oil, Porter 1; projected sec. 25, T. 11 8., R. 2 E.; T.D. 2985 feet. Monterey Oil, Porter 2; projected sec. 30, T. 11 8., R. 3 E.; T.D. 3977 feet. Texas, Blake 1; projected sec. 11, T. 12 8., R. 1 E.;T.D. 2463 feet. Western Gulf Oil, Johnson 1; projected sec. 29, T. 12 8., R. 2 E.; T.D. 3198 feet. Texas—Seaboard Oil, Robert Blohm 2; projected sec. 34, T. 12 8., R. 2 E.; T.D. 2188 feet. Jergens Oil, Blohm 1; projected sec. 34, T. 12 8., R. 2 E.;T.D.1935 feet. Occidental Petroleum, Bingaman 1; projected sec. 34, T. 12 8., R. 3 E.; T.D. 7136 feet. Bayside Development, Vierra 1; sec. 7, T. 13 8., R. 2 E.; T.D. 7916 feet. Texas, Pieri 1; projected sec. 19, T. 13 8., R. 2 E.; T.D. 3291 feet. Texas, Davies 1; projected sec. 34, T. 13 8., R. 2 E.; T.D. 2219 feet. ZAYANTE FAU LT Miocene and Pliocene "/ Vinson?“ / \\\ Gravity data by JD. Rietman,_J.C. Clark, and California Division of Mines and COMPUTE D BOUGUER G RAVITY’ RESIDUAL BOUGUE R GRAVITY P (density): 1.97 P =2.47 P =2.67 Geology E APPROXIMATE MEAN DECLINAIION,1973 SCALE 1:125 000 2 O 2 4 6 8 10 MILES I : l : l 1 l—-———-—I )——-————l 2 O 2 4 6 8 10 KILOMETERS l—l l——-i I———-—-——l l—l l———————l CONTOUR INTERVAL 200 FEET DATUM IS MEAN SEA LEVEL Geology compiled by J. C. Clark, 1970 '3 S 3 fl I- '3 I- < (3 _l 3 U) -’ "" C: U) I) < < 3 ‘0 u.I |— < < LL LLI <1: < > _I UJ u. m o: |_ r e I2 95: 5 5' o 4 3 D E LL 2 3 Lu IJJ <2): <3: < Z Z < < 0 0 u. I— < < 2 LL D: E E 3 z > < 5 > w m < I < ‘0 BI 16 CI Tb T" w A . . B 8 N 08' T55 Tem KM 2000' 2000' Trb Ts Oal 2000' . 2000 2000 0a D II Tsj l o: I Kgr SEA LEVEL SEA LEVEL SEA LEVEL SEA LEVEL , SEA LEVEL 2000' 2000' 2000' 2000' 2000' 4000' 4000' 4000' 4000' 4000' 6000' 6000' 6000' 6000' 6000' 8000' 8000' 8000' 8000' 8000' 10,000' 10,000' 10,000' 10,000' 10,000. 12,000' 12,000' 12,000' 12,000' 12,000' 14,000' 14,000' 14,000' 14,000' 14,000' 16,000' 16,000’ 16,000' 16,000' 16,000' GEOLOGIC CROSS SECTIONS A’ B B’ C C, 7 0 o 7 7 o 0 7 L 0 — 10 10 7 7 10 10 7 _ 10 / ' 7’7 20 20 7 7 20 2o 7 - 20 RESIDUAL BOUGUER GRAVITY "’ w RESIDUAL BOUGUER GRAVITY m E E "S. —30§ _30 g 30% 430% E E E .E .E E a e .5 740 g —40 g 40m —40 E (D (D 0 COMPUTED BOUGUER GRAVITY 77 50 COMPUTED BOUGUER GRAVITY W 50 50 — _ 5o ' l — 60 60 — i 60 60 fl _ 60 \ ‘d/ 70 70 7O 70 70 BOUGUER GRAVITY PROFILES A’ B B’ C c' e 2000' 2000' — 2000' 2000' 7 L 2000' — SEA LEVEL SEA LEVEL ‘ SEA LEVEL SEA LEVEL * —— SEA LEVEL 7 2000' 2000' 7 2000' 2000' 7 “1'97 L 2000: P=2.17 7 4000' 4000' 7 4000' 4000' 7 7 4000' — 6000' 6000' _ 6000' 6000' — __ 6000' 7 8000’ 8000' 7 8000' 8000' 7 _ 8000' — 10,000' 10,000' m 10,000' 10,000' —I .2 10,000' 7 12,000' 12,000' 7 12,000' 12,000' 7 P =2.67 7 12,000' — 14,000' 14,000' w 14.000’ 14.000’ “ 7 14,000' 16'000' 16'000' 16'000 16'000 INTERIOR—GEOLOGICAL SURVEY, WASHINGTON, 0.c.71973—G72 300 16000, DENSITY MODELS GEOLOGIC AND GRAVITY MAPS OF THE SANTA CRUZ—SAN JUAN BAUTISTA AREA, SANTA CRUZ, SANTA CLARA, MONTEREY, AND SAN BENITO COUNTIES, CALIFORNIA Holocene Pleistocene Pliocene Miocene Oligocene Eocene Paleocene Pliocene Palocene Upper Jurassic/.7) Upper Miocene and Lower Pliocene PROFESSIONAL PAPER 783 PLATE 1 EXPLANATION SOUTHWEST OF SAN ANDREAS FAULT Oal Surficial deposits Qal, alluvium Os, sand dunes Ols, landslide material River-terrace deposit Includes fan deposits northeast of Watsonville Tl Marine-terrace deposit Aromas Red Sands of Allen (1946) Ligh t—brown to red sands; nonmarine UNCONFORM/TY Purisima Formation Interbedded yellow sandstone and siltstone; marine. Includes Tsc east and south of Scotts Valley a a Santa Cruz Mudstone of Clark (1966b) Pale-yellowish-brown siliceous organic mudstone; marine. West of Santa Cruz locally includes Tsm at base Santa Margarita Sandstone Yellowish-gray to white friable arkosic sandstone; marine Upper Miocene Middle Miocene UNCONFORMITY Tm Monterey Formation Olive-gray, weathers light gray, organic mudstone; marine Lompico Sandstone of Clark (1966b) Yellowish~gray arkosic sandstone; marine UNCONFORM/TY SAN JUAN BAUTISTA AREA BEAN HILL-BEAR CREEK AREA Lambert Shale Dusky-yellowish-brown organic mudstone; marine Volcanic rocks Dacitic and andesitic flows and agglomera te; in terbedded ligh t- brown arkosic sandstone Lower Miocene Vaqueros Sandstone Yellowish-gray arkosic sandstone; marine Tbs. pillow basalt flows Red beds of Kerr and Schenck (19 25) Red breccia and conglomerate; in- terbedded red and yellow arko- sic sandstone; nonmarine Pinecate Formation of Kerr and Schenck (1925) Yellow arkosic sandstone; inter- bedded conglomerate; marine Zayante Sandstone Yellowish-orange arkosic sand- stone; interbedded green and red siltstone and conglomerate; nonmarine San Juan Bautista Formation of Kerr and Schenck (1925) Buff sandstone; interbedded gray siltstone; marine San Lorenzo Formation Light-gray nodular mudstone, lo- calu grading to arkosic sand- stone, and olive-gray shale; ma- rine Butano Sandstone Yellowish-gray arkosic sandstone; interbedded olive-gray siltstone; in terbedded conglomerate in lower part; marine UNCONFORMITY Locatelli Formation of Cummings, Touring, and Brabb (1962) Pale-yellowish-brown nodular siltstone, arkosic sandstone locally at base; marine NONCONFORM/TY Granitic rocks Predominantly quartz diorite and adamellite, ranging from gabbro to granite Metasedimentary rocks Schists, quartzites, marbles, and calc-silicate rocks. Include Gabilan Limestone in Gabilan Range NORTHEAST OF SAN ANDREAS FAULT 1: £0155." m , _ . . 5 Landslide material E o m Qal Alluvium Q) . ,7 S 8 Qt o E . g River-terrace deposit m : M a 8 - 'E San Benito Gravels of Lawson (1893) E Poorly consolidated sands and gravels; nonmarine UNCONFORMITY Q) R N D g Purisima Formation Porly consolidated sands, silts, clays, and gravels; marine and nonmarine UNCONFORMITY Mudstone Dar-gray, weathers light gray, siliceous organic mudstone; marine. Includes Monterey Shale ofAllen (I946) Sandstone Yellowish-gray arkosic sandstone; interbedded dark-gray mudstone; marine. Includes Monterey Sandstone of Allen (I 946) Eocene/.7), Oligocene, and Miocene Mudstone Olive-gray nodular mudstone; marine or Eocene Undifferentiated sedimentary rocks [nterbedded conglomerate, feldspathic graywacke, and shale; marine. Near Lama Prieta locally include extrusive(?) volcanic rocks Upper Cretaceous flflA/Afifi A AAr—R Ultrabasic intrusive rocks Predominantly serpentinite intrusive into Franciscan Formation and along fault zones (’1 3 - o N U E 9. N . . i3 Franctscan Formation a Lithic and feldspathic graywacke, siltstone, and altered mafic volcanic rocks; mi- g nor chert, limestone, and conglomerate b Diabase Altered diabase along San Andreas fault Contact, approximately located —'—# _?.__?. ....... - Fault Dashed where approximately located; dotted where concealed; queried where doubtful. U, relatively upthrown side; D, downthrown side; arrows show relative horizontal movement (—__+_. ......... Crest of anticline Showing direction of plunge. Dashed where inferred; dotted where concealed (__,j_____ Trough of syncline Showing direction of plunge. Dashed where inferred #5+6 Abandoned exploratory oil well Red indicates well reported to have reached crystalline basement 0 Gravity station +1o———- Isogals Lines of equal complete Bouguer anomaly in milligals; dashed in areas of poor control; contour interval 2 milligals. + indicates gravity high, 7 gravity low RH ; __.___. _/ W V WWWW—J OUATERNARY TERTIARY PALEOZOIC(?) CR ETACEOUS OUATERNARY TERTIARY OR QUATERNARY TERTIARY JURASSIC, CRETACEOUS AND CRETACEOUS, CRETACEOUS AND YOUNGER JURASS|C(?I AGE UNKNOWN Early Paleozoic Brachiopods of the Moose River Synclinorium, Maine GEOLOGICAL SURVEY PROFESSIONAL PAPER 784 Early Paleozoic Brachiopods of the Moose River Synclinorium, Maine By ARTHUR J. BOUCOT GEOLOGICAL SURVEY PROFESSIONAL PAPER 784 Descriptions are provided for large Early Devonian and Late Silurian brachiopod faunas of northwestern Maine, together with small Early Silurian and Middle Ordovician faunas 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. 73-600004 For sale by the Superintendent of Documents, US. Government Printing Office, Washington, DC. 20402 Price: $1.75. domestic postpaid; SLSO, GPO Bookstore Stock No. 2401-00325 Abstract Introduction _____________________________________ Age of the faunas _______________________________ Provinciality Systematic paleontology __________________________ Superfamily Orthacea ________________________ Genus Orthostrophia Hall, 1883 ____________ Genus Orthambonites Pander, 1830 ________ Genus Dolerorthis Schuchert and Cooper, 1931 Genus Ptychopleurella Schuchert and Cooper, 1931 Genus Valoourea Raymond, 1911 __________ Superfamily Dalmanellacea ___________________ Genus Resse'rella Bancroft, 1928 ___________ Genus Isorthis Kozlowski, 1929 ____________ Genus Levenea, Schuchert and Cooper, 1931 _ Genus Dicaelosia King, 1850 _______________ Genus Discomyorthis Johnson, 1970 ________ Genus Dalejina Havlicek, 1953 _____________ Genus Platyorthis Schuchert and Cooper, 1931 Superfamily Enteletacea ______________________ Genus Salopina Boucot, 1960 ______________ Superfamily Pentamemcea ____________________ Genus Gypidula Hall, 1867 ________________ Genus Siebe’rella Oehlert, 1887 _____________ Genus Pentamerus Sowerby, 1812 __________ Superfamily Plectambonitacea _________________ Genus Sowerbyites Teichert, 1937 __________ Superfamily Strophomenacea __________________ Genus Leptaena Dalman, 1828 _____________ Genus Stropheodonta Hall, 1852 ___________ Genus Leptostrophia Hall, 1892 ____________ Genus Protoleptostrophia Caster, 1939 ______ Genus Strophonella Hall, 1879 _____________ Superfamily Davidsoniacea ____________________ Genus Schuchertella Girty, 1904 ___________ Genus Hipparionyx Vanuxem, 1842 ________ Genus Leptaem’sca Beecher, 1890 ___________ Superfamily Chonetacea ______________________ Genus Cyrtom‘scus Boucot and Harper, 1968- Genus Eccentricosta Berdan, 1963 __________ Genus Dawsonelloides Boucot and Harper, 1968 Genus Eodevonam'a Breger, 1906 ___________ Genus Chonostrophiella Boucot and Amsden, 1964 __________________________________ Superfamily Rhynchonellacea __________________ Genus Cupularostrum Sartenaer, 1961 ______ CONTENTS Page U‘lhhfiCoI-‘HH O1 :0wa 10 11 11 12 16 16 18 18 19 19 20 20 20 20 20 20 21 21 22 23 24 24 24 25 25 25 25 26 27 28 29 29 Systematic paleontology—Continued Genus Ancillotoechia Havlicek, 1959 ________ Genus Sphaem'rhynchia Cooper and Muir- Wood, 1951 ____________________________ Genus Sulcatina Schmidt, 1964 ____________ Genus Costelli’rostra Cooper, 1942 __________ Genus Eatonia Hall, 1857 _________________ Genus Machaemm'a, Cooper, 1955 ___________ Superfamily Atrypacea _______________________ Genus Atrypa Dalman, 1828 _______________ Genus Lissatrypa Twenhofel, 1914 _________ Genus Nanospi’ra Amsden, 1951 ____________ Superfamily Dayiacea ________________________ Genus Leptoceolia Hall, 1859 ______________ Genus Coelospira Hall, 1863 _______________ Superfamily Delthyridacea ____________________ Genus Hedeina, Boucot, 1957 ______________ Genus Delthym‘s Dalman, 1828 _____________ Genus Howellella Kozlowski, 1946 __________ Genus Acrospiri'fer Helmbrecht and Wede- kind, 1923 _____________________________ Genus Antisgm’rifer Williams and Breger, 1916 Genus Costellispirifer Boucot, n. gen ________ Genus Mucrospirifer Grabau, 1931 _________ Genus Costispirifer Cooper, 1942 ___________ Genus Kozlowskiellina Boucot, 1958 ________ Genus Metaplasia Hall and Clarke, 1893 _~__ Genus Plicoplasia Boucot, 1959 _____________ Superfamily Cyrtinacea ______________________ Genus Cy'rtina Davidson, 1858 _____________ Superfamily Athyridacea _____________________ Genus Men’sta Suess, 1851 ________________ Genus Men'stella Hall, 1860 _______________ Genus Charionoides Boucot, Johnson, and Staten, 1964 ___________________________ Genus Nucleospira Hall, 1859 ______________ Genus Protathym's Kozlowski, 1929 ________ Superfamily Terebratulacea ___________________ Genus Nanothyris Cloud, 1942 _____________ Genus Rensselaem‘a Hall, 1859 Genus Beachia Hall and Clarke, 1893 Genus Cloudothyris Boucot and Johnson, 1968 Genus Amphigem'a Hall, 1867 Genus Globithyn‘s Cloud, 1942 _____________ Genus Mutationella Kozlowski, 1929 References cited __________________________________ Index Page 30 31 33 34 34 35 36 36 37 37 38 38 38 39 39 39 39 41 50 51 61 62 62 63 63 63 63 63 64 64 65 65 65 65 66 69 69 70 72 73 76 79 III IV CONTENTS ILLUSTRATIONS [Plates follow index] PLATE 1. Orthostrophia, Schizoramma, 0rthambonites?, and Dolerorthis. 2. Dolerorthis, Dolerorthis?, Ptychopleurella, Valcourea, Resserella, and Iso'rthis. 3. Isorthis, Levenea, unidentified orthoid, Dicaelosia, and Discomyorthis. 4. Discomyorthis, Dalejina, and Platyorthis. 5. Salopina and Gypidula. 6. Gypidula, Sieberella, Pentamerusl Sowerbyites?, Leptaena, and Stropheodonta. 7. Leptost'rophia, Leptostrophia?, and Protoleptostrophia. 8. Protoleptostrophia and Strophonella. 9. “Schuchertella,” Hipparionyx, Leptaem’sca, and Cyrtoniscus. 10. Eccentricosta and Dawsonelloides. 11. Dawsonelloides, Eodevonaria, and Chonostrophiella. 12. Chonostrophiella, Cupularostrum, Cupularostrum?, Ancillotoechia, “Ancillotoechia,” and Sphaer’irhynchia. 13. Sphaerirhynchia, Sphaem'rhynchia?, and Sulcatina. 14. Costellirostra, Eatom‘a, Maahaeram'a, and Atrypa. 15. Aim/pa, Lissatrypa, Nanospi'ra?, Leptocoelia, Coelospira, and Hedeina. 16. Delthym‘s, Howellella?, “Howellella,” and Acrospim'fe'r. 17. Acrospim'fer. 18. Acrospirifer, Antispirifer, Costellispi'rifer, “Muc’rospirifer,” and Costispirifer. 19. Megakozlowskiella, Metaplasia, Plicoplasia, and Cyrti‘nal 20. Men'sta, Meristella, Charionoides, Nucleospi’ra, and Protathy’r'is. 21. Nanothyris, Rensselaeria, and Beach/id. 22. Beachia, Cloudothym's, and Amphigem'a. 23. Globithyris and Mutationella. Page FIGURES 1—10. Scattergrams for measurements of: 1. Dolerorthis hobbstownensis ___________________________________________________________ 1 2. Ptychopleurella sp ____________________________________________________________________ 8 3. Valcom'ea sp ________________________________________________________________________ 9 4. Discomyorthis musculosa solaris _______________________________________________________ 14 5. Platyorthis plamocon'uexa ______________________________________________________________ 17 6. Salopina hitchcocki ________________‘ ___________________________________________________ 18 7. Cyrtom'scus nectus ___________________________________________________________________ 26 8. Dawsonelloides comadensis ____________________________________________________________ 27 9. Eodevonam’a arcuata _________________________________________________________________ 28 10. Chonostrophiella complanata __________________________________________________________ 29 11. Serial sections of Ancillotoechia haragtmensis from the Haragan Shale of Oklahoma _____________ 30 12. Serial sections of Diabolirhynchia acinus from the Waldron Shale of Indiana ___________________ 31 13. Serial sections of “Rhynchonella” bidentata from Silurian rocks near Klinteham, Gotland, Sweden -. 32 14. Serial sections of “Ancillotoechia” cf. “A.” altisulcata __________________________________________ 32 15—19. Scatterg'rams for measurements of: 15. “Howellella” tomhegomensis ___________________________________________________________ 42 16. Acrospirifer mu'rchisoni ______________________________________________________________ 44 17. Acrospirifer atlant'icus _______________________________________________________________ 48 18. Antispim'fer harroldi _________________________________________________________________ 52 19. Costellispiriferperimele _______________________________________________________________ 60 20. Serial sections of Protathyris sp ______________________________________________________________ 66 21—25. Scattergrams for measurements of: 21. Nanothym‘s hodgei ___________________________________________________________________ 68 22. Amphigem'a par'va ___________________________________________________________________ 71 23. Globithym‘s callida ____________________________________________________________________ 72 24. Globithy’ris diam'a ____________________________________________________________________ 74 25. Mutationella parlinensis ______________________________________________________________ 75 TABLE Page TABLE 1. Rib branching pattern for brachial valves of Dolerorthis hobbstownensis Boucot, n. sp ___________________ 6 EARLY PALEOZOIC BRACHIOPODS OF THE MOOSE RIVER SYNCLINORIUM, MAINE By ARTHUR J. BOUCOT ABSTRACT Brachiopods of the Moose River synclinorium of Maine are described from beds of Middle Ordovician, Early Silurian, Late Silurian, and Early Devonian age, with emphasis on the Early Devonian brachiopods because of their relative abundance and because of their distribution in several zones and several marine communities. New genus Costellispirifer, a multiribbed delthyrid, is defined. New species are Dolerorthis hobbstowensis, Siebe'rella beckensis, Cupularostrum macrocosta, M achaeraria mainensis, “Howellella” tomhegtmensis, and Nanothym's hodgei. INTRODUCTION The stratigraphic and structural geology of the Moose River synclinorium has been dealt with earlier by Boucot (1961, 1969). Oliver (1960) de- scribed the Devonian corals and Stumm (1962) the Silurian corals; Boucot and Yochelson (1966) de- scribed the gastropods; and Whittington and Camp- bell (1967) described some of the Silurian trilobites. Earlier work on the faunas, chiefly brachiopods, in- cludes the studies of Clarke (1907, 1909), Schuchert (in Pirsson and Schuchert, 1914), Williams and Breger (1916, in their report on the Chapman Sand- stone fauna), and Go-ldring (1933). Miscellaneous reports by Boucot and others have dealt with par- ticular parts of the brachiopod fauna in connection with various biologic studies: (Boucot, 1959a, deals with species of Metaplasia and Plicoplasia; Boucot and‘others, 1963, deal with species of Mutationella; Boucot and Amsden, 1964, deal with species of Chonostrophiella; Boucot and Johnson, 1967a, deal with occurrences of Coelospim; and Boucot and Harper, 1968, deal with occurrences of Dawson- elloz‘des, Cyrtom'scus, and Eodevonaria). Walmsley, Boucot, and Harper (1969) deal with Salopina from the Moose River synclinorium. The original purpose of the paleontologic and geo- logic work in the Moose River synclinorium was to determine if the undifferentiated Early Devonian fauna as described by Clarke (1907, 1909) and Wil- liams and Breger (1916) contained more than one horizon. Based on his own thesis studies (Cloud, 1942) and an extensive field trip in the area in 1943, Preston E. Cloud, Jr. had suspected that more than one horizon was present and suggested the problem to Boucot in 1948 as a promising subject for a dis- sertation. The purpose of this report is to document the brachiopod fauna. Extensive work on Appa- lachian province Early Devonian brachiopod faunas since the completion of the Moose River studies in early 1956 has made possible effective comparison of them with similar faunas occurring elsewhere in the eastern half of North America. The locality numbers used in this report are from the US. Geological Survey’s locality registers. Those with the suffix “SD” are from the Silurian-Devonian catalog, those with “CO” from the Cambrian-Ordo- vician catalog. Details of USGS locs. 2690—SD to 4017—SD are given in Boucot (1969, appendix 1, p. 105-117) and of USGS locs. 4841—SD, 4843-SD, 5583—SD, 5586—SD, 5587—SD, and 5995—SD are given in Albee and Boudette (1972). US National Museum brachiopo-d catalog numbers are assigned to all the illustrated, figured, and measured specimens. Boucot and Heath (Boucot, 1969) summarize the geology of the Moose River and Roach River syncli- noria, including the structure, stratigraphy, fossil localities, and animal communities. Plate 30 of that report is a generalized geologic map on which the fossil localities are shown. Stratigraphic nomencla- ture corresponds with that used in the descriptive part of this report. AGE OF THE FAUNAS Early Paleozoic faunas of Middle Ordovician, Early Silurian, Late Silurian, and Early Devonian age are associated within the Moose River syncli- norium. The Middle Ordovician fauna, which in- cludes the brachiopods Orthambonites? sp., Val- coureu sp., and Sowerbyites? sp., was previously concluded to be of early to middle Middle Ordovician age as evidenced by the brachiopods. Early Silurian fossils are not known in the Moose River synclinorium proper, but have been recorded at Limestone Hill in the Stratton quadrangle, which lies on the southeastern side of the synclinorium (Boucot, 1969) . On Limestone Hill the occurrence of Pentamerus? sp., Dicaelosz'a sp., and a species of Amphistrophia similar to A. funiculata suggests a late late Llandovery or possibly very early Wenlock age, from the occurrence of Dicaelosz’a (late Llan- dbvery-Wenlock type) as well as Amphistrophia 1 l 2 E RLY PALEOZOIC BRACHIOPODS OF THE MOOSE RIVER SYNCLINORIUM, MAINE similar to A. funiculata. The occurrence of Atrypa “reticularis,” Howellella, and Eospirifer precludes the possibilit of these being of pre-late Llandovery age, as these genera are unknown elsewhere in the world earlier than the late Llandovery. The Late Silurian faunas are from four sources: Lobster Lake Formation, USGS loc. 3488—SD of the Hardwood Mountain Formation, the remainder of the Hardwood Mountain Formation, and lower con- glomerate member of the Hobbstown Formation. Brachiopods are very rare in the Lobster Lake Formation and were not studied in connection with this report. They include uncollected, poorly pre- served specimens of Kirkidium cf. K. knighti from the southwest shore of Lobster Lake, and a Gypidulz similar to the one present in the Hardwood Mountain Formation from the north-east shore (USNM loc. 12113). Stumm (1962, p. 2, 3, 4, 6) and Oliver (1962, p. 11) both consider that the corals of the Lobster Lake Formation are similar to those of the Ludlow and Pridoli age Hardwood Mountain and Mont Wis- sick Formations. The evidence from both the brachi- opods and corals is consistent with a Ludlow age, but a Pridoli age cannot be ruled out, at least in part. This conclusion also applies for at least parts of both the Hardwood Mountain and Mont Wissick Formations. In View of the probable Pridoli age of part of the Hardwood Mountain, a Pridoli age, at least in part, for any “Ludlow” age unit in northern Maine must be considered likely, unless compelling evidence to the contrary is available. USGS loc. 3488—SD has been assigned to the Hard— wood Mountain Formation, which represents a single exposure of calcareous siltstone in an area ad- j acent to extensive exposures of a lithologically simi- lar unit. However, the fauna of the locality is com- pletely difierent from that found elsewhere in the Hardwood Mountain. Inspection of table 8 in Boucot (1969) shows that among the 13 brachiopod taxa recognized at this locality only three have been recognized elsewhere in the formation, and these three are the ubiquitous forms, Atrypa “reticulam’s,” Lissatrypa, and Coelospim. The locality has a lower age limit of late Wenlock, as shown by the presence among the brachiopod-s of Coelospim and Sulcatina. The ostracodes and bryozoans (fenestellids are abun- dant) are relatively undiagnostic or are new. Whit- tington and Campbell (1967) conclude that the trilobite fauna from USGS 10c. 3488—SD may be of late Wenlock or early Ludlow age. Structurally the locality is inferred to be near the base of the Hard- wood Mountain Formation. Summing up all of these data, the author assigns at this time an inferred late Wenlock or Ludlow age for the locality. The Hardwood Mountain Formation as presently mapped (1969) consists of three discontinuous areas; The Spencer Mountain outlier and adjacent southwest prong of the Moose River synclinorium, the area to the north of Little Big Wood Pond, and a few exposures on the eastern flank of Sally Moun- tain. The exposures in both the Sally Mountain and Little Big Wood Pond areas have yielded Eccentri- costa, a genus restricted in the central Appalachians to beds of Pridoli age. Unfortunately Eccentricosta is virually unknown from an evolutionary point of View and is difiicult to use in determining the age of the containing beds in the northern Appalachians. The occurrence (Klapper in Boucot, 1969) of very latest Ludlow through Pridoli age conodonts at one locality on the north shore of Little Big Wood Pond makes it likely that the northern exposures of the formation which contain Eccentricosta are indeed of Pridoli age. Corals from the Little Big Wood Pond area were concluded by both Stumm (1962) and Oliver (1962), to be of “Ludlow” age, but this age determination is an earlier one made when the dis- tinction between strata of Ludlow and Pridoli age was not yet practicable within the northern Appa— lachians. J. M. Berdan, in Boucot (1969) considers that ostracodes from the Hardwood Mountain For- mation of the Little Big Wood Pond area can be best correlated with the Pridoli age Tonoloway Lime- stone. Considering all this evidence and the fact that the conodont collectiOn was made within about ten feet of the unit’s base, the author concludes that the Hardwood Mountain Formation of the Sally Moun- tain and Little Big Wood Pond area is largely, if not entirely, of Pridoli age. The Hardwood Mountain Formation of the Spen- cer Mountain outlier and of the adjacent synclinori- um to the southeast does not contain any brachio- pods of zonal value, but the ostracodes (see Berdan in Boucot, 1969) do indicate a correlation with the Pridoli age Tonoloway Limestone. The overlying fauna recovered from the basal few inches of the Hobbstown Formation is certainly Silurian, as shown by the presence in the Appalachians of such brachiopod genera as Dolero'rthis, Ptychopleurella, Resse’rella, and Delthym's (Delthym's). The ostra- code Limbinam’a? cf. L.? mm‘icata (Berdan in Bou— cot, 1969) suggests a Tonoloway, that is, Pridoli cor- relation. The available evidence suggests that the Hardwood Mountain Formation of the Spencer Mountain outlier and adjacent synclinorium is at least partly, if not almost entirely of Pridoli age, but a Ludlow or even a late Wenlock age for a part of the formation seems reasonable at this time. The fauna obtained from the basal few inches of PROVINCIALITY 3 the Hobbstown Formation’s lower conglomerate member is considered to be of probable Pridoli age. Other considerations (Boucot, 1969) make it likely that this fauna from the basal few inches was re- worked from Silurian sediment and redeposited in an essentially Lower Devonian rock unit. The Lower Devonian faunas of the Moose River synclinorium may be divided into three age units: a lower unit of middle or late Helderberg age (see Bou- cot and Johnson, 1967b) , a middle unit of Oriskany age, and an upper unit of Schoharie age. Strata of Esopus age that contain recognizable marine fossils have not been found in this region. Strata of middle or late Helderberg age containing datable fossils include the basal Seboomook Forma- tion near Beck Pond, the Bear Pond Limestone Mem- ber of the Seboomook Formation, the Beck Pond Limestone, and the Parker Bog Formation. Boucot and Johnson (1967b) have summarized the argu- ments for the middle or late Helderberg age of the Beck Pond Limestone and the Bear Pond Limestone Member of the Seboomook. The scanty fauna from the Parker Bog Formation is consistent with a middle or late Helderberg age but cannot be used to prove it. Similarly scanty are the basal Seboomook fossils of Helderberg age in the Beck Pond area (these shells occur in a stratigraphic position below the Bear Pond Limestone Member), although the occurrence of Spimplasia would appear to rule out an early Helderberg age. Strata of Oriskany age within the synclinorium include the Tarratine Formation and most of the Seboomook Formation. The presence of Rensselaeria, Hippam'onyx, Beachia, Costispirifer, and Dawson- elloides is consistent with an Oriskany age for the Tarratine Formation. The intertonguing relation-s of the Seboomook and the Tarratine (Boucot, 1969) and the similarity of the fauna indicate that the greater part of the Seboomook is of Oriskany age in this region. The Schoharie age beds in the Moose River syncli- norium include the Tomhegan Formation and the Kineo Rhyolite. Only the Tomhegan has yielded diag- nostic fossils. The presence of small Amphigenia, Eodevonuriu arcuatw, Acrosm'm'fer atlcmtz'cus, and Cham'onoides dom's indicate that the Tomhegan fauna is of Schoharie age. PROVINCIALITY The Middle Ordovician Kennebec Formation has yielded a very small fauna (Boucot, 1969), but all the elements present suggest an association with Central and Southern Appalachian rather than with northern European forms. Other shelly «faunas from northern Maine of Ordovician age (Neuman, 1968) differ sharply from that of the Kennebec Formation. This evidence indicates that Maine was a boundary region during the Ordovician and included Euro- pean-type faunas in both Aroostook and Penobscot Counties to the northeast and Appalachian type-s in Somerset County to the southwest. The evidence ob- tained by Lespérance (1968) for the Late Ordovician trilobites of Gaspé is similar in that it indicates that Gaspé lay near a boundary between Appalachian- and European-type regions for at least part of the Ordovician. The Silurian (see Boucot and others, 1969) was a time of great cosmopolitanism for the brachiopod faunas. Toward the end of the Silurian this situa- tion began to deteriorate into a more provincial con- dition. In Somerset County, the presence of the endemic Appalachian chonetid genus Eccentricosta is consistent with this conclusion, as the containing beds are probably of Pridoli, that is, latest Silurian age. The ostracode fauna of the central Appalachians differs largely from that of northern Europe, and the occurrence of central Appalachian-type ostracodes in the probably Pridoli-age strata of Somerset County is in great contrast to the Baltic-type fauna of coast- al Maine (Berdan, p. 39, 40 and Martinsson, p. 41- 43, in Berry and Boucot, 1970). These data indicate that during Pridoli time there was a boundary be- tween an Old World province-type fauna in coastal Maine and adjacent New Brunswick, and the Appa- lachian province precursors in northern Main-e, ad- jacent Quebec and through Gaspé. It is emphasized, however, that during Pridoli time most of the brachi- opods of northern Maine, adjacent Quebec, and Gaspé belonged to relatively cosmopolitan genera. The scanty pre-Pridoli age Silurian fauna in this region from Limestone Hill is concluded to be of late late Llandovery age and is a typical Silurian cosmopolitan shelly fauna. The Helderberg faunas of the Moose River syncli- norium contain a variety of brachiopods, all of which have Appalachian province affinities or belong to widespread taxa. The same situation holds for faunas of Oriskany and Schoharie age. The only anomaly is the abundance of Mutationella in the Oriskany age faunas, because this genus, which also occurs in the Rhenish community of the Old World province, can be interpreted to be an Old World genus, but more probably it is a member of a rela- tively near-shore community (the “Mutationella. community” of Boucot and Johnson, 1967b) which also occurs in the Old World province. The same problem is posed by the abundant representatives of Globithym’s which occur in the widespread globithy- 4 EARLY PALEOZOIC BRACHIOPODS OF THE MOOSE RIVER SYNCLINORIUM, MAINE rid community of both Oriskany and Schoharie age in the Moose River synclinorium. The earliest repre- sentatives in this region of an undoubted Old World province brachiopod fauna occur about fifty miles to the northwest at Saint George, Quebec, Where several genera of Old World province afl‘inity occur in the Onondaga (that is, Eifelian) age Famine Limestone. Oliver’s (1967) analysis of the coral fauna in eastern North America during the Helderberg through Onondaga interval is similar to the brachio- pod analysis presented here. SYSTEMATIC PALEONTOLOGY PHYLUM BRACHIOPODA Class ARTICULATA Suborder ORTHOIDEA Superfunily ORTHACEA Family ORTHIDAE Subfamily ORTHOSTROPHIINAE Genus ORTHOSTROPHIA Hall, 1883 As pointed out by Williams (1951, p. 91), Ortho- strophia is predominantly orthid in its general form. Williams emphasized the orthid nature of the cardi- nalia associated with plectorthid-type pallial marks. Williams (1951, p. 90), after finding a dolerorthid (his Schizoramma cf. subplicata) with plectorthid- type pallial marks, concluded that pallial marks were of less value in making family assignments than car- dinalia. However, Williams, in reassigning Ortho- strophia to the Hesperorthinae, ignored the external and internal differences between Orthostrophia and the genera of the Hesperorthinae which constitute valid grounds for leaving the former in a distinct subfamily, albeit of the Orthidae rather than the Plectorthidae. More recently Williams (1965) classi- fied Orthostrophia with the Orthinae. Probably the most prominent feature of Ortho- strophia vis-a-vis the hesperorthinids is the presence in the former of a reversal of fold and sulcus from the presence in the umbonal region of a pedicle valve fold to a prominent sulcus anteriorly and corres- ponding features in the brachial valve. Orthostro- phia has accessory ridges parallelling the linear cardinal process in a manner reminiscent of true Schizoramma (pl. 1, figs. 13, 14). Orthoslrophia cf. 0. brownsporlensis Amsden, 1949 Plate 1, figures 1—5 Exterior—Shells transverse in outline. Straight hinge line position of maximum width. Lateral mar- gins relatively straight, normal to hinge line posteri- orly, round evenly into anterior margin. Anterior margin evenly rounded, crenulate, unisulcate. Pedi- cle valve bears low fold in umbonal region which becomes transformed into prominent sulcus anteri- orly; brachial valve bears low sulcus in umbonal region which becomes transformed into fold an- teriorly. Radial costellae increase by both bifurca- tion and implantation. Concentric filae, growth lines. Pedicle valve in‘terarea strongly apsacline, about twice length anacline brachial valve interarea. Both valves evenly convex. Pedicle valve about twice as deep as brachial valve. Pedicle valve interior—Dental lamellae very short, almost completely obsolescent. Hinge teeth stout. Muscle field very short, divided into median sector bounded laterally by two lateral sectors, all concentrically striated. Radial ornamentation im- press prominent except in delthyrial cavity. At bot- tom of delthyrial cavity is low subcircular ridge which may have bounded adjustor attachment area. Brachial valve interior.——Cardinalia consist of stubby brachiophores bounding edge of notothyrial cavity, linear cardinal process, pair of thin ridges paralleling closely cardinal process margins. Low myophragm extends anteriorly from cardinal proc- ess margins. Low myophragm extends anteriorly from cardinal process; bisects very deeply impressed adductor field. Adductor field quadripartite. Comparison—O. cf. 0. brownsportensis is more finely costellate than 0. strophomenoides and bears a narrower fold and sulcus anteriorly. Occurrence.—USGS loc. 4843—SD, Attean quad- rangle, Somerset County, Maine. Stratigraphic location—Hardwood Mountain For- mation (Upper Silurian). Figured specimens.—USNM 160109—160111. Orthoshophia cf. 0. slrophomenoides (Hall, 1857) Plate 1, figures 6—12 Exterior—Shell gently biconvex, with pedicle valve possessing greater convexity. Outline subcir- cular, greatest width near midlength. Brachial valve sulcate posteriorly, bears fold anteriorly; pedicle valve lacks sulcus posteriorly, develops one anterior- ly in large shells. Hinge line straight. Pedicle valve interarea about three times as long as that of brachi- al valve. Pedicle valve interarea. apsacline, that of brachial valve anacline. Both valves multicostellate; costellae appear to originate largely by bifurcation. Anterior co-mmissure uniplicate, crenulate. Pedicle valve interior—Interior prepared impres- sion does not show interior details very satisfactori- ly. Short dental lamellae present; laterally bound muscle attachment area. Latter restricted to del- thyrial cavity, appears tripartite, with median por- tion slightly more depressed than tw0 lateral portions. Medial impression expands anteriorly, probably adductors attachment site. Muscle field SYSTEMATIC PALEONTOLOGY 5 crossed by striations subparallel to hinge line. No trace of pallial marks found on specimen studied. External costellae impression marked, particular- ly in antero-median part of shell; this feature may be partly associated with deformation of the specimen. Brachial valve interior—Anterior portion dental sockets formed by stubby brachiophores, posterior portion by interarea edge. Linear cardinal process grades anteriorly into short myophragm which has triangular cross section. Subcircular adductor at- tachment area divided medially by myophragm, bounded laterally by low ridges, divided transverse- ly by low ridge normal to myophragm. Pallial mark- ings not identified on available material. Impression of costellae on interior marked in anterior and lateral portions of shell. Valves large, several are 35 to 40 mm wide. Comparison—Material from the Beck Pond Lime- stone most closely resembles O. strophomenoidcs from the New Scotland Formation of eastern New York, but poor preservation and inadequate number of specimens preclude positive specific identification. Occurrence.—USGS loc. 3499—SD, Spencer Lake quadrangle, Somerset County, Maine. Stratigraphic location—Beck Pond Limestone (Lower Devonian). Distribution—Species is known only from eastern North America and Oklahoma. Figured specimens.—USNM 125785B, 125786A— B. Unfigured specimens.—USNM 125787A, 125788B, 125789C, 125790D, 125791E, 125792A. Subfnmily ORTHINAE Genus ORTHAMBONITES Pander, 1830 Orthambonites? sp. Plate 1, figures 15—17 Exterior.——Unequally biconvex shell, brachial valve gently convex, pedicle valve subpyramidal. Greatest width at straight hinge line. Pedicle valve interarea steeply apsacline, brachial valve interarea appears anacline. Pedicle valve interarea much longer than that of brachial valve. Lateral margins gently rounded. Anterior margin rectimarginate, crenulate. Radiating costae; no secondary costae evident on specimen studied. Comparison—The single impression of the ex- terior available is que-stionably assigned to Ortham- bonites on the basis of convexity of the brachial valve which is unlike the flat brachial valve of Hes- perorthis. Pedicle valve interarea is too poorly pre- served for the determination of the presence or ab- sence of a pseudod-eltidium. Occurrence.—USGS 10c. 4317—00, Brassua Lake quadrangle, Somerset County, Maine. Stratigraphic location.—Kennebec Formation (middle Middle Ordovician). Figured specimen.—-USNM 125753. Family HES PERORTHIDAE. Subfamily DOLERORTHINAE Genus DOLERORTHIS Schuchert and Cooper, 1931 Dolerortliis hobbstownensis Boucot, n. 5]). Plate 1, figures 18—26 Exterior—Shells unequally biconvex, pedicle valve subconical, brachial valve gently convex; ellip- tical to subcircular in outline. Brachial valve sulcus more prominent in early than in late growth stages. Hinge line straight, lateral margins rounded, great- est width occurring abOut midlength. Pedicle valve interarea steeply apsacline, in crushed specimens it appears gently apsacline; brachial valve interarea gently anacline. Pedicle valve interarea about three times as long as that of brachial valve. Sides of del- thyrium subparallel but in crushed specimens they commonly include angle of about 45°. Number of pri- mary plications (table 1) originating near beak usually six or seven each side of midline. Two secon- dary plications occur in brachial valve sulcus; simi— larly on either side of pedicle valve medial primary plication. Secondary plications bifurcate a few milli- meters anterior to umbo. Secondary plications as- sociated with other primary plications originate slightly anterior to those in sulcus or associated with median primary of pedicle valve. Those originating on brachial valve do so on medial side of primary plication, whereas reverse true of pedicle valve. Ter- tiary plications uncommon except in third primary which gives rise to tertiary plication in about 50 percent of specimens examined. Interspaces crossed by prominent filae. Anterior commissure of small shells sulcate, but in larger specimens becomes al- most rectimarginate, crenulate. Pedicle valve interior—Short dental lamellae di— verge laterally from anterior face of palintrope and bound muscle attachment area. Hinge teeth stubby and occur medially on hinge line; they are supported in part by dental lamellae. Muscle field located entire- ly within delthyrial cavity, lies on pad of secondary material which floors this part of valve and raises it slightly above level of rest of valve. Anterior margin crenulated, both anterior and lateral portions of shell bear impression of external coarse costellae. Brachial valve interior.—Cardinalia consist of linear, blade-like cardinal process resting on noto- thyrial platform bounded by short, stout brachio— phores whose posterior faces are in same plane as interarea. Dental sockets floored by fulcral plates, EARLY PALEOZOIC BRACHIOPODS OF THE MOOSE RIVER SYNCLINORIUM, MAINE TABLE 1.—Rz’b branching pattern for brachial valves of Dolerorthis hobbstownensis Bouaot, n. sp [From collection of specimens numbered 1A55, from base of Hobbstown Formation, USGS loc. 3479~SD, Spencer Lake quadrangle, Somerset County. Maine. 1. 2, 3 . . . primary costella; 12 median secondary costella; la tertiary to median secondary costella; 23, 33 . . . respective primary. L and R refer to left and right side of valve; 1 and 2 refer to one or more costellae of that order in that pasition] lateral secondary costella to Specl- 1 12 13 2 23 3 33 4 4a 5 53 6 menLR LR LR LR LR LR LR LR LR LRLR LR LR LR 1 _ 1 1 1 1 __ __ 1 1 1 1 1 1 __ 1 1 1 1 __ 1 1 __ __ 1 1 1 __ __ __ 5 _ 1 1 1 1 __ _- 1 1 1 1 1 1 2 1 1 1 __ __ 1 1 __ __ 1 1 __ __ __ __ 7 _ 1 1 1 1 __ __ 1 1 2 1 1 1 2 1 1 __ 1 __ 1 __ __ __ __ __ __ __ __ -_ 9 _ 1 1 1 1 1 __ 1 1 1 1 1 1 2 2 1 1 __ 1 1 1 __ __ __ 1 __ 1 __ -- 10 _ 1 1 1 1 __ -_ 1 1 1 1 __ 1 __ 1 __ _- __ -_ -- _- __ __ __ __ __ __ __ __ 11 _ 1 1 1 1 __ 1 1 1 1 __ 1 1 1 __ 1 1 __ __ 1 1 1 __ __ 1 __ __ __ __ 22 _ 1 1 1 1 __ __ 1 1 __ __ 1 1 1 2 1 1 __ __ 1 1 __ __ 1 1 __ 1 __ __ 24 _ 1 1 1 1 __ __ __ 1 __ 1 __ 1 __ 2 __ 1 __ __ __ 1 __ __ __ 1 __ __ __ __ 25 _ 1 1 1 1 _- __ 1 1 1 __ 1 _- 2 __ l __ __ __ 1 __ __ __ __ __ -_ __ __ __ 27 - 1 1 1 1 _- -- 1 1 1 1 1 1 __ 1 1 __ 1 __ __ 1 __ __ __ __ __ __ __ __ 28 _ 1 1 1 1 __ __ 1 1 1 1 1 1 1 1 1 1 1 1 1 1 __ __ 1 1 __ __ __ __ 29 _ 1 1 __ __ __ __ 1 1 __ __ 1 1 _- -- 1 1 __ __ __ __ __ __ __ __ __ __ __ __ 30 _ 1 1 1 1 -_ 1 1 1 1 1 1 1 2 2 1 1 1 1 1 1 _- __ 1 1 __ -_ __ -_ 31 _ 1 1 1 1 __ __ 1 1 1 1 1 1 2 1 1 1 1 1 1 1 _- __ 1 1 1 __ 1 __ 39 - 1 1 1 1 __ __ 1 1 __ __ 1 1 __ -_ 1 1 __ __ 1 1 __ __ __ 1 __ __ __ __ 42 _ 1 1 1 1 __ __ 1 1 1 1 1 1 1 -- 1 1 __ 1 1 __ __ __ 1 __ __ -_ __ __ 47 _ 1 1 1 1 __ __ 1 1 1 1 1 1 1 __ 1 1 1 __ __ 1 __ __ __ 1 __ __ __ __ 49 _ 1 1 1 1 __ _- 1 1 2 1 1 1 2 2 1 1 __ __ 1 1 __ __ 1 1 __ __ __ __ 50 _ 1 1 1 1 1 __ 1 1 __ 1 1 1 1 2 1 1 __ 1 1 1 __ __ 1 1 __ __ __ __ 51 _ 1 1 1 1 __ 1 1 1 1 2 1 1 2 2 1 1 -_ __ 1 __ __ __ __ __ __ __ __ __ 54 _ 1 1 1 1 __ __ __ 1 __ 1 -_ 1 __ 1 __ 1 __ __ __ 1 __ a- __ 1 __ __ __ __ 55 _ __ __ __ -_ __ __ 1 __ __ __ 1 __ 2 __ 1 __ 2 __ 1 __ __ _- __ __ __ __ __ _- laterally formed by outer face of brachiophor‘es and 20 by inner face of interarea. Low rounded median sep- _ tum extends from anterior face of notothyrial plat- form to about midlength. Adductor impressions _. 1; .' poorly impressed, but noticeable posteriorly, fading .-', anteriorly. Anterior margin crenulated, and anterior ' ‘ '3 ' and lateral portions of shell impressed by external 10 ,. costellae. _ ' Measurements—Scatter diagrams of length ver- . ~' sus Width for both pedicle and brachial valves are E Brachia. “We essentially linear (fig. 1). The plot for pedicle valves ui-J ' exhibits greater degree of dispersion than that for E brachial valves. This greater dispersion is probably :4 0 due to the pedicle valve having been a far more con- 3 vex valve which upon crushing (most of specimens __ 20 are crushed) has had its linear dimensions subjected E ' to a greater amount of alteration than the relatively E ' ‘ . flat brachial valve. The slope of both plots is about -| -' ; n . . . . . . - '.' a. 55° It IS unlikely that there IS a radlcal change 1n :-'-. slope of the growth curve from small specimens . . (smaller than available for study) to large specimens 10 , because the extrapolated growth line would approxi- . ' mately pass through the origin. A greater degree of -°: dispersion accompanies increase in overall size, but Pediclé valve this does not indicate a greater degree of variation but rather about the same percentage variation. Both samples are strongly skewed towards larger sizes 00 1o 20 30 which suggests that abstraction of smaller specimens from sample occurred, although it is unclear whether this occurred during transportation or in situ. Comparison—D. hobbstown ensis is more coarsely costellate than the previously described species of Spencer Lake quadrangle, Somerset County, Maine. WIDTH, IN MILLIMETERS FIGURE 1.—Comparison of length versus width for brachial and pedicle valves of Dolerorthis hobbstownensis from the base of the Hobbstown Formation. Locality 3479—SD, SYSTEMATIC PALEONTOLOGY 7 Doler orthis sensu stricto. The rib-branching pattern of D. hobbstowrceusis is less complicated than that. of the previously described species. Occurrence.-—USGS locs. 2712—SD, 3479—SD, Spencer Lake quadrangle, Somerset County, Maine. Stratigraphic location—Base of Hobbs-town For- mation (Upper Silurian). Possibly Late Silurian shells reworked in basal part of a Devonian unit (Boucot, 1969, p. 39-40). Figured specimens.-—USNM 125796A, B (holo- typ'e), 125797B, 125903A. Measured paratypes.—USNM 125799. Unmeasured specimens.—USNM 126272, 126273. Dolerorlhis cf. D. llami (Amsden, 1951) Plate 2, figures 1—5 Exterior—Shell unequally biconvex, pedicle valve subconical, brachial valve gently convex. Brachial valve sulcate. Shell outline su‘bcircular to elliptical. Anterior commissure sulcate, crenulate. Straight hinge line, greatest width near midlength. Pedicle valve interarea steeply apsacline, brachial valve in- terarea anacline. Costellae which increase by bifur- cation. Brachial valve sulcus contains two pairs secondary costellae, one pair tertiary costellae. Fine filar ornamentation concentrically crosses inter- spaces, absent from crests of costellae. Pedicle valve interior—Short dental plates later- ally bound low pad secondary material in delthyrial cavity on which are situated impressions of muscle field. Hinge teeth nature not clear on one specimen available for study. External costellae impressions reflected on anterior and lateral portions of interior. Brachial valve interior.—Cardinalia consist of linear, bladelike cardinal process situated on noto- thyrial platform lateral to which are located rela- tively short anteriorly directed brachiophores. Mus- cle field nature diflicult to ascertain on one specimen available but paired adductor impressions near an- terior face noto‘thyrial platform are medially di- vided by low myOphragm. External costellae impres- sion reflected on anterior and lateral portions of interior. Comparisou.—The material from Maine is inade- quate for definitely determining specific position; most closely resembles D. hami. Occurrence.—-USGS 10c. 3479—SD, Spencer Lake quadrangle; loc. 5995—SD, Attean quadrangle, Som- erset County, Maine. Stratigraphic locatiou.——Base of Hobbstown For- mation and Hardwood Mountain Formation (Upper Silurian). Distribution.—Central Oklahoma; Maine. northern Figured specimens.—USNM 125794A-B, 125795A—B, 160112. Dolerorthis? 3]). Plate 2, figures 6—8 Exterior—Gently convex pedicle valve (no brach- ial valve available) has subcircular outline. Hinge line straight, greatest width at hinge line. Pedicle valve sulcate. Lateral extremities rounded towards anterior. Anterior commissure gently uniplicate, crenulate. Numerous bifurcating costellae crossed by filae restricted to interspaces. Pedicle valve interior—Hinge teeth located on medial margin of hinge line, appear short. Dental plates Short, laterally bound muscle attachment area. Muscle field located on pad secondary material that raises delthyrial cavity floor slightly above rest of valve. Muscle field nature not elucidated from frag- mentary specimen available. External costellae im- pression leaves imprint on anterior, lateral portions of shell. Comparison—The two fragmentary specimens available are insufficient to assign material generical- ly. Relatively fine costellae, nature of pedicle valve muscle field, presence of filae in interspaces between costellae are suggestive of Dolerorthis. Most closely resembles forms like D. rustica osilierzsis (Schrenk, 1858), unlike pauciplicate forms assigned to genus. Occurrence.—USGS 10c. 3479-SD, Spencer Lake quadrangle, Somerset County, Maine. Stratigraphic location—Base of Hobbstown For- mation (Upper Silurian). Figured specimens.——USNM 1258OOB—C. Unfigured specimens.—USNM 125800A. Subfamily GLYPTORTHINAE Genus PTYCHOPLEURELLA Schuchert and Cooper, 1931 Ptychopleurella sp. Plate 2, figures 9-17 Exterior.—Unequally biconvex, pedicle valve sub- conical, brachial valve convex. Brachial valve bears deep median sulcus. Hinge line straight, maximum width slightly anterior of midlength. Shell outline transversely elliptical, width about one and one-half times length. Pedicle valve interarea apsacline, brachial valve interarea orthocline. Anterior com- missure rectimarginate, crenulate. Delthyrium sides subparallel, apparently unmodified by delthyrial structures. Primary costellae bifurcate anteriorly giving rise to secondary and tertiary costellae. Small specimens (less than a few millimeters long) do not possess secondary or tertiary costellae). Three to four primary plications present either side midline. C'oarse, radial ornamentation crossed by fine, con- centric lamellae. 8 EARLY PALEOZOIC BRACHIOPODS OF THE MOOSE RIVER SYNCLINORIUM, MAINE Pedicle valve interior—Short dental plates lateral- ly bound delthyrial cavity, intersect palintrope an- terior face lateral to delthyrial margin. Hinge teeth short, located medial part hinge line. Secondary ma- terial pad laid down in delthyrial cavity slightly an- teriorly, greatest thickness material anterior simu- lating spondylium. Pad secondary material narrows medially to diamond-shaped ridge. Median portion muscle field slightly elevated and parallel-sided; cov- ering about one-half total area muscle attachment. Paired, narrow diductor impressions lateral to medi- an scar. Small pedicle \callist at posterior part del- thyrial cavity. Anterior and lateral margins strong- ly crenulated. Brachial valve interior.—-Cardinalia consist of rounded cardinal process (narrow, linear cardinal process present in small individuals) rising from notothyrial platform laterally bounded by short, stout brachiophores laterally fused to valve posteri- or margin. Dental sockets bounded anteromedially by brachiophores, posteriorly by interarea, are rela- tively narrow. Broad myophragm, rounded in cross section, descends anteriorly from notothyrial plat- form to point slightly anterior of midlength. Ad- ductor impressions divided medially by low myo- phragm and laterally by paired, laterally diverging myophragms. Laterally diverging myophragms marked posteriorly but decrease in height anteriorly and serve to divide adductor impressions into two scar pairs. Lateral pair adductor impressions about half length of medial pair, both pairs relatively elon- gate in outline. Anterior and lateral margins strong- ly crenulate. M easarements.—Scatter diagram of length versus width of brachial valves is essentially linear (fig. 2), although the short size interval represented by sam- ple precludes conclusions as to linearity or departure therefrom of growth curve as a whole. Sample ap- pears normally distributed which suggests that small specimens have been abstracted at growth site or that this portion of the original population has un- dergone transport and sorting. Slope of growth line about 50° and amount of dispersion not great for this type of orthoid. Comparison—Adequate comparative material is not available with which to ascertain specific identi— ty of the Maine material. It is not markedly dissimi- lar to previously described Ptychopleurella s. s. spe- cies; likely that group of species assigned to this genus needs revision. Occurrence—USGS loc. 3479—SD, Spencer Lake quadrangle, Somerset County, Maine. Stratigraphic location—Base of Hobbstown For- matiOn (Upper Silurian). 20 10 LENGTH, IN MILLIMETERS 0 10 20 WIDTH, IN MILLIMETERS FIGURE 2.-—Relationship of length versus width in impressions of interior of brachial valves of Ptychopleurella sp. USGS 10c. 3479—SD, Spencer Lake quad- rangle, Somerset County, Maine. Distribution—Genus known from Europe and North America. Figured specimens.—USNM 125801A—B, 125802A, 125803B, 125804. Measured specimens.—USNM 125801—125803. Unmeasured specimens.——USNM 125805. Family DINORTHIDAE Genus VALCOUREA Raymond, 1911 Valcourea sp. Plate 2, figures 18—22 Exterior—Pedicle valve concave, brachial valve evenly convex, latter has greater degree of curvature than former. Greatest width at straight hinge line. Pedicle valve interarea steeply apsacline, brachial valve interarea anacline. Pedicle valve interarea about three to four times longer than that of brachi- al valve. Valve outline subcircular to elliptical, but deformation helps determine outline. A study of relationship between length and Width in a number of species of Valcourea and in material from Maine (fig. 3) indicates that degree of variability in ma- terial from Maine is far greater than in other shell populations. The much greater variability of Maine specimens should probably not be ascribed to bio- logic cause but rather to deformation. Relation be- tween length and width changes with size. Width in- creases at greater rate than length, that is, shells tend to become subcircular with increasing size from originally transverse condition. Delthyrium closed by convex pseudodeltidium and notothyrium closed by convex chilidium. Lateral margins of valves even- ly rounded. Brachial valve bears broad sulcus and pedicle valve corresponding low fold. Shells stro- SYSTEMATIC PALEONTOLOGY 9 phomenoid in appearance. Costellae increase by bi- furcation, crossed by fine concentric lines. Anterior commissure faintly uniplicate and crenulate. Pedicle valve interior—Short dental plates later- ally bound deeply impressed muscle field. Hinge teeth short, rounded, supported by dental lamellae. Muscle field tripartite, consisting of median, narrow ad- ductor impression with subparallel sides, laterally bounded by relatively elongate diductor impressions that extend anteriorly beyond adductor impression area. Anterior part adductor impression slightly elevated above corresponding parts of diductor impressions. Brachial valve interior—Cardinalia consist of rounded cardinal process anteriorly joining with rounded median septum and short, stubby brachio- p‘hores. Measurements.—Scatter diagram of length versus width for specimens from Maine shows an abnor- mally great dispersion for an orthoid brachiopod (fig. 3). To check this seemingly abnormal distribu- 20 10 SHELL LENGTH, IN MILLIMETERS 0o 10 20 30 SHELL WIDTH, IN MILLIMETERS FIGURE 3.—Relationship of length versus width of shell impressions of Valcourea sp., Kennebec Formation, USGS 10c. 4317—00, Brassua Lake quadrangle, Somer- set County, Maine. tion, five undistorted samples belonging to four spe- cies of the genus were measured. The study of five plots obtained from undistorted samples showed con- clusively that the sample from Maine had an abnor- mally high degree of dispersion in terms of biologic variation. Abnormal degree of dispersion is ascribed to distortion of specimens by mechanical means. The growth line slope for specimens from Maine is rela- tively meaningless because the amount of mechani- cal deformation undergone by specimens has com- pletely obscured and altered their form. The sample from Maine is normally distributed, but it is not possible to ascribe distribution to either transporta- tion or abstraction of small specimens in situ. Discussion—The material from Maine has under- gone a great amount of deformation making it im- possible to assign specifically because the form of the exterior is the basis used for discriminating between species of the genus (Cooper, 1956). Occurrence—USGS loc. 4317—CO, Brassua Lake quadrangle, Somerset County, Maine. Stratigraphic location—Kennebec (middle Middle Ordovician). Distribution—The genus is restricted to North America and Scotland. Figured specimens.—USNM 125750A, B; 125751, 125752A-C (A, B are unfigured external impres- sions of the figured impression of the interior). Measured specimens.—USNM 125750—125752, 125754—125766, 125768—125771, 126328. Unfigured specimens.—USNM 125767N, 1257758, 125776T. Formation Suborder DALMANELLOIDEA Superfamily DALMANELLACEA Family DALMANELLIDAE Genus RESSERELLA Bancroft, 1928 Resserelln sp. Plate 2, figures 23—27 Exterior—Brachial valve relatively flat, pedicle valve convex, navieulate in form. Straight hinge line, greatest width anterior of binge line near midlength. Brachial valve interarea very steeply hypercline, pedicle interarea gently apsacline. Delthyrium and notothyrium apparently unmodified by plates. An- terior margin rectimarginate, crenulate. Lateral margins evenly rounded, shell outline slightly elon- gate to subcircular. Costellae increase by bifurcation. Pedicle valve interior.—Short dental lamellae lat— erally bound posterior half of muscle field. Hinge teeth elongate, terminally crenulate. Muscle field tri- partite with narrow, median, parallel-sided ad- ductor impression laterally flanked by anteriorly ex- panding diductor impressions. Anterior and lateral shall margins strongly crenulate. Brachial valve interior.——Cardinalia consist of prostrate, posteriorly directed cardinal process later- ally flanked by stubby brachiophores and anteriorly joining low myophragm with rounded cross section. Cardinal process cleft medially by narrow slit. Den- tal sockets floored with crenulated fulcral plates. Medial face dental socket formed from brachiophore and posterior face by hinge line. Adductor impres- sion divided medially by myophragm, bounded later- ally by low ridges of secondary material that extends anteriorly from brachiophore base. Adductor im- pressions quadrilobate, divided by low ridge secon- 10 EARLY PALEOZOIC BRACHIOPODS OF THE MOOSE RIVER SYNCLINORIUM, MAINE dary material extending normal to midline. Anterior pair larger; whole adductor field elliptical in outline. Anterior and lateral margins of shell strongly crenulate. Discussion—The available material is inadequate for specific determination. Occurrence—USGS 10c. 3479—SD, Spencer Lake quadrangle, Somerset County, Maine. Stratigraphic location.——Base of Hobbstown For- mation (Upper Silurian). Figured specimens—USN M 125772—125774. Uufigured specimens.—USNM 125778—125780, 126112. Resserrella? sp. Discussiou.—Single specimen from Hardwood Mountain Formation assigned questionably to Res- serella because specimen has characteristic flat brachial valve and naviculate pedicle valve of genus. Occurrence.—USGS 10c. 3488—SD, Spencer Lake quadrangle, Somerset County, Main-e. Stratigraphic location—Hardwood Mountain For- mation (Upper Silurian). Unfigured specimeu.—USNM 125777. Subfamily ISORTHINAE Genus [SORTHIS Kozlowski, I929 Isorthis sp. 1 Plate 2, figures 28—32 Exterior.—Unequally biconvex shells, pedicle valve having greater degree of convexity. Straight hinge line, greatest shell width near midlength. Pedi- cle valve interarea apsacline, brachial valve inter- area anacline. Shell outline subcircular. Brachial valve faintly sulcate, Anterior margin gently sulcate, crenulate. Pedicle valve interarea about two to three times as long as that of brachial valve. Shell lateral margins rounded. Costellae in- crease by bifurcation, crossed by concentric growth lines. Shell punctate. Pedicle valve iuterior.—Hinge teeth bladelike, supported by short dental lamellae that form lateral margins of elongate muscle field. Muscle field con- tains elongate median adductor track whose anterior portion is raised up from floor of valve by secondary material, and paired, lateral, elongate diductor im- pressions which extend slightly anterior of adductor track margin. Shell interior relatively smooth, does not reflect external costellae. Brachial valve iuterior.——Cardinalia consists of small prostrate cardinal process located between two large, bladelike, laterally diverging brachiophores. Brachiophores connected posteriorly to fulcral plates which serve to floor dental sockets. Posterior face of dental socket formed by palintrope. Brachio‘phores supported basally by brachiophore plates which di- verge laterally, parallel to brachiophores. Sharp ridges extend anteriorly from brachiophore plates which surround muscle field laterally and finally con- verge medially before attaching to rounded, low myo- phragm which bisects muscle field. Myophragm extends from base of cardinal process to about mid- length or to position slightly posterior. Muscle field quadripartite adductor impressions transversely di- vided by low ridges uniting medially with myo— phragm. Anterior and lateral margins finely crenulated. Comparison—Moose River synclino-rium isorthids represented by relatively inadequate material, but meaningful comparisons may still be made. 1. sp. 3 possesses a relatively narrow median adductor track in the pedicle valve which contrasts greatly with the relatively broad median adductor tracks of I. sp. 1 and 2. Subcircular adductor field in brachial valve of I. cf. I. perelegaus contrasts strongly with the longi- tudinally elongate adductor field present in 1. sp. 1, 2, and 3. Pedicle muscle field in I. sp. 2 widens an- teriorly in manner, differing considerably from I. sp. 1. 1. sp. 1 has pedicle valve internal morphology similar in all regards to I. arcuaria, but inadequate material available makes specific determination premature. Occurrence.—USGS 10c. 3479—SD, Spencer Lake quadrangle, Somerset County, Maine. Stratigraphic location—Base of Hobbstown For- mation (Upper Silurian). Figured specimeu8.—USNM 125827—125829. Uufigured specimens.—USNM 125830, 125831. Isorthis sp. 2 Plate 2, figures 33—35 Discussion—Single brachial valve from Hard— wood Mountain FormatiOn and single pedicle valve from Parker Bog Formation have characteristic internal features usually associated with Isorthis. Material too poor for specific identification. Comparison—See I. sp. 1. Occurrence.—USGS 10c. 3495—SD and 3487—SD, Spencer Lake quadrangle, Somerset County, Maine. Stratigraphic l0cati0u.—Hardwood Mountain For- matiOn (Upper Silurian), and Parker Bog Forma- tion (Lower Devonian). Figured specimens.—USNM 125831A, 125826. Uufigured specimens.—USNM 126329. lsortllis sp. 3 Plate 3, figures 1—16 Exterior—Valves unequally biconvex with pedi- cle valve about twice as deep as brachial valve. Valve outline tends to be transversely suboval. Hinge line SYSTEMATIC PALEONTOLOGY 11 straight and less than maximum width which occurs about midlength. External ornament not preserved. Pedicle valve interior—Hinge teeth supported by short, slightly divergent dental lamellae which are continuous anteriorly as subparallel or slightly an- teriorly converging muscle bounding ridges. Di- ductor tracks enclosed by these ridges relatively narrow and rounded, separated by rounded ridge- like myophragm medially. Diduotor tracks extend about 1/3 to 215 length of valve and not deeply im- pressed or elevated on platform anteriorly. Anterior very faintly crenulated by costae at margin of valves. Brachial valve interior—Sockets expand and di- verge anterolaterally with what appear to be poorly developed fulcral plates at bases. Cardinal process base narrow, nearly linear; distal end not exposed. Adductor muscle scars enclosed by relatively broad- ly set-apart muscle bounding ridges that may or may not continue around posterior and anterior pairs. Two pairs about equal size and delineated by small ridges between posterior and anterior pairs; ridges are normal to midline. Low myophragm divides sub- triangular posterior pair adductors. Dorsal margins, like those of pedicle valve, faintly crenulated by im- press of costellae. Comparisou.—See I . sp. 1. Occurrence.—USGS loc. 5587—SD, 5583—SD, 5586— SD, Attean quadrangle, Somerset County, Maine. Stratigraphic location—Hardwood Mountain Formation (Upper Silurian). Figured specimens.—USNM 160113—160119. lsorthis cf. I. perelegans (Hall, 1859) Plate 3, figure 17 Discussion—Single brachial valve closely resem- bles Isorthis perelegaus; convex, subcircular, slight- ly wider than long. Cardinal process flanked by wide- ly diverging brachiophores that join anterolaterally with broadly spaced muscle bounding ridges which enclose rhomboidal muscle impression consisting of posterior and anterior adductor pairs. Comparisou.—See I . sp. 1. Occurrence.—USGS 10c. 3499—SD, Spencer Lake quadrangle, Somerset County, Maine. Stratigraphic location—Beck Pond Limestone (Lower Devonian). Figured specimeu.—USNM 125813. Genus LEVENEA Schuchert and Cooper, 1931 Levenea sp. Plate 3, figures 18-20 Exterior.—Biconvex shells with pedicle valve hav- ing greater convexity. Outline subcircular. Hinge line straight; greatest width about midlength. Later- al margins rounded; anterior commissure rectimar- ginate, crenulate. Radial costellae. Pedicle valve interior—Dental lamellae short and almost obsolete. Hinge teeth form not shown on Maine material. Muscle field tripartite with broad, anteriorly expanding median adductor track, whose anterior portion is raised up from valve floor by pad of secondary material, and relatively narrow, paired diductor tracks inclined almost normally to shell, placed in secondary material which fills valve pos- terior portion. Pallial marks paired; diverge antero- laterally from point where diductor impressions join adductor impression. Brachial valve iuterior.—Cardinalia consist of cardinal process on bladelike base flanked late-rally by stout brachiophores. Anteriorly brachiophores give rise to bladelike brachiophore plates. Anterior- ly from brachiophore plates are subparallel muscle- bounding ridges. Muscle field medially divided by prominent, rounded myophragm. Comparison—Available material too poorly pre- served and fragmentary for specific identification, but form resembles L. subeariuata (Hall. 1857). Occurrence—USGS 10c. 3499-SD, Spencer Lake quadrangle, Somerset County, Maine. Stratigraphic location—Beck Pond Limestone (Lower Devonian). Figured specimeu3.—USNM 125811B, 125814. Unfigured specimeu3.—USNM 125815 A—G. 125812, Unidentified orthoid Plate 3, figure 21 Discussion—The only fossil recovered from the Lobster Mountain volcanics indicates a Middle Ordo- vician or younger age (presumably pre-Givetian in the absence of Givetian or younger marine beds in this region). Occurrence—USGS 10c. 3280—SD, North East Carry quadrangle, Piscataquis County, Maine. Figured specimeu.—USNM 126473. Family DlCAELOSllDAE Genus DlCAELOSlA King, 1850 Dicaelosia sp. Plate 3, figure 22 Exterior—Small shells with strongly emarginate anterior margin; straight hinge line much narrower than greatest width which occurs forward of mid- length. Shells biconvex with pedicle valve having somewhat greater convexity than brachial valve. Valves sulcate. Costellae and concentric growth lines. Comparison—No information regarding specific identity of Maine material is available because of 12 EARLY PALEOZOIC BRACHIOPODS OF THE MOOSE RIVER SYNCLINORIUM, MAINE fragmentary condition. No interiors of either valve are available for study. Occurrence.—USGS 10c. 3479—SD, 3488—SD, and 3499—SD, Spencer Lake quadrangle, Somerset Coun- ty, Maine. Stratigraphic location—Hardwood Mountain For- mation (Upper Silurian), base of Hobbstown For- mation (Upper Silurian), and Beck Pond Limestone (Lower Devonian). Figured specimen.—USNM 125810. Unfigured specimen.—USNM 126330. Family RHIPIDOMELLIDAE Subfamily RHlPlDOMELLlNAE Genus DlSCOMYORTHlS Johnson, 1970 Discomyorthis musculosa solaris (Clarke, 1907) Plate 3, figures 23—25 plate 4, figures 1—3 Rhipidomella musculosa Hall var. solaris Clarke, 1907, p. 284, figs 1, 3, not fig. 2; Clarke, 1909, p. 88, pl. 21, figs. 8, 10—11, not fig. 9. Exterior.—Subcircular to transversely elliptical shells with narrow hinge line and interarea. Brachi- al valve more convex than almost flat pedicle valve. Pedicle valve commonly bears broad, low median sul- cus. Greatest width near midlength. Interarea con- cave, gently apsacline in pedicle valve orthocline in brachial valve. Pedicle valve interarea about three times as long as that of brachial valve. Delthyrium unmodified by plates, but filled by cardinal process. Pedicle valve beak slightly incurved. Anterior com- missure uniplicate and crenulate. Costellae increase by bifurcation. Growth lines concentric. Shell punctate. Pedicle valve interior—Stout hinge teeth border delthyrial cavity and project dorsally from anterior portion narrow interarea. Hinge teeth supported by short dental lamellae; almost obsolescent in large specimens. At base of dental lamellae lateral ridge continues anterolaterally to point slightly posterior of midlength and then turns anteromedially almost to surround muscle field. Muscle attachment area large, subcircular to transversely elliptical in outline, commonly reaches three-quarters of distance to an— terior margin. Muscle field bisected by sharp low myophragm. Large, flabellate diductor impressions surround small, elongate, posteriorly situated ad- ductor impressions. Several radially disposed myophragms traverse diductor impressions. Posterior face delthyrial cavity occupied by pedicle callist; which is located on slight- ly raised pad of secondary material. Umbonal cavi- ties commonly almost completely filled with secon- dary material. Anterior and lateral shell margins crenulated by fiat crenulae which are grooved on upper faces. Valve interior, where not occupied by muscle impressions, relatively smooth due to deposi- tion of secondary material. Brachial valve interior.——Cardinalia consist of erect cardinal process which swells distally, has trifid posterior face, and joined posterolaterally with bladelike brachiophores. Brachiophores supported by brachiophore plates which form outer edge of dental sockets. Extending anteriorly from base of cardinal process is broad, low myophragm which bisects ad- ductor impressions. Four adductor impressions, but posterior and anterior pairs not well differentiated. They extend anteriorly to position short of mid- length. Anterior and lateral margins bear flat crenu- lae which may be grooved on upper faces. Measurements—The lengthzwidth scattergrams for Maine specimens show essentially linear growth curve with slope about 50° (figs. 4C, D). Similar material from three Appalachian localities has a much smaller degree of dispersion than the Maine material, strongly suggesting that the Maine ma- terial is mechanically deformed to a large degree. The Maine material is normally distributed, which suggests that either small specimens have been ab- stracted or tranSportation and sorting has taken place. Degree of dispersion of the more highly con— vex brachial valves is greater than the relatively flat pedicle valves from Maine which suggests crushing has been important in producing disper- sion. Amount of absolute variation increases with size increase, but there is no suggestion that per- centage variation increases with size. Distribution of three samples from the Appa- lachians suggests all have similar slopes and growth curves; degree of dispersion is of the same order of magnitude. Relation between length and width of the diductor muscle field of the pedicle valve was studied (fig. 4E). The growth curve slope is about 40° but the degree of dispersion is great which prevents, with present samples, estimating amount of departure from linearity. Considerable variation between di- ductor length and width suggests that this relation is not useful in specific characterization of rhipi- domellids. It can be inferred that size and form of diductor field were not critical to the animal. Figure 4F shows the relationship between pedicle valve width and interarea width. Available informa- tion suggests a relatively linear relation; amount of dispersion is small. Slopes show a significant degree of difference between samples. Differences in slope are about 13° for D. musculosa solaris and about 25° for D. oblata. This relation emphasizes that inter- area width appears to be a rhipidomellid character which remains relatively constant in any one sam- SYSTEMATIC PALEONTOLOGY 13 ple and has a wide variation span within a generic grouping. Inadequate information is available to determine whether differences should be accorded specific or generic rank. Relationship between angle included by diductor muscle field (lateral ridge angle) and pedicle valve width is very variable (fig. 4G). The angle does not change consistently with change in specimen size. Degree of dispersion between samples is very great, but the significance of this diiference is not apparent. Relationship of left lateral ridge angle to right lateral ridge angle is random, as expected, of bi- laterally symmetrical shells (fig. 4H); all samples exhibit about the same degree of dispersion and about the same angular relationships. Relationship of pedicle valve width and width of diductor field is linear (fig. 4B) , having a low degree of dispersion, indicating muscle field increases linearly with increase in shell width. Relationship between brachial valve width and angle included be- tween the brachiophores is essentially random (fig. 4A) , suggesting it remains constant with shell width increase. , Discussion—This subspecies was first described by Clarke (1907), as a variety of Rhipidomella musculosa (Hall, 1857), from material collected by Olof Nylander. In the original description, “Moose- head Lake, Baker Brook Point; Brassua Lake, east side; Moose River at Stony Brook, Me.” listed as localities (Clarke, 1907, p. 284). Later Clarke gave the same locality list (Clarke, 1909, p. 88) ; on page 208 of same volume, however, localities were as- signed to four figured specimens (three previously figured in Clarke, 1907)—figures 8, 10, and 11 cited from Tomhegan Point, Moosehead Lake and figure 9 from Jackman farm. Because rocks and fossils of Oriskany age only occur on Tomhegan Point and Discomyorthis musculosa sola-ris is limited in range to beds of Schoharie age in this region, it seems clear that the specimens represented by figures 8, 10, and 11 were collected from beds of Schoharie age (probably on Baker Brook Point). Figure 9, repre- senting a brachial valve, possesses the cardinalia and other characters of Platyorthis; therefore, it is reasonable that it was collected from beds of Oris- kany age at J ackman farm. Clarke’s original description indicates that he founded his variety [sic] upon the characters of the pedicle valve: “These are all small shells with the enormous adductor [sic—adductor-diductor] scar in a state of high development. The shells are somewhat less circular, more transverse than in the New York and Grande Gréve Oriskany specimens of R. muscu- losa, but their specific identity is not greatly veiled” (Clarke, 1907, p. 284). It is recommended that D. musculosa solaris (Clarke, 1907) be retained for the form whose pedicle valves are figured in accordance with his implied intent. Comparison.—Discomyorthis musculosa has a slightly wider interarea than D. musculosa solaris, but relationships between pedicle valve length and Width and pedicle valve muscle field are very similar in both forms. D. oblata tends to have a smaller mus- cle field in the pedicle valve and wider interarea than does D. musculosa solaris. Discomyorthis alsa, al- though having characteristic marginal crenulations of Dalejina and Discomyorthis, does have a far wider interarea than do Discomyorthis oblata, Discomyor- this musculasa, or the form from Maine. Interarea of Discomyorthis eryna is similar to Discomyorthis alsa and may prove to be conspecific when adequate material of both species is studied. Occurrence.—USGS locs., 2723—SD, 2730-SD, 2750—SD, 2752—SD, 2814—SD, 2820—SD, 2839-SD, 2840—SD, 2842—SD, 2852-SD, 2873—SD. Brassua Lake quadrangle, Somerset County, Maine. Stratigraphic location.—Tomhegan Formation (Lower Devonian). Distribution—Northern Maine (Tomhegan For- mation), northern New Hampshire (Littleton For- mation), possibly Green Pond area of New Jersey if Kanouse Sandstone “Schizophoria sp. cf. S. striatula (Schlotheim) ” Weller (1903) belongs to this species, and possibly “R. alsa? Hall” of Dunbar (1919) from western Tennessee (Camden Chert). Holotype.—The interior impression of a pedicle valve figured by Clarke (1907, p. 284, right side of page, NYSM 8505) is here selected as the holotype. Figured specimens.—USNM 125806-125809. Unfigured specimens.—USNM 126317-126327. Measured specimens.—USNM 125806—125809, 125816. Discomyorthis sp. Plate 4, figure 10 Discussion—Specifically unidentifiable Discomy- orthis Specimens have been obtained from Beck Pond Limestone (moderate size, up to 3%; cm wide, rela- tively short hinge line, relatively restricted pedicle valve muscle field) and Discomyorthis sp. from McKenney Ponds Member of Tarratine Formation (moderate size, flat pedicle valve, relatively convex brachial valve, short hinge line, broadly flabellate pedicle valve muscle field). Occurrence.——USGS locs. 3499—SD, Spencer Lake quadrangle, 2810—SD, 2806—SD, 2864—SD, Pierce Pond quadrangle, Somerset County, Maine. Stratigraphic location—Beck Pond Limestone 14 EARLY PALEOZOIC BRACHIOPODS OF THE MOOSE RIVER SYNCLINORIUM, MAINE 40 40 A E a? I'- LLI LL] ’— E . . Lu 3 30 E 30 :1 :1 5 . 2 a z I - 3:~ .' I- . g 20 , : ° 5 20 5 . . . . . . - g _ < - ' - .J . : . > . . . < .. ‘2 _' . . > _ . <_( 10 . L3 10 I - 9 2 E} g - ' a o 0 40 5o 60 7o 80 90 o 10 20 30 BRACHIOPHORE ANGLE, IN DEGREES DIDUCTOR WIDTH, IN MILLIMETERS (D [r (n E 30 g 30 LLI C I— D : Lu 3 E - :' - j - E _ 0' ' . u z . - E - _ 20 - . . ' z 20 _ ' g ‘ .' ..|‘.- . S . . ': If: ' . . .n E . 2 _. . . u g . 0 , u o ">4 ~.:'-- 3 10 :,.,-‘ 4 10 .' '- 3 .‘l ' ‘ g .1 :- <( - I 4 i E .' _] 5 2 < 0 0 o g o 10 20 3o 40 E o 10 20 30 4o BFIACHIAL VALVE WIDTH, IN MILLIMETERS PEDICLE VALVE WIDTH, IN MILLIMETEFIS m 30‘. E E '— In E . .J 2' 2 2O _ Z . - -. I. - ~ I- - ' o . .. Z . ' 5 10 .."."’ m o O . ' "s . I- -' O :> 9 D o 0 1o 20 30 DIDUCTOR WIDTH, IN MILLIMETERS FIGURE 4.-—Scattergrams for Discomyorthis musculosa solaris (Clarke, 1907) from the Tomheg'an Formation. USGS 10c. 2750—SD, Brassua Lake quadrangle, Somerset County, Maine. PEDICLE VALVE WIDTH, IN MILLIMETERS LEFT LATERAL RIDGE ANGLE, IN DEGREES SYSTEMATIC PALEONTOLOGY 40 40 VJ 0: Lu .— E 30 j 30 :' E ; Z 1‘ 20 '5 2o ' '. E IJJ > _I < > 10 L_ul 10 E2 0 LIJ O. o o o 10 20 50 so INTERAREAWIDTH, IN MILLIMETERS 60 . . 50 . 4o 30 20! 20 30 4o 50 60 RIGHT LATERAL RIDGE ANGLE, IN DEGREES FIGURE 4.—Continued. LATERAL RIDGE ANGLE, IN DEGREES LEFT BRACHIOPHORE ANGLE, IN DEGREES 70 80 90 100 110 50 4o 30 20 20 30 40 RIGHT BRACHIOPHORE ANGLE, IN DEGREES 15 16 EARLY PALEOZOIC BRACHIOPODS OF THE MOOSE RIVER SYNCLINORIUM, MAINE (Lower Devonian) and McKenney Ponds Member of Tarratine Formation (Lower Devonian). Figured specimen.—USNM 125823. Unfigured specimens—U S N M 125822A,B. 125821A,B, Genus DALEJINA Havlicek, 1953 Dalejina 3p. 1 Plate 4, figures 6—9 Discussion—The base of the Hobbstown Forma- tion yielded one pedicle valve and one brachial valve possessing characters of Dalejina including flat cren— ulations with grooves on inner faces. Paucity of ma— terial and present difficulty in accurately defining already named species make it inadvisable to identi- fy specifically Maine material. This subequally bi- convex form, having short hinge line, small size, relatively small and nonflabellate pedicle muscle field, is similar to the Silurian species D. cliftonensis, D. newsomensis, and D. subtriangularis. Costellae of the form from the Hobbstown Formation are somewhat coarser than in the form from the underlying Hard- wood Mountain Formation (species 2). The outline of Hobbstown Formation specimens is more like that of D. subtriomgalaris than of any other species men- tioned above. Occurrence.—USGS 10c. 3479—SD, Spencer Lake quadrangle, Somerset County, Maine. Stratigraphic location—Base of Hobbstown For- mation (Upper Silurian). Figured specimens.—USNM 125821, 125822. Daleiina sp. 2 Plate 4, figures 4, 5 Discussion.—A few poorly preserved specimens having Dalejina characters, including crenulate in- ternal margin on which crenulae are flat and have grooved upper faces, were obtained from several localities in the Hardwood Mountain Formation. Pedicle and brachial valves sub-equally biconvex, short hinge line, relatively small and nonflabellate pedicle valve muscle field, and relatively fine costel- lae. Shell outline subcircular. Specimens similar to Silurian species enumerated under description of Dalejina sp. 1 from the base of the Hobbstown For- mation; like it, inadequate for making specific identi- fication. Hardwood Mountain Formation material has somewhat finer costellae than those from 'the base of the Hobbstown Formation. 0ccurrence.—USGS locs. 2950—SD and 3469-SD, Spencer Lake quadrangle, Somerset County, Maine. Stratigraphic location—Hardwood Mountain For- mation (Upper Silurian). Figured specimen.—USNM 125824. Unfigared specimens.—USNM 126346—126349, 126440-126442. Daleiinn? :1). or Discomyortllis'.’ 5p. Discussion—Parker Bog Formation yielded few very poorly preserved, highly deformed rhipidomel- lids that probably belong to Dalejina or Discomyor- this. Not possible to make positive assignment with- out information regarding crenulations along inter- nal margins. Specimens have relatively short hinge line. Occurrence.—USGS loc. 3477—SD, Spencer Lake quadrangle, Somerset County, Maine. Stratigraphic location—Parker Bog Formation (Lower Devonian). Subfamily PLATYORTHlNAE Genus PLATYORTHIS Schuchert and Cooper, 1931 Discussion—In addition to the distinguishing characters of Platyorthis, described by Schuchert and Cooper (1931, p. 246) including the plane-con- vex profile and large, posteriorly reclining cardinal process, bifurcating crenulations occur along inter- nal margins. These crenulations are unlike those of other members of the Rhipidomellidae, and there- fore, serve to distinguish even fragmentary platyor- thid material from other genera of the family. Platyortllis planoconvexa (Hall, 1859) Plate 4, figures 11—18 Platyorthis planoconvexa (Hall, 1859), see Boucot, 1959b, p. 739—740, pl. 90, figs. 3—10. Exterior.—Brachial valve flat, pedicle valve con- vex and has carinate form. Shell outline subcircular with greatest width near midlength. Hinge line straight, very narrow, interarea small. Pedicle valve interarea concave, gently anacline to orthocline; brachial valve interarea anacline. Delthyrium un- modified, occupied by projecting cardinal proceSs of opposite valve. Pedicle valve interarea about twice as long as that of brachial valve. Anterior commis- sure rectimarginate and crenulate. Pedicle valve beak slightly incurved. Costellae increase by bifurca- tion; crossed by concentric growth lines. Shell punctate. Pedicle valve interior.——Stout, bladelike, laterally diverging hinge teeth supported by short dental lamellae. Hinge teeth posteriorly connected by small, horizontal plates flooring crural fossettes with pos- terior valve margin. Dental lamellae inner faces in- dented by crural fossettes. Umbonal cavities com- monly almost filled with secondary material; conse- quently dental lamellae tend to become obsolete in larger specimens. Muscle field large, flabellate, com- monly includes angle of about 30°, reaches from two- SYSTEMATIC PALEONTOLOGY 17 thirds to three-quarters distance to anterior margin, bisected by low, narrow myophragm. Muscle impres- sions divided into pair large diductor scars which expand anteriorly; divided by number of radially disposed, short myophragms into number of sectors. Small, elongate adductor impressions, poorly im- pressed posteriorly, abut pedicle callist. Pedicle callist located on posterior wall of delthyrial cavity upon small pad secondary material. Internal mar- gins bear crenulae that bifurcate peripherally. Mus- cle field in large specimens bordered laterally by low, rounded ridge of secondary material that extends anteriorly frOm base of dental lamellae. Brachial valve inten‘or.—Gardinalia consist of posteriorly directed, terminally bifid cardinal process fused laterally with laterally diverging brachio- phores. Brachiophores stout, supported by short brachiophore plates. Dental sockets anteriorly bor— dered by brachiophores and posteriorly by low ridge slightly anterior of interarea. Low ridges diverge in slightly anterolateral direction. Muscle area transversely elliptical to subcircular in outline; restricted to posterior half of valve. Di- vided by low, rounded myophragm extending an- teriorly to about midlength. Paired adductor impres- sions elliptical in outline; may be transversely divided by pair of low, transverse ridges into four impressions. Lateral portions muscle field bounded by low ridge of secondary material that extends an- teriorly from base of brachiophore plates. Measurements.——Relationship between brachial valve length and width suggests growth is relatively linear. Dispersion of Maine material is several times greater (fig. 5A) than undeformed material from New York (fig. 53) suggesting that Maine material had linear dimensions altered mechanically. Maine material is normally distributed. Comparison.—Orthis lucia Billings, 1874 is a j un- ior synonym of Platyorthis planoconvexa, the former being described from the Grande Gréve Limestone of Oriskany age. No criteria were observed that would enable one to discriminate between the platyorthids found in northern Maine and those known elsewhere in North America. Occurrence.—USGS locs. 2721-SD, 2834—SD, 2691—SD, 2813—SD, 2777—SD, 2803-SD, 2860—SD, 3471—SD, 2862-SD, 2770-SD, 2710—SD, 2857—SD, 2870—SD, 2760-SD, 3474—SD, 2771-SD, 3093—SD, 2810—SD, 3486—SD, 2806-SD, 3229—SD, Somerset and Piscataquis Coun- ties, Maine. Stratigraphic locatiom-Seboomook Formation (Lower Devonian), McKenney Ponds Member of the 2701—SD, 2698-SD, 2719—SD, 2711—SD, 3482-SD, 2700—SD, 2767 —SD, 2751—SD, 40 30 20 LENGTH, IN MILLIMETERS o 20 10 0 10 2O 30 WIDTH, IN MILLIMETERS FIGURE 5.—Comparison of length versus width of brachial valves of Platyorthis planoconvexa (Hall, 1859). A, lower sandstone of the Tarratine Formation, USGS 10c. 2701— SD, Brassua Lake quadrangle, Somerset County, Maine. B, Glenerie Limestone, on N.Y. Rte. 9W, 1 miles north of Glenerie and 1 mile south 0f Cockburn, N.Y. Tarratine Formation, and the sandstones of the Tar- ratine Formation (Lower Devonian). 18 EARLY PALEOZOIC BRACHIOPODS OF THE MOOSE RIVER SYNCLINORIUM, MAINE Distribution—The species is known from eastern North America. The European species Platyorth'ls circularis is similar in morphology. Figured specimens.—USNM 125825. Measured specimens—125817. Unfigured specimens.—-USNM 125817, 126345, 125930—125939, 126350—126367. Superfamily ENTELETACEA Family SCHIZOPHORIIDAE Subfamily DRABOVIINE Genus SALOPINA Boucot, 1960 Salopina hitchcocki Walmsley, Boucot, and Harper, 1969 Plate 5, figures 1—11) Salopina hitchcocki, Walmsley, Boucot, and Harper, 1969, p. 512—513, pl. 71, figs. 1—4; pl. 72, figs. 1—5. Exterior.—Unequally biconvex shells with evenly convex brachial valve and subconical pedicle valve. Hinge line straight, greatest width occurs about midlength. Lateral margins of shell rounded. Shell outline transversely elliptical. Brachial valve may bear low sulcus. Anterior commissure rectimargin- ate, crenulate; sulcate shells slightly emarginate. Brachial valve interarea steeply anacline; pedicle valve interare-a steeply apsacline. Pedicle valve inter- area about four to five times as long as that of brachial valve. Delthyrium unmodified. Costellae in- crease by bifurcation; crossed by concentric growth lines. Pedicle valve interior—Hinge teeth stout, located on medial edge of interarea. Hinge teeth supported by short dental l-amellae which laterally bound cor- date muscle field. Muscle field restricted to delthyri- al cavity, located on slightly raised pad of secondary material. Ordinarily muscle field indivisible, on few specimens discernible median adxductor track lateral- ly bounded by elongate diductor impressions. In- ternal margins shell crenulate. External costellae impression not present on larger shells coated with secondary material, but evident on smaller specimens. Brachial valve interior.—Cardinalia consist of linear, small, prostrate cardinal process laterally bounded by laterally divergent brachiOphores. Brachiopho-res basally supported by laterally diver- gent brachiophore plates and posteriorly supported by fulcral plates. Fulcral plates floor dental sockets. Low, rounded myophragm extends anteriorly from cardinal process in some specimens (where deposi- tion secondary material progressed) and serves, in these specimens, to separate deeply impressed mus- cle field. Such a myo-phragm commonly absent when specimen lacks deeply impressed muscle field. Muscle field quadripartite, subcircular outline, transversely 125819, 125820, subdivided by low ridges which join medially with myophragm. Muscle impressions subequal, anterior adductors larger, bounded laterally and anteriorly by ridge which begins at base of brachiophore plates, extends anteriorly then turns medially to join myo- phragm. Valve interior may be strongly crenulated by impression of external costellae, or smooth ex- cept for crenulations along internal margins in speci- mens where secondary material deposited. Measuremeats—Relationships of length and Width of both valves are obscure in the sample studied because of the great amount of dispersion and relatively small size range of specimens (fig. 6). Relationship does appear linear. The amount of dis- persion appears more than predicted for a small orthoid suggesting that physical deformation strong- ly affected specimens. Comparison—See Walmsley, Boucot, and Harper (1969) for discussion and comparison of the species assigned to Salopina. 20 Brachial valves 10 20 Pedicle valves LENGTH, IN MILLIMETERS 0 1O 20 WIDTH, IN MILLIMETERS FIGURE 6.—Comparison of length versus width of brachial and pedicle valves of Salopina hitchcocki Walmsley, Boucot, and Harper, 1969. Tarratine Forma- tion, Somerset County, Maine. SYSTEMATIC PALEONTOLOGY 19 Occurrence—USGS locs. 2705-SD, 2720—SD, 2719—SD, 2722—SD, 2704—SD, 2711—SD, 2721—SD, 2743—SD, 2775—SD, 2861—SD, 2862—SD, 2872—SD, 2811—SD, 2729—SD, 2731—SD, 2890—SD, 2860—SD, 2727—SD, Somerset County, and 2884—SD, Pisca- taquis County, Maine. Stratigraphic l0cati0n.—T‘arratine Formation (Lower Devonian), and Seboomook Formation (Lower Devonian). Holotype.—USNM 125782. Figured specimens.—USNM 125793A, 125798. Unfigured specimens.—USNM 126331—126344. Measured specimens—125784. 125781—125783, Suborder PENTAMEROIDEA Superfamily PENTAMERACEA Family GYPIDULIDAE Subfamily GYPIDULINAE Genus GYPIDULA Hall, 1867 Gypidula sp. 1 Plate 5, figures 12—14 Exterior.—Unequally biconvex shells, pedicle valve possessing greater degree convexity. Neither fold nor sulcus appears present. Shell outline subcir- cular with lateral and anterior margins rounded. Anterior commissure apparently rectimarginate. Hinge line short, probably straight. Pedicle valve interarea incurved, very narrow, anacline; that of brachial valve not available. Delthyrium appears open, unmodified. Shell smooth; does not appear to possess fine ornamentation, spines, or granules. Pedicle valve interior—Small, posrteriorly located spondylium consists of medially convergent dental lamellae supported by short median septum. Median septum striated by curved growth lines. Nature of hinge teeth not ascertained. Umbonal cavity very deep with spondylium near level of commissure. Aniter‘iormost part of spondylium is free. Brachial valve interior—Outer portions brachial plates discrete, extend anteriorly about one—third length valve. Inner portions brachial plates short, apparently expand posteromedially. Comparison—Nonsulcate G. sp. 1 opposed to sul- cate G. sp. 2. Occurrence—USGS 10c. 3488-SD, Spencer Lake quadrangle, Somerset County, Maine. Stratigraphic location—Hardwood Mountain For- mation (Upper Silurian). Figured specimens.—USNM 126508, 126509A. Unfigured specimens.—USNM 126373, 126374. Gypidula sp. 2 Plate 5, figures 15—23 Exterior.—Unequally biconvex shells, pedicle valve having greater degree convexity. Brachial valve bears broad, shallow sulcus, pedicle valve cor- responding fold. Shell outline subcircular t0 ellipti- cal. Anterolateral margins rounded, hinge line short. Pedicle valve beak strongly incurved. Interarea very restricted. Shell exterior costate, about four costae in brachial valve sulcus, four on each flank. Anterior margin sulcate, crenulate. Brachial valve interior—Outer portions brachial plates discrete. Comparison—G. sp. 2 (two specimens available for study) ornamentation resembles G. coeymahen- sis, but lack of material precludes specific identifi- cation. Discussion—The specimen referred to by Wood- ard (1951, p. 76) and dated as “Siluro-Devonian” is shown on plate 5, figures 15—19. Occurrence—USGS loc. 2728—SD, Spencer Lake quadrangle, USGS loc. 5587-SD, Attean quadrangle, Somerset County, Maine. Stratigraphic location—Hardwood Mountain For- mation (Upper Silurian). Figured specimens.—USNM 125840, 160120. Gypidula sp. 2? Plate 6, figures 1—2 Exterior.—Unequally biconvex shells with pedicle valve having greater degree convexity. Neither fold nor sulcus appear present on material. Shell outline subcircular, anterior and lateral margins rounded. Hinge line short. Pedicle valve beak strongly in- curved. Shell exterior costate, about fourteen low, rounded costae each valve. Pedicle valve interior—Small spondylium located posteriorly is supported by median septum extending anteriorly about one-half length of valve. Spondyli— um formed by convergent dental lamellae. Brachial valve interior—Short outer portions brachial plates rest on valve floor, slightly lateral to low, short myophragm which bisects (very poorly impressed) muscle field. Outer portions brachial plates long and extend anteriorly about one-third length of valve, somewhat concave medially. Posteri- or portions inner brachial plates turn laterally to form broad plates. Discussion—The single silicified brachial valve from USGS loc. 3485-SD has internal characters of Gypidula. Pedicle valve from loc. 3496—SD assigned to Gypidula because of similar ornamentation (no brachial valve recovered at this locality). Occurrence—USGS locs. 3496—SD, 3485—SD, Spencer Lake quadrangle, Somerset and Franklin Counties, Maine. Stratigraphic location—Hardwood Mountain For- mation (Upper Silurian). 20 EARLY PALEOZOIC BRACHIOPODS OF THE MOOSE RIVER SYNCLINORIUM, MAINE Figured specimen.—USNM 125842. Unfigared specimens.—USNM 126373, 126374. Genus SIEBERELLA Oelllert, 1887 Sieberella beckensis Boucot, n. 5]). Plate 6, figures 3—8 Exterior.—Biconvex shells, pedicle valve has greatest degree convexity. Outline transversely ellip- tical. Brachial valve bears broad, low sulcus, pedicle valve has corresponding fold. Hinge line straight, greatest width near midleng'th. Pedicle valve inter- area steeply apsacline, slightly concave. Anterior commissure sulcate, crenulate. Pedicle valve fold and corresponding brachial valve sulcus develop at length about 6 mm. Brachial valve beak incurved. Angular costae, increase by bifurcation, crossed by concentric growth lines. About three costae on fold, four to six on each flank. Sulcus commonly contains two costae. Pedicle valve interior—Small spondylium consists pair small medially conjunct dental lamellae sup- ported by median septum. Median septum extends anteriorly about one-half valve length. Anterior and lateral portions interior reflect external ornamentation. Brachial valve interior—Outer plates medially conjunct, supported by median septum, inner plates turn laterally to form hinge plates. Dental sockets not worked out. Interior impressed by costae. Comparison.——Sieberella beckensis is distinguished from S. sieberi by latter’s greater number of costae on fold and sulcus, five to six and four to five respec- tively. S. roemeri has relatively lower, more rounded costae than S. beckensis. Occurrence.——USGS 10c. 3499—SD, Spencer Lake quadrangle, Somerset County, Maine. Stratigraphic location—Beck Pond Limestone (Lower Devonian). Holotype.—USNM 125841A. Figured specimens.—USNM 125841, 125839, 125837, 125838. Unfigared specimens.—USNM 126375. Family PENTAMERIDAE Subfamily PENTAMERINAE Genus PENTAMERUS Sowerhy, 1812 Pentamerus? sp. Plate 6, figures 9—10 Exterior—Evenly convex pedicle valve (no brachial valves found), intera’rea almost entirely ob- solete. Beak incurved, anacline. Shell exterior smooth, open delthyrium appears unmodified. Cardi- nal margin terebratulid. Pedicle valve interior.—Medially convergent den- tal lamellae form spondylium which extends anteri- orly at least one—third valve length. Posteriormost portion delthyrial cavity longitudinally wrinkled. Remarks—Pending recovery of a brachial valve, generic identification is uncertain; both Pentamerus and Pentameroides have similar pedicle valves. Occurrence—USGS loc. 3475-SD, southern end of summit of Limestone Hill, Stratton quadrangle, Somerset County, Maine. Stratigraphic location.—Lime-silicate hornfels of Early Silurian age. Figured specimens.—USNM 125835—125836. Unfigured specimens.—USNM 126372. Suborder STROPHOMENOIDEA Superfamily PLECTAMBONITACEA Family PLECTAMBONITIDAE Subfamily PLECTAMBONITINAE Genus SOWERBYITES Teicherl, 1937 Sowerbyiles? sp. Plate 6, figure 11 Exterior—Brachial valve gently concave, pedicle valve evenly convex. Greatest width at straight hinge line. Lateral and anterior margins rounded. Anterior commissure rectimarginate. Pedicle valve interarea apsacline, that of brachial valve anacline. Radiating coarse and fine costellae crossed by concentric growth lines. Remarks—A few Kennebec Formation specimens possess exteriors resembling Sowerbyites; in absence of information regarding internal structures, gen- eric assignment these shells uncertain. Occurrence—USGS loc. 4317—CO, Brassua Lake qaudrangle, Somerset County, Maine. Stratigraphic location—Kennebec (middle Middle Ordovician). Figured specimens.—USNM 126152. Unfigured specimens.—USNM 126676. Superfamily STROPHOMENACEA Family LEPTAENIDAE Genus LEPTAENA Dalman, 1828 Leplaena "rhomboidalis" (Wilckens, 1769) Plate 6, figures 12—16 Leptaena “rhomboidalis” (Wilckens, 1769), see Boucot, 1959b, pl. 96, figs. 1—2. Exterior—Posterior portion brachial valve flat, anterior portion geniculate, shell has concave ap- pearance. Posterior portion pedicle valve flat to gent- ly convex, anterior portion geniculate. Shell outline transversely elliptical. Maximum Width situated at straight hinge line. Fine, radiating costellae crossed by concentric wrinkles. Anterior and lateral margins rounded. Pedicle valve interior—Hinge teeth stubby. Mus- cle field deeply impressed, restricted to posterior half of valve, subcircular in outline, pair of narrow, medi- an impressions flanked laterally by elliptical, rela- tively large, diductor impressions. Valve interior pustulose and uncorrugated due to deposition secon- dary material. Formation SYSTEMATIC PALEONTOLOGY 21 Brachial valve interior—Cardinalia consist of posteriorly bilobed cardinal process laterally sup- ported by buttress plates, medially joining median septum which extends anteriorly to about midlength. Muscle field deeply impressed, consists of subcircu- lar, posterior adductor impressions and elongate anterior adductor impressions laterally bounded by ridge secondary material. Pair of lateral septa occur at anterior part of muscle field. Occurrence—USGS 10c. 3601—SD, 3400-SD, 2806—- SD, 3497—SD, 2730—SD, 348‘8—SD, 3475—SD, 3479— SD, 3477—SD, 3499—SD, Somerset County, Maine. Stratigraphic locatiou.—Lime-silicate hornfel-s of Early Silurian age, base of Hobbstown Formation (Upper Silurian), Hardwood Mountain Formation (Upper Silurian), Parker Bog Formation (Lower Devonian), Back Pond Formation (Lower Devoni- an), McKenney Ponds Member of the Tarratine Formation (Lower Devonian), and the Tomhegan Formation (Lower Devonian). Figured specimens.—USNM 126127. Uufigured specimens.—USNM 127376—127383. Family STROPHEODONTIDAE Subfamily STROPHEODONTINAE Genus STROPHEODONTA Hall, 1852 Slropheodonta cf. 5. demissa (Conrad, 1842) Plate 6, figures 17—19 Exterior—Pedicle valve convex, brachial valve concave. Maximum width at straight hinge line. Out- line subcircular. Posterior part lateral margin straight, anterior part rounded as is anterior mar- gin. Anterior commissure rectimarginate, crenulate. Pedicle valve interarea orthocline, that of brachial valve hypercline. Interareas both valves short. Cos- tellae with rounded cross sections separated by rounded interspaces. Costellae originate by bifurca- tion, crossed by concentric growth lines and filae. Brachial valve interior—Hinge line wholly den- ticulate. Cardinalia consist of bilobed cardinal process. Muscle field outline subcircular, extends anteriorly to about midlength. Median, small, elon- gate adductor impressions anteriorly pounded by breviseptum, laterally and anteriorly bounded by two pairs adductor impressions. Peripheral regions valve pustulose. Comparison.—Ornamentation and form of Maine specimens are similar to S. demissa. Occurrence—USGS loc. 2820-SD, 2750—SD, Bras- sua Lake quadrangle, Somerset County, Maine. Stratigraphic location—Tomhegan Formation (Lower Devonian). Figured specimens.—USNM 126145, 126144. 126141, 126146, afl. Stropheodonta Ip. Plate 6, figures 20—23 Exterior—Pedicle valve very convex and inflated. Maximum width located near midlengtn. Hinge line straight. Posterior part lateral margin relatively straight, anterior part and anterior margin rounded. Anterior commissure rectimarginate. Interarea rela- tively short, orthocline. Costellae coarser in umbonal region than peripherally, part of this apparent weak- ening may be due to abrasion. Outline variable. Pseu- dodelbi‘dium covers delthyrium. Pedicle valve interior—Hinge line wholly denticu- late. No dental lamellae observed although thicken- ings lateral to muscle field present. Two pits present lateral to well-developed median process. Muscle field weakly impressed to well impressed, paired postero- medial adductors separated by narrow depression, more broadly flabellate anterolateral diductor impressions. Comparison—This material does not belong to Stropheodouta s. s. as defined by Williams (1953, p. 34—35) because it lacks the coarse costellation of that genus. Unfortunately the available material is inade- quate for specific description. Occurrence.—USGS locs. 2820—SD, 2750—SD, Brassua Lake quadrangle, Somerset County, Maine. Stratigraphic location—Tomhegan Formation (Lower Devonian). Figured specimens.—USNM 126126, 126153, 126125. Genus LEPTOSTROPHIA Hall, 1892 Leplostrophia cf. L. magnifies (Hall, 1857) Plate 7, figures 1—9) Exterior—Pedicle valve gently convex, brachial valve gently concave. Hinge line straight, maximum width near midlength. Lateral and anterior margins rounded. Anterior commissure rectimarginate. Pedi- cle valve interarea apsacline, relatively long, brachi- al valve interarea anacline, relatively short. Delthyrium open, includes angle of about 60°—90°. Delthyrium apex occupied by secondary material. Shell outline subcircular. Fine costellae crossed by weak concentric wrinkles. Pedicle valve interior.-—Hinge line wholly denticu- late. Cardinal process pits well defined, separated medially by ventral process. Ventral process narrow, posteriorly fills delthyrial cavity, forms small pseu- dvospondylium. Prominent lateral ridges diverge from delthyrial cavity margins, laterally bound flabellate muscle field. Lateral ridges include angle of 30°—90°. Flabellate muscle field medially divided by low rounded ridge which extends anteriorly from ventral process to about midlength or position Slight- 22 EARLY PALEOZOIC BRACHIOPODS OF THE MOOSE RIVER SYNCLINORIUM, MAINE ly anterior. Adductor impressions small, elliptical, situated just anterior of ventral prOcess. Large, flabellate diductor impressions divided into number of radial sectors by series of myophragms. Umbonal cavities strongly pustulose. Brachial valve interior.—Cardinal process lobes stout, laterally flanked by large socket plates. Small chilidium located between posterior portion cardinal process lobes. Wide low rounded ridge descends from cardinal process lobes to position posterior of mid- length, becoming progressively narrower as it pro- ceeds anteriorly. Adductor impressions relatively narrow, extend anteriorly to position slightly pos- terior of midlength. Umbonal region very pustulose lateral of muscle field, remainder of interior crenu- lated by external ornamentation impress. Comparison—The Maine specimens most strong- ly resemble L. magnifica, but the material is too poor- ly preserved for positive specific identification. No specimens have been assigned to Leptostrophia in the absence of the diagnostic brachial valve interior. Occurrence.—USGS loc. 2700—SD, 2705—SD, 2710— SD, 2813—SD, Somerset County, Maine. Stratigraphic location—Lower sandstone of the Tarratine Formation (Lower Devonian). Figured specimens.-—USNM 126121-126123, 126142, 126158, 126159. Unfigured specimens.—USNM 126155. Lepiostropllin? 5]). Plate 7, figures 10, 11 Remarks—A few specimens from several pre- Moose River Group formations possess internal fea- tures that suggest assignment of the specimens to Leptostrophia. Cardinal process lobes well developed, flanked laterally by prominent socket plates. Hinge line at least partly denticulate. Pedicle valve muscu- lature well impressed, bounding ridges subparallel anteriorly. None of this material well enough pre- served to be specifically or generically assigned with confidence. Occurrence.—USGS locs. 3477—SD. 3499-SD, 3483—SD, Spencer Lake and Pierce Pond quad- rangles, Somerset County, Maine. Stratigraphic location.—Hardwood Mountain For— mation (Upper Silurian), Parker Bog Formation (Lower Devonian), and Beck Pond Limestone (Low- er Devonian). Figured specimeu3.—USNM 126120, 126147. Leplostropllia 5]). or Protoleplostrophia :1). Plate 7, figures 12, 13 Discussiou.——Leptostrophiid pedicle valves are abundant in the Lower Devonian formations, but, unfortunately, diagnostic brachial valves are rela- tively uncommon. In my opinion it is not possible to discriminate pedicle valves of Leptostrophia from those of Protoleptostrophia. The diagnostic charac- ters of these genera are discussed under Protolepto- straphia. Occurrence.—USGS 2861—SD, 2832—SD, 2819—SD, 2890—SD, 2725—SD, 2761—SD, 27 08—SD, 2727—SD, locs. 2760—SD, 2883—SD, 2864—SD, 2877—SD, 2870—SD, 2711—SD, 2698—SD, 2749—SD, 2793—SD, 2722—SD, 2785—SD, 2831—SD, 2823—SD, 2847—SD, 2860—SD, 2834—SD, 2777—SD, 2‘697—SD, 2775—SD, 2858—SD, Somerset and Piscataquis Counties, Maine. Stratigraphic locatiou.——McKenney Ponds Mem- ber, upper and lower sandstones of Tarratine For- mation (Lower Devonian), and Seboomook Forma- tion (Lower Devonian), Kineo Rhyolite (Lower Devonian). Figured specimens.—USNM 126118. Genus PROTOLEPTOSTROPHIA Caster, 1939 Comparisou.——The discrimination of Leptostro- phia and Protoleptostrophia is based primarily (Wil- liams, 1953, p. 41) on the absence of socket plates and a prominent chilidium in the latter; and their presence in former. No reliable criteria for discrimi- nating between pedicle valves of these two closely related genera have yet been determined. 2857—SD, 3092—SD, 3088—SD, 2691—SD, 2795—SD, Protoleptoslrophin cf. P. blainvillei (Billings, 1874) Plate 8, figures 1—8 Exterior—Pedicle valve gently convex, brachial valve flat. Maximum width at straight hinge line. Lateral and anterior margins rounded. Anterior commissure rectimarginate. Pedicle valve interarea apsacline, brachial valve interarea anacline. Shell outline transversely elliptical. Fine costellae crossed by concentric growth line and subdued wrinkles. Pedicle valve interior—Cardinal process pits well defined, medially separated by short ventral process. Hinge line wholly denticulate. Lateral ridges bound flabellate muscle field, originate lateral to cardinal process pits, diverge at angle of about 90° Umbonal cavities pustulose. Muscle field large, flabellate, ex- tends to about midlength or a position slightly anterior and about half as wide as maximum width. Muscle impressions consist of pair small, elliptical adductor impressions located just anterior of ven- tral process and large, flabellate diductor field di- vided into radial sectors by number of short myo- phragms. Low ridge originates posteriorly at base of ventral process, bisects muscle field medially. Peripheral regions of interior crenulated by impress of external ornamentation. SYSTEMATIC PALEONTOLOGY 23 Brachial valve interior.—Cardinalia consist of stout cardinal process lobes joined anteriorly by broad notothyrial platform, laterally by lateral ridges. Notothyrial platform narrows anteriorly to median ridge which bisects muscle field. Lateral ridges diverge at angle about 90°, bound deeply im- pressed muscle field. Muscle field consists of anterior- ly expanding adductor impressions restricted to pos- terior third of valve. Umbonal regions very pustu- lose, remainder of valve crenulated by impress external ornamentation. Comparison.——Maine specimens are similar to P. blainvillei, but adequate comparative material is not available to make positive specific identification. Occurrence.——USGS locs. 2730—SD, 2750—SD, 2814—SD, 2820—SD, 2839-SD, 2840—SD, 2852—SD, 2873—SD, Brassua Lake quadrangle, Somerset Coun- ty, Maine. Stratigraphic location.—Tomhegan Formation (Lower Devonian). Figured specimens.—USNM 126140, 126137, 126154, 126136. Unfigured specimens.——USNM 127384, 127393— 127395. Prololeptostropllin :p. Plate 8, figures 9—13 Exterior—Pedicle valve gently convex, brachial valve flat. Maximum width at straight hinge line. Outline subcircular to transversely elliptical. Anteri- or and lateral margins rounded. Anterior commis- sure rectimarginate. Pedicle valve interarea apsa- cline, relatively long; brachial valve interarea ana- cline, relatively short. Fine costellae crossed by con- centric growth lines and weak wrinkles. Pedicle valve interior.—Cardinal process pits deeply excavated, medially separated by short ven- tral process. Lateral ridges bound cardinal process pits laterally and laterally border flabellate muscle field. Narrow ridge extends from ventral process to muscle field front. Paired, elongate, small posteriorly situated adductor impressions and large, flabellate diductor impressions divided into radial sectors by short myophragms. Muscle field of variable length; may extend three-quarters distance to anterior mar- gin and half valve width. Umbonal regions strongly pustulose. Valve peripheral portions crenulated by external ornamentation impress. Hinge line wholly denticulate. Brachial valve interior—Paired cardinal process lobes anteriorly bounded by broad, flat, notothyrial platform which narrows anteriorly into median ridge. Cardinal process lobes laterally bounded by diverging lateral ridges which bound muscle field and diverge at angle of about 90° or more. Umbonal region strongly pustulose, remainder of valve crenu- lated by external ornamentation impress. Muscle field deeply impressed, consists of pair of anteriorly expanding adductor impressions restricted to pos- terior half of valve. Comparison.-——Maine material is too poorly pre- served to be specifically identified. Occurrence.———USGS locs. 2824—SD. 2699—SD, 2696—SD, 2719—SD, 2721—SD, 3090—SD, 2810—SD, 2767-SD, 2729—SD, 2770—SD, 2720—SD, 2718—SD, 2771—SD, 2701—SD, 2731—SD, Somerset County, Maine. Stratigraphic location—Lower sandstone of the Tarratine Formation (Lower Devonian). Figured specimens.—USNM 126132-126135. Genus STROPHONELLA Hall, 1879 Strophonella cf. 5. punclulifen (Conrad, 1838) Plate 8, figures 14—18 Exterior—Pedicle valve gently concave, brachial valve gently convex in large specimens; reverse con- dition true in small specimens, that is resupinate. Maximum width at straight hinge line. Pedicle valve interarea apsacline, brachial valve interarea steeply anacline. Interarea both valves moderate length. Lateral and anterior margins rounded. Costellae crossed by concentric growth lines. Small pseudo- deltidium appears to be present at delthyrium apex. Brachial valve chilidium not certain. Anterior com- missure rectimarginate. Pedicle valve interior—Hinge line denticulate about half its length. Small cardinal process pits lo- cated either side of short ventral process. Muscle field not discernible on specimen studied. Interior crenulated by external ornamentation impress. Brachial valve interior—Cardinal process lobes flanked laterally by narrow socket plates which par- allel hinge line. Short notothyrial platform extends anteriorly from cardinal process lobes and narrows rapidly. Very short lateral ridges bound muscle field. Muscle field restricted to valve posterior, consists of subcircular, small adductor impressions. Valve in- terior very pustulose except periphery which is crenulated by external ornamentation impress. Comparison—Strongly resembles stronhonellids from Lower Devonian like S. punctulifera and S. ampla. Pedicle valve musculature is not well marked in the petaloid fashion characteristic of many stro- phonellids, but this may be due to the small size of the specimen studied. Dental plates are absent on the only pedicle interior studied. Occurrence.—USGS loc. 3477—SD, Pierce Pond quadrangle, Somerset County, Maine. 24 EARLY PALEOZOIC BRACHIOPODS OF THE MOOSE RIVER SYNCLINORIUM, MAINE Stratigraphic location—Parker Bog Formation (Lower Devonian). Figured specimens.——USNM 126143. Unfigured specimens.—USNM 126161. Superfunily DAVIDSONIACEA Funny SCHUCHERTELLIDAE Subfamily SCHUCHERTELLINAE Genus SCHUCHERTELLA Girty, 1904 “Schuchertella” becnflensis (Clarke, 1900) Plate 9, figures 1—11 Orthotetes becraftensis (Clarke, 1900), p. 51—52, pl. 7, figs. 15—28. Exterior.—Unequally biconvex shells, brachial valve gently convex, pedicle valve subconical. Great- est width near midlength. Hinge line straight. Pedi- cle valve interarea steeply apsacline to catacline, brachial valve interarea steeply anacline. Delthyrium closed by convex pseudodeltidium; apical portion notothyrium bounded by ribbonlike chilidium. Pedi- cle valve interarea relatively long, brachial valve interarea short. Shells subsemicircular in outline. Closely spaced costellae, rounded in cross section, separated by r0unded interspaces. Costellae originate by implantation and bifurcation. Lateral and anteri- or margins rounded. Anterior commissure rectimar- ginate, crenulate. Pedicle valve interior—Stout hinge teeth border sides of delthyrium, supported basally by own tracks. Track may be anteriorly supported by lamellar de- posit of secondary material which simulates dental lamella. Muscle field semicircular, poorly impressed; consists of bilobate field restricted to posterior third of valve. Peripheral regions of interior crenulated by impress of external ornamentation, remainder smooth due to deposition of secondary material. Brachial valve interior.—Bilobed cardinal process, posteriorly extends behind hinge line, fused anterior- ly to laterally diverging buttress plates which include angle of about 90°. Dental sockets excavated between interarea and socket plates. Adductor muscle field lightly impressed, extends anteriorly to about mid- length. Interior crenulated by external ornamenta- tion impress. Comparison—Maine material appears to be “S.” becraftensis. Generic assignment not certain because of no modern consideration of taxonomy and morph- ology of Silurian-Devonian orthotetaceans. Occurrence.—USGS locs. 2721—SD, 2719—SD, 2720—SD, 2810-SD, 3091—SD, 2722—SD, 2806-SD, 2777—SD, 2767—SD, 2761—SD, Somerset County, Maine. Stratigraphic location.—McKenney Ponds Mem- ber of the Tarratine Formation, lower sandstone 126119, 126157, of the Tarratine Formation, and Seboomook Formation. Figured specimens.—USNM 126139, 126124, 126138, 126160, 126162, 126151. Unfigured specimens.——-USNM 127367, 127369, 127371. 126156, "Scllucllerlella” 5]). Plate 9, figures 12—14 Brachial valve exterior.—Brachial valve evenly convex. Maximum width is slightly posterior of mid- length. Straight hinge line, interarea steeply ana- cline, short. Ribbonlike chilidium borders notothyri- um. Costellae originate by bifurcation and implanta- tion. Lateral and anterior margins rounded. Anteri— or commissure rectimarginate, crenulate. Brachial valve interior.—Posteriorly bilobed car- dinal process anteriorly fused with buttress plates, anteromedially joined by short, rounded septumlike ridge. Socket plates diverge at angle of about 90°. Muscle field impressed, consists of elongate, elliptical adductor impressions medially divided by median ridge. Muscle field restricted to posterior third of valve. Comparison.-—Maine material is too fragmentary to be specifically identified. Occurrence.——USGS locs. 2814—SD, 2820—SD, 2750—SD, 2730-SD, Brassua Lake quadrangle, Som- erset County, Maine. Stratigraphic location.—Tomhegan Formation. Figured specimens.—USNM 126148. Unfigured specimens.—-—USNM 127362—127366, 127368, 127370, 127372, 127373. Genus HIPPARIONYX Vanuxem, 1842 Hipparionyx 51). I Plate 9, figures 15—17 Exterior.—Unequally biconvex shells with highly convex brachial valve, almost flat pedicle valve. Shell outlines sulbcircular. Greatest width near midlength, hinge line very narrow. Lateral, posterior, and an- terior margins rounded. Anterior commissure recti- marginate, crenulate. Rounded costellae. Pedicle valve interior—Muscle field flabellate, ex- tends about two—thirds distance to anterior margin and one-half distance to lateral margins. Posterior portion muscle field bounded by lateral ridges origi- nating posteriorly at base of hinge teeth. Muscle field tripartite, consists of posterior, medially situ- ated, elongate adduotor impression anterior of which are paired, elliptical diduotor impressions medially separated by low ridge. Peripheral region of valve crenulated by external ornamentation impress. Brachial valve interior.—Posteriorly bifid cardinal process anteriorly fused with laterally diverging SYSTEMATIC PALEONTOLOGY 25 buttress plates. Myophragm situated at anterior edge of cardinal process. Muscle field impression not discernible. Comparisou.—Maine material is too poorly pre- served to be specifically identifiable. Occurrence.—USGS locs. 2806—SD, 2810—SD, Pierce Pond quadrangle, Somerset County, Maine. Stratigraphic locatiou.——McKenney Ponds mem- ber of the T‘arratine Formation (LOWer Devonian). Figured specimens.—USNM 126150, 126149. Unfigured specimens.—USNM 127374, 127375. Family STROPHOMENIDAE Subfamily LEPTAENOIDEINAE Genus LEPTAENISCA Beecher, 1890 Leptaenisca sp. Plate 9, figures 18—20 Pedicle valve exterior.—Pedicle valve highly con- vex. Valve outline asymmetrical. Broad cicatrix which may have been formed by attachment against the stem of a crinoid present on one specimen. Radi- ating coarse and fine costellae crossed by concentric growth lamellae. Lateral and anterior margins ir- regularly rounded. Maximum width is slightly an- terior of midlength. Shell transversely elliptical. Hinge line relatively straight. Anterior commissure rectimarginate. Pedicle valve interior—Muscle field strongly im- pressed, consists of paired, elongate scars restricted to um'bonal region. Rounded ridge medially separates muscle scars. Muscle field laterally bounded by ridge of secondary material. Umbonal cavities pustulose, due to presence of pseudopunctae. Comparison—Maine specimens are not well enough preserved to be identified specifically. How- ever, they do resemble L. australis Kindle. Occurrence.—-—USGS loc. 3499-SD, Spencer Lake quadrangle, Somerset County, Maine. Stratigraphic locatiom—Beck Pond Limestone (Lower Devonian). Figured specimens.—USNM 126128, 126129. Suborder CHONETOIDEA Superfamily CHONETACEA Family ANOPLHDAE Subfamily ANOPLIINAE Genus CYRTONISCUS Boucot and Harper, 1968 Cyrtoniscus nectus (Clark, 1907) Plate 9, figures 21—37 Cgrtoniscus nectus (Clarke, 1907), see Boucot and Harper, 1968, p. 172—173, pl. 30, figs. 6—13. Exterior.——Pedicle valve strongly convex posteri- orly, changes convexity abruptly at length of about 2 mm. Small shells, posterior part of larger shells subhemispherical. Shell outline circular to trans- versely elliptical (deformation greatly distorted out- line of specimens from Maine). Hinge line straight, maximum width located near midlength. Brachial valve gently concave, does not show abrupt change in convexity from posterior to anterior as opposed to shape of pedicle valve. Interareas virtually unde- veloped except in medial part of pedicle valve where anacline. Delthyrial and notothyrial cavities appear unmodified. Anterior margin rectimarginate, crenu- late. Anterior and lateral margins rounded. Radial costae, crossed by fine concentric lines, costae in- creasing anteriorly by bifurcation and implantation. The relation between length and width shown in figure 7. One pair of spines extends laterally from ears; from two to ten spines occur along posterior margin of pedicle valve. Brachial valve interior.—Posteriorly directed, ter- minally bifid cardinal process, which seals off delthy- rium and laterally directed socket ridges. Dental sockets formed medially by edge of cardinal process, anteriorly by socket ridges, and posteriorly by poor- ly developed interarea. Median septum not present although cardinal process continues anteriorly for short distance as notothyrial platform. Accessory septa radiate from cardinalia, reach almost to anteri- or margin. Anderidia minute. Interior highly pustu- lose. External ornamentation impress general. Pedicle valve interior—Stubby hinge teeth con- tinuous with reduced interarea, unsupported by den- tal plates. Narrow median septum divides muscle field, extends anteriorly to position where change in shell convexity takes place. Hollow spine bases pres- ent along hinge line. External ornamentation impres- sion present near anterior margin. Measurements.—Relationship between length and width of pedicle valve (fig. 7) is linear, with a high degree of dispersion which is probably a function of mechanical deformation. Occurrence.—USGS locs. 2723—SD, 2750—SD, 2752—SD, 2814—SD, 2820-SD, 2842—SD, 2852-SD, 2873—SD, Brassua Lake quadrangle, Somerset Coun- ty, Maine. Stratigraphic locatiou.—Tomhegan Formation (Lower Devonian) . Figured specimens.—USNM 126222, 126223, 126227—126229, 126231, 126247, 126248, 127391, 127392. Unfigured specimens.—USNM 127361, 127396. Funily CHONETIDAE Subfamily CHONETINAE Genus ECCENTRICOSTA Berdan, 1963 Eccentricosta :1). Plate 10, figures 1—7 Exterior—Shells concavo-convex in cross section, pedicle valve genrtly convex. Maximum width proba- bly near hinge line. lateral and anterior margins 26 EARLY PALEOZOIC BRACHIOPODS OF THE MOOSE RIVER SYNCLINORIUM, MAINE 20 .t' '. ' _..." ' . “4|... -- . \__..._:.. . . _. “Aw- LENGTH, IN MILLIMETERS 8 ' M o I . I I 2"”? (":5 0 10 20 WIDTH, IN MILLIMETERS FIGURE 7.—Comparison of length versus width for pedicle valves of Cyrtonis- cus nectus (Clarke, 1907). Tomehegan Formation, USGS loc. 2852—SD, Bras- sua Lake quadrangle, Somerset County, Maine evenly rounded and crenulate. Pedicle valve inter- area apsacline, planar. Deltidium fills delthyrium. Deltidium very convex posteriorly. Brachial valve interarea not observed; presumably very narrow. Both valves transverse in outline. Relatively coarse, undulating costellae originate by implantation pe- ripherally and along posterior margin posteriorly. About five or six short spines on each side of pedicle valve midline. Interior.—Internal features poorly preserved on available specimens; little noted except presence of short, thin septum in interior of pedicle valve beak. Remarks.—Berdan (1963, p. 254) noted that cos- tellae originate by bifurcation; I would prefer im- plantation and bifurcation, as few defined junctions were observed in material illustrated by Berdan or in specimens from Maine. Maine specimens are too poorly preserved to be identified specifically. Occurrence.—USGS locs. 4841—SD, 5583—SD, 5586—SD, 5995—SD, Attean quadrangle, Somerset County, Maine. Stratigraphic location.—Hardwood Mountain For- mation (Upper Silurian). Distribution—The genus is known from the cen- tral Appalachians (Berdan 1963, p. 256) , and I have observed it in Late Silurian collection-s from western Gaspé, the Fish River Lake quadrangle in Maine, and near Marbleton in the Eastern Townships of Quebec. Figured specimens.—USNM 160121, 160126. Genus DAWSONELLOIDES Boncot and Harper, 1968 Dawsonelloidel canldensis (Billings, 1874) Plate 10, figures 8—12; plate 11, figures 1—7 Dawsonelloides canadensis (Billings, 1874), see Boucot and Harper, 1968, p. 166—167, pl. 28, figs. 17—21; pl. 29, figs. 1—3. Exterior—Pedicle valve gently convex, brachial valve gently concave. Maximum width located at straight hinge line. Shell subsemicircular to quad- rate in outline. Posterior portion lateral margin straight, anterior portion and lateral margin round- ed. Pedicle valve interarea apsacline, brachial valve interarea steeply anacline to hypercline. Interareas both valves moderately long. Convex pseudodeltidi- um occupies apex of delthyrium. Discrete chilidial plates border notothyrium. Hollow spines border lower edge of pedicle valve interarea. Number of spines variable, does not exceed ten in specimens studied. Anterior margin rectimarginate. Fine cos- tellae interrupted by concentric growth lines, give reticulate appearance. Pedicle valve interior—Hinge teeth stout, unsup- ported by dental lamellae. D‘istally and medially hinge teeth deeply grooved to receive socket ridges of brachial valve for articulatory purposes. Median septum extends to about midlength anteriorly. Medi- an septum low in delthyrial cavity, but rises rapidly to form winglike blade, descends abruptly at about one-quarter distance to anterior margin, then ex- tends to about midlength as low septum. Delthyrial cavity partly filled by secondary shell material. Mus- cle bounding ridges laterally border flabellate di- ductor field. Muscle field about half width of shell, extends anteriorly to about midlength. Interior of valve very pustulose, due to projecting pseudopunc- tae. Radiating from posteromedial portion of shell are ridges separated by broad, rounded interspaces. Ridges bifurcate anteriorly. Brachial valve interior—Posterior facing cardi- nal process consisting of two cardinal process lobes fused medially in large specimens. Short chilidial plates occur posterolaterally of cardinal process lobes. Anteriorly cardinal process lobes join median septum which extends to about midlength. Lateral to cardinal process lobes are socket plates which curve and extend parallel to hinge line. Pair of anderidia in posteromedial portion of interior laterally enclose paired adductor field. Anderidia include angle of about 60°. Interior very pustulose, Ridges radiate from umbonal region, separated by broad inter- spaces. Ridges bifurcate peripherally. Measuremcuts—Relationship between length and width of pedicle valves appears linear (fig. 8), but small size of sample precludes definite conclusion. SYSTEMATIC PALEONTOLOGY 27 50 40 3O 20 WIDTH, IN MILLIMETERS 10 O 10 2O 30 LENGTH, IN MILLIMETERS FIGURE 8.—Comparison of length versus width of pedicle valves of Dawsonelloides canadensis (Billings, 1874). Lower sandstone of Tarratine Formation, USGC 10c. 2701—SD, Brassua Lake quadrangle, Somerset County, Maine. Occurrence.——USGS 2711—SD, 2705-SD, 2704—SD, 2708—SD, 2861—SD, 2872—SD, 2801—SD, 2696—SD, 2700—SD, 2812—SD, 2766—SD, 2813—SD, 3486—SD, 2857-SD, 3090—SD, 2770—SD, 2767—SD, 2777—SD, 2803—SD, 2799—SD, 2807—SD, 2771—SD, 2830~SD, 3482-SD, 2890—SD, 2706—SD, Piscataquis and Somerset Counties, Maine. Stratigraphic location.—Seboomook Formation (Lower Devonian), and sandstones of the Tarratine Formation. Figured specimens.——USNM 126250, 126245, 126253, 126239, 126251, 126244, 126252, 126246, 126233. locs. 2701-SD, 2699—SD, 2743-SD, 2860—SD, 2740—SD, 2849—SD, 2797-SD, 2710—SD, 2703-SD, 2725-SD, 2847—SD, 2776—SD, 3089—SD, 2783—SD, 2802—SD, 3476—SD, Unfigured specimens.—USNM 126249. Measured specimens.—USNM 126705. Genus EODEVONARIA Breger, 1906 Eodevonnria arcunta (Hall, 1857) Plate 11, figures 8—16 Eodevonaria arcuam (Hall, 1857), see Boucot and Harper, 1968, p. 153—156, pl. 27, figs. 1—7. Exterior—The pedicle valve is highly convex and the brachial valve is gently concave. Body cavity nar- row. Hinge line straight, maximum Width located at hinge line. Shell outline subcircular to elliptical (de- formation has changed outline of many specimens). Ears flattened. Pedicle valve interarea orthocline to gently apsacline, brachial valve interarea hypercline. Delthyrium completely closed by convex pseudodelti- dium, notothyrium closed by convex chilidium. Pedi- cle valve interarea about three times as long as that of brachial valve. Anterior margin rectimarginate. Spines originate at base of pedicle valve interarea, but very rare and seldom observed. Hinge line den- ticulation visible at junction of two interareas. Cos- tellae originate largely by bifurcation. Costellae un- dulate gently, do not increase in width anteriorly. Costellae low, rounded in cross section, separate-d by interspaces of similar form. Number of spines along posterior margin of pedicle valve has not been ob- served to exceed eight. Pedicle valve interven—Two stubby, laterally di- rected hinge teeth situated anterior of delthyrial cavity and hinge line, just bordering edges of del- thyrium. LOW median septum extends from delthyri- al cavity to about midlength. Hollow spine bases rarely seen along junction of interarea and bottom of valve. Two posterolateral ridges diverge from mid- line and interlock with corresponding anterolaterally directed posterior ridges of brachial valve. Interior of ears pustulose, remainder of interior bears im- press of external ornamentation in addition to being pustulose. Muscle field very weakly impressed, con- sists of two pairs of elongate impressions paralleling median septum, plus small pair posteriorly located, oval impressions. Brachial valve interior—Fused, prostrate, pos- teriorly directed cardinal process lobes. Lateral to base of cardinal process lobes are anterolaterally directed socket ridges. Low platform of secondary material, anterior of fused cardinalia, trifurcates anterior‘ly into median septum extending to mid- length and into two anderidia which enclose muscle field. Near anterior portion of median septum may be developed pair of accessory septa. Valve interior pustulose posteriorly, peripherally cren‘ulated by im- press of external ornamentation. Muscle field con- 28 EARLY PALEOZOIC BRACHIOPODS OF THE MOOSE RIVER SYNCLINORIUM, MAINE sists of paired, elongate adductor impressions which parallel median septum. Measurements.——The relationship between length and width of pedicle valves is essentially random with a high degree of dispersion (fig. 9). This condi- tion reflects the fact that the specimens were se- lected from a very small size range and that they have been subjected to a high degree of mechanical deformation. 20 (I) I UJ ‘— w . 2 .. ’ __J '-'.' t i 'u t 2 1o z ' -‘ I. '— o z I.“ .J 00 10 20 WIDTH, IN MILLIMETERS FIGURE 9.—Comparison of length versus width of pedicle valves of Eodevonaria arcuata (Hall, 1857). Tomhegan For- mation, USGS loc. 2820-SD, Brassua Lake, quadrangle, Somerset County, Maine. Occurrence—USGS locs. 2820—SD, 2750—SD, 2852—SD, 2814—SD, 2839—SD, 2873-SD, 2840—SD, Brassua Lake quadrangle, Somerset County, Maine. Stratigraphic location—Tomhegan Formation. Distribution—Eastern North America. Figured specimens.—USNM 126230, 126235, 126232, 126238, 126226. Unfigured specimens.—USNM 126224. Measured specimens.—126677A—126677AN. Subfumily CHONOSTROPHIELLINAE Genus CHONOSTROPHIELLA Boucot and Amsden, 1964 Chonostrophiella complaints (Hall, 1857) Plate 11, figures 17—25; plate 12, figures 1, 2 Chonostrophiella complanata (Hall, 1857), see Boucot and Amsden, 1964, p. 881—882, pl. 141, figs. 1—11. Exterior.—Pedicle valve flat or gently concave ex- cept in strongly concave peripheral region. Brachial valve gently convex. Maximum width located on straight hinge line. Pedicle valve interarea catacline to steeply apsacline, brachial valve interarea steeply anacline. Pedicle valve interarea relatively long, brachial valve interarea relatively short. Delthyrium 126234, usually open, includes angle of about 90° except in few specimens that have a small pseudodeltidium located apically. Notothyrium modified laterally by small chilidial plates which extend medially almost to midline but have not been observed in conjunct state. Fine, radial costellae originate by both bifur- cation and implantation, crossed by closely spaced concentric ornamentation that results in granulose effect. Shell outline transversely elliptical. Posterior part of lateral margin straight but anterior part and anterior margin gently rounded. Anterior commis- sure rectimarginate. Base of pedicle valve interarea bears short spines (three to seven on each side of midline). Pedicle valve interior.—Very short dental lamellae support spatulate hinge teeth. Dental lamellae di- verge at angle between about 90° and 120°. Dental lamellae enclose posterior part of poorly impressed muscle field. Muscle impressions posterolaterally delineated by bounding ridges which extend from base of dental lamellae. Diduetors compose bulk of muscle field; adductor impressions scarcely recog- nizable, restricted to posterior part of muscle field. Adductor impressions elongate, situated just anteri- or of delthyrial cavity. Muscle field divided medially by anteriorly bladelike septum with triangular cross section posteriorly. Median septum extends anterior- ly to about midlength, muscle field about half as wide as maximum width. Hollow spine bases origi- nate at base of palintrope, extend medially until reaching exterior where they diverge laterally. Re- gion posterolateral to muscle field pustulose. Exter- nal ornamentation impression present peripherally. Brachial valve interior—Cardinal process lobes, striate both posteriorly and medially, bounded pos- teriorly by chilidial plates, anteriorly fused with buttress plates. Muscle impressions consist of two pairs of adductors, one posterolateral and other an- teromedial in position, separated from each other by low, trifid, platform of secondary material. Dental sockets not crenulated, laterally directed, bounded by interarea and short socket ridges. Costellae im- pressed on interior, become more pronounced periph- erally. Muscle field restricted to posterior quarter of valve. Measurements.—Relation between length and width of pedicle valves is linear with a relatively low degree of dispersion (fig. 10). Comparison—C. dawsomi may be junior synonym of C. complanata but adequate material of former is not available for comparison. Adequate comparative material of other species of Chonostmphiella is not available. Occurrence—USGS locs. 2718—SD, 2720—SD, SYSTEMATIC PALEONTOLOGY 29 30 20 10 LENGTH, IN MILLIMETERS 20 30 40 WIDTH, IN MILLIMETERS FIGURE 10.—Comparison of length versus width of pedicle valves of Chance- trophiella complanata (Hall, 1857). Lower sandstone of Tarratine Formation, USGS loc. 2718-SD, Long Pond quadrangle, Somerset County, Maine. 2767-SD, 2719—SD, 2691—SD, 2721—SD, 2722—SD, 3090-SD, 2796—S-D, 2727—SD, 2777—SD, 2766—SD, 3088—SD, 2824—SD, 2862—SD, 3482—SD, 2783—SD, 2861-SD, 2890-SD, Somerset and Piscataquis Coun- ties, Maine. Stratigraphic location.——Seboomook Formation (Lower Devonian), and upper and lower sandstones of the Tarratine Formation (Lower Devonian). Distribution—Eastern North America. Figured specimens.——USNM 126254, 126237, 126240, 126241, 126236, 126225, 126243, 126242. Suborder RHYNCHONELLOIDEA Superfnmily RHYNCHONELLACEA Family TRIGONIRHYNCHHDAE Genus CUPULAROSTRUM Sartenaer, 1961 Cupulnrostrum macrocosta Boucot, n. sp. Plate 12, figures 3—11‘ Exterior.—Unequally biconvex shells with brachi- al valve having greater convexity. Cardinal margin terebratulid, pedicle foramen submesothyrid in posi- tion. Hinge line very short, curved in anterolateral direction. Deltidial plates conjunct. Anterior com- missure uniplicate, strongly crenvulate. Pedicle valve beak straight. Moderately sized ribs with four to five in sulcus of pedicle valve, seven to nine on either flank. Brachial valve bears eight to ten ribs either side of median line. Shell outline pentagonal to subcircular with greatest width at about mid- length. Lateral margins rounded, anterior margin greatly rounded. Shells large, many reaching length of 3 cm. Pedicle valve inten’or.—Dental lamellae short, sup- port short, bladelike, laterally directed hinge teeth. Muscle field impression barely discernible except in large specimens. Paired diductor scars enclose medi- an, diamond-shaped adductor scar which narrows anteriorly at about one-third distance to anterior margin. Well-developed pedicle callist located at rear of delthyrial cavity. Muscle area no more than one- third of maximum width, commonly somewhat less. Interior strongly plicated by impress of external ornamentation. Brachial valve interior.—Discrete hinge plates slightly overlap lateral sides of small septalium, lat- ter supported by median septum which extends an- teriorly about one-half length of valve. Laterally directed dental sockets crenulated by about seven to ten ridges distinctly impressed on anterior and bot- tom of sockets. Muscle field not distinctly impressed. Interior strongly plicated by impress of external ornamentation. Comparison—Externally C. macvocosta resembles Billings’ (1874) “Rhynchonella” excellens and “R.” dryape, both from Gaspé limestone N o. 3. “R.” dry- ope has only three costae in sulcus, whereas “R.” excellens has much finer costae than C. macrocosta. 30 EARLY PALEOZOIC BRAC'HIOPODS OF THE MOOSE RIVER SYNCLINORIUM, MAINE C. macrocosta reaches much larger size than the type species C. recticostatum Sartenaer, 1961, which only has three and sometimes four costae in sulcus. Discussion.—-Hall and Clarke’s (1894) definition of this genus (as Camarotoechia) makes necessary a complete acquaintance with the structure of cardi- nalia before a species can be generically assigned. External form and ornamentation are not diagnos- tic; true Cupularostram has septalium commonly roofed over by a perforate hinge plate, also crenulate dental sockets. Occurrence.—USGS locs. 2750—SD, 2814—SD, 2852—SD, 2820—SD, 2840—SD, 2842—SD, B‘rassua Lake quadrangle, Somerset County, Maine. Stratigraphic location.—Tomhegan Formation (Lower Devonian). Holotype.—USNM 125851. Figured specimens—USNM 125852, 125855, 125857, 125904. Cupularostrum? sp. Plate 12, figures 12—21 Exterior.—Unequally biconvex shells, brachial valve having greater convexity. Shell outline sub- circular, greatest width near midlength. Cardinal margins terebratulid, beak suberect. Hinge line very short, curved in anterolateral direction. Anterior commissure uniplicate, crenulate. Brachial valve bears well—developed fold, pedicle valve has corre- sponding sulcus. Costae, about three in sulcus, four on fold, five on each flank. Brachial valve interior—Serial sections show sep- talium covered with hinge plate anterior of small apical foramen. Median septum supports septalium. Interior strongly plicated by external ornamenta- tion. Nature of dental sockets not elucidated by serial sections. Pedicle valve interior—Small dental lamellae situated either side weakly impressed muscle field which is situated in delthyrial cavity. Interior strongly plicated by impress of external ornamen- tation. Occurrence.—USGS 10c. 3601—SD, Spencer Lake quadrangle, Somerset County, Maine. Stratigraphic location—Beck Pond Limestone (Lower Devonian). ‘Figiired specimens—USN M 125862, 125859. Unfigiired specimens—USNM 125860. Genus ANCILLOTOECHIA Havlicek, 1959 Ancilloloechia 1]). Plate 12, figures 22—26 Discussion.——A few poorly preserved specimens from Maine possess external form and ornamenta- tion of Ancillotoechia. The material is too poor to be specifically assigned. During the study of Early Devonian rhynchonellids from Maine, here assigned to Ancillotoechia, the close similarity of external features of the rhynchonellid species bidentata Hi- singer, bialveata Hall, and acinum Hall was investi- gated. Hall (1863, p. 215) drew attention to these similarities by comparing the three species. Serial sectioning of Amsden’s species Ancillotoechia hara- ganensis (close to and possibly synonymous with A. bialveata (Hall)), “Rhyrichonella.” bidentata, and “R.” acinus shows that haraganensis belongs to Ancillotoechia (fig. 11) ; that bidentata belongs to an undescribed homeomorphic genus (fig. 13) ; and that acinus belongs to Diabolirhynchia Drot, 1964 (fig. 12). Occurence.——USGS loc. 3499-SD, Spencer Lake quadrangle, Somerset County, Maine. Stratigraphic location—Beck Pond Limestone (Lower Devonian). Figured specimen—USNM 125902. Unfigured specimen—USNM 125901. “Ancillotoecllil” cf. "A." allisulcntn (Amsden, 1951) Plate 12, figures 27—31 Exterior.—Unequally biconvex shells having gently convex pedicle valve, sulbpyramidal brachial valve. Brachial valve highly convex, made subpyra- midal by fold which extends upwards from rest of valve, near midlength, at angle of about 30°. Pedicle valve has broad, low sulcus and very high tongue @ 13.4 @ 13.3 @ 13.1 13.0 13.3 0 13‘ 12.9 FIGURE 11.—Serial sections of three specimens of Ancilla- toechia haraganerisis (Amsden) USNM 125858, X 4.5. Haragan Shale, White Mound, 3 miles west of Nebo, Okla. Numbers are measurements in millimeters from posterior end of shell. SYSTEMATIC PALEONTOLOGY 31 10.15 10.65 10.4 @ 10.1 10.0 Q“? dd FIGURE 12.———Serial sections of Diabolirhynchia acinus (Hall, 1863), X 4.5, USNM 126049. Waldron Shale, Waldron, Indiana. Numbers are measurements in milli- meters from posterior end of shell. anteriorly corresponding to high fold in brachial valve. Shells pentagonal in outline, greatest width near midlength. Lateral margins rounded into very gently rounded anterior margin. Anterior margin uniplicate, strongly crenulate. Pedicle valve beak incurved, terebratulid. Fold bears about four angu- lar costae, sulcus contains about three. Each flank bears about seven costae. Neither fold nor sulcus developed in umbonal regions of shell. Concentric growth lines present on valves. Pedicle valve interior—Serial sections (fig. 14) show valve possesses short dental lamellae border- ing delthyrial cavity and muscle fields poorly im- pressed. Brachial valve interior—Septalium formed by median septum that bifurcates into laterally diverg- ing branches posteriorly. Borders of septalium over- lapped by discrete hinge plates. Dental sockets roofed over posteriorly by outer edges of binge plates. Muscle field poorly impressed. Occurrence.—USGS 10c. 3488—SD, Spencer Lake quadrangle, Somerset County, Maine. Stratigraphic location—Hardwood Mountain For- mation (Upper Silurian). Figured specimens—USNM 125854. Family UNCINULIDAE Genus SPHAERIRHYNCHIA Cooper and Muir-Wood, 1951 Splllerirllyncllia 3p. 1 Plate 12, figures 32—36; plate 13, figures 26—28 Exterior.——Unequally biconvex shells with brach- ial valve about twice convexity of pedicle valve. Shell outline subcircular to longitudinally elliptical. Anterior and lateral margins rounded, greatest width near midlength. Brachial valve bears broad, low fold, pedicle valve corresponding sulcus. Pedicle valve has long tongue anteriorly which extends up to join fold on brachial valve. Anterior margin uni- plicate, crenulate. Cardinal margin terebratulid in form, beak of pedicle valve appears incurved. Small, rounded costae, about ten costae in fold and sulcus, about ten costae each flank. Pedicle valve interior—Short dental lamellae lo- cated either side of delthyrial cavity. Serial sections indicate muscle field not very deeply impressed. 3 Q . @ 8.75 696 @ WQ <33 83 a a o 9.1 8.6 FIGURE 13.—Serial sections of “Rhynahonella” bidentata (Hisinger, 1826), X 4.5, USNM 125880. Silurian beds near Klinteham, Gotland, Sweden. Numbers are measure- ments in millimeters from posterior end of beak. Brachial valve interior.—Septalium supported by median septum. Discrete hinge plates project over medial edge of cruralivum. Lateral edges of hinge plates project over medial edge of dental sockets. Comparison—Maine material is too scanty for specific identification. Occurrence.—USGS 10c. 3470—SD, 3469—SD, Spencer Lake quadrangle, locality 5586—SD, Attean quadrangle, Somerset County, Maine. Stratigraphic location—Hardwood Mountain For- mation (Upper Silurian). Figured specimens—USNM 125874, 160129. Unfigured specimens—USNM 125875, 126385. 160127— Spllaerirhynchia sp. 2 Plate 13, figures 1—6 Exterior.—Unequally biconvex shells with brach- ial valve having greater degree convexity. Shell Outline longitudinally elliptical, maximum width lo- cated about two-thirds the distance to anterior mar— gin. Beaks attenuated, pedicle valve incurved. Pedi— cle valve cardinal margin terebratulid in form. Brachial valve bears low fo-ld, pedicle valve corre- sponding sulcus. Anterior commissure uniplicate, crenulate. Relatively coarse, rounded costae, two on fold in small specimens to six in large specimens. EARLY PALEOZOIC BRACHIOPODS OF THE MOOSE RIVER SYNCLINORIUM, MAINE cc 9: ac: 10.25 10.15 Ce cc 3% 10.7 10.5 FIGURE 14.—Serial sections of “Ancillotoechia” cf. “A.” altisulcata (Amsden, 1951), parallel to hinge line distance from beak given in millimeters; X 4.5. Hardwood Moun- tain Formation, USGS loc. 3488—SD, Spencer Lake quad- rangle, Somerset County, Maine. SYSTEMATIC PALEONTOLOGY 33 Sulcus of small specimens contains one costa, flanks bear from six to eight costae each. Lateral and an- terior margins rounded. Radial ornamentation crossed by concentric growth lines. Pedicle valve—Serial sections of pedicle valve show presence of short dental lamellae bounding delthyrial cavity. They are partly obsolete due to deposition of secondary material in umbonal cavities. Muscle field very deeply impressed into secondary material deposited in posterior portion of shell. Brachial valve interior.—Posteriorly located, small septalium supported by long median septum extends to about midlength. Interior of valve strongly pli- cated by impress of costae except in umbonal region. Comparison.—S. sp. 2 has much coarser costae than S. sp. 1 and resembles S. sp. 3 in this regard. S. sp. 2 has a more deeply impressed pedicle valve muscle field than S. sp. 3. Not enough material of any of these forms is available for sound specific identification. Occurrence.—USGS loc. 3499-SD, Spencer Lake quadrangle, Somerset County, Maine. Stratigraphic location—Beck Pond Limestone (Lower Devonian). Figured specimens.——USNM 125877, 125876. Unfigured specimens.—USNM 125878, 125879, 126389. Sphnerirllyncllin sp. 3 Plate 13, figures 7—12 Exterior—Unequally biconvex shells, brachial valve having about one and one-half times convexity of pedicle valve. Valve outline subcircular, greatest width near midlength. Anterior and lateral margins rounded. Brachial valve bears low, broad fold, pedi- cle valve corresponding sulcus. Pedicle valve beak incurved, cardinal margins terebratulid in form. Anterior commissure uniplicate, crenulate. Low, rounded costae with about Six on fold, five in sulcus, about ten to twelve each flank. Pedicle valve interior—Dental lamellae very short, almost entirely obsolete due to deposition of secondary material in umbonal cavities. Muscle field extends about one-third distance to anterior margin, very elongate in form. Well-developed pedicle callist located at posterior of delthyrial cavity. Anterior of pedicle callist a pair of very narrow anterolaterally striated diductor impressions. Lateral to media] di- ductor impression pair roughened, elongate impres- sions which may have served as site for pair of lateral diductors. Peripherial parts of valve plicated by impress of external ornamentation. Brachial valve interior—Serial sections, septalium supported by long median septum. show Occurrence.—USGS loc. 3601—SD, Spencer Lake quadrangle, Somerset County, Maine. Stratigraphic locatiou.—Beck Pond Limestone (Lower Devonian). Figured specimens.——USNM 125870, 125871. Sphnerirhyncllia? sp. Plate 13, figures 13—18 Exterior.—Unequally biconvex shells, brachial valve about twice convexity of pedicle valve. Speci- men outline transversely elliptical although so de- formed that it is difl‘icult to be sure of original form. Maximum width located near midlength. Pedicle valve bears low, broad fold, pedicle valve correspond- ing sulcus. Lateral and anterior margins rounded. Anterior commissure uniplicate, crenulate. Fold bears about four to five low, rounded costae, sulcus contains about three to four costae. Costae in tongue of pedicle valve medially grooved. Each flank bears about 17 costae. Pedicle valve interior—Dental lamellae very short, obsolete due to deposition of secondary ma- terial in umbonal cavities. Pedicle callist deeply im- pressed, occupies posterior part delthyrial cavity. Diductor impressions deeply impressed, abut an- terior face pedicle callist. Diductors very narrow, medially separated by low myophragm, extend an- teriorly about one-third length of valve. Impression of costae noticeable in peripheral region of valve, absent posteromedially where overlain by secondary material. Brachial valve interior—Small, posteriorly lo— cated septalium supported by median septum that extends anteriorly about one-third length of valve. Septalium part roofed over by conjunct hinge plates as in Cupularostrum. Dental sockets medially roofed over by projecting edges of hinge plates. Species is similar to Sphaerirhynchia except as regards con- j unct hinge plates; form very unlike Cupularostrum which does not approach almost globular shape of these shells. Occurrence—USGS loc. 3488—SD, Spencer Lake quadrangle, Somerset County, Maine. Stratigraphic location—Hardwood Mountain For- mation (Upper Silurian). Figured specimens.—USNM 127385A, B. Family UNCERTAIN Genus SULCATINA Schmidt, 1964 Sulc-finn sp. Plate 13, figures 19—25 Exterior.—Unequally biconvex shells with brach- ial valve about 11/; times convexity of pedicle valve. Greatest width about three-quarters of distance to anterior margin. Shell outline transversely elliptical. 34 EARLY PALEOZOIC BRACHIOPODS OF THE MOOSE RIVER SYNCLINORIUM, MAINE Pedicle valve beak strongly incurved, margin tere- bratulid in form. Lateral margins strongly rounded, anterior margin gently rounded. Brachial valve has pronounced broad fold which bears about five to six rounded costae. Pedicle valve has corresponding sul- cus which commonly contains about four or five costae. Flanks each bear about nine to eleven costae. Anterior commissure uniplicate, crenulate. On speci- mens studied costae broaden anteriorly, not ob- served to bifurcate or increase by implantation. Hinge line very short, curved in anterolateral direc- tion. Brachial valve beak completely concealed be- neath incurved pedicle valve beak. Pedicle valve interior—Dental lamellae very short, almost entirely obsolete in large specimens owing to deposition of secondary material in um- bonal cavities. They support stout, plate-like hinge teeth which project beyond relatively short hinge line. Crural fossettes relatively narrow, laterally directed, bordered by hinge line and hinge teeth. Muscle field deeply impressed, consists of subcircu- lar diductor impression, medially divided by low myophragm, which anteriorly and laterally sur- rounds small, posteriorly located, elongate adductor impressions. Behind myophragm, anterior or promi- nent pedicle callist which occupies posterior wall and base of delthyrial cavity, are two pairs of elongate adjustor impressions. Area lateral to muscle field somewhat roughened by deposits of secondary material, and muscle field itself surrounded by low ridge secondary material which begins at base of dental lamellae and continues in anterior and then ‘anteromedial direction before becoming very faint in front of myophragm. Brachial valve interior.—Bifid cardinal process whose lobes curve to form scoop—shaped structure. Anterior part of structure nearly vertical but pos- teriorly swing medially and become inclined before joining near midline. Lateral faces of lobes have steps which parallel bottom of elongate, antero- laterally directed dental sockets which are roofed medially by projecting cardinal process lobes. Noto- thyrial cavity, between cardinal process lobes, has secondary material laid down anteriorly forming lozenge-shaped outline. Anterior of notothyrial cav- ity is low, sharp myophragm or ridge which extends about one-half length of valve. On either side of cardinalia and slightly anterior are deeply impressed, subcircular impressions of posterior adductor scars. Anterior of posterior adductor scars are elongate, paired anterior adductor impressions which extend to about midlength. Comparison—S. sulcata (Cooper, 1942) has more costae on the fold and sulcus than does the material from Maine. S. tennesseensis (Feerste) has about the same number of costae but is more subcircular in outline than the Maine material; as Tennessee specimens are smaller; they may be an earlier growth stage of the Maine material. Adequate com- parative specimens are not available to specifically identify the Maine material. Occurrence.—USGS loc. 3488—SD, Spencer Lake quadrangle, Somerset County, Maine. Stratigraphic location.—Hardwood Mountain For- mation (Upper Silurian). Figured specimens.—USNM 125850, 125853A, 125847, 125848. Unfigured specimens.—USNM 125849, 126388. Family EATONllDAE Genus COSTELLIROSTRA Cooper, 1942 Costellirostn 8]). Plate 14, figures 1~5 Exterior.—Unequally biconvex shells with brachi- a1 valve having greatest convexity. Brachial valve develops relatively narrow fold anterior of mid- length. Fold rises abruptly at angle of about 30° from rapidly descending anterior part of valve. Pedicle valve bears relatively shallow sulcus corre- sponding to fold, anterior part of pedicle valve pro- longed into a tongue. Specimen outline subtriangular due to attenuated beak. Greatest width slightly an- terior of midlength. Lateral and anterior margins rounded. Pedicle valve beak incurved, cardinal mar- gins terebratulid in form. Relatively low, fine striae barely visible on Maine material. Pedicle valve interior.——Imperfectly preserved pedicle valve interior shows pair deeply impressed adductor impressions, small in size, sunk into pit near anterior part of muscle field. Brachial valve interior—Median septum extends to about midlength on Maine material, nature of cardinalia not established. Comparison.—-External form and ornamentation, coupled with the nature of the muscle field in the pedicle valve, justify assignment to Costellirostra, but specific identification is not possible. Occurrence.—USGS 10c. 280'6—SD, Pierce Pond quadrangle, Somerset County, Maine. Stratigraphic location.——-McKenney Ponds Mem- ber of the Tarratine Formation (Lower Devonian). Figured specimen.——USN M 125885. Unfigured specimens.—USNM 125884, 125886. Genus EATONIA Hall, 1857 Eatoniu cf. E. medialis (Vanuxem, 1842) Plate 14, figures 6—13 Exterior.—-Unequally biconvex shells with brachi- a1 valve very inflated, pedicle valve gently convex. SYSTEMATIC PALEONTOLOGY 35 Brachial valve bears prominent fold which originates just anterior of umbo and rises at angle of about 30° from descending anterior part of valve. Shell outline transversely elliptical, greatest width near midlength. Pedicle valve beak incurved, hinge line curved anterolaterally. Rounded costae, four on fold, three in sulcus, and six on each flank. Lateral mar- gins rounded, but anterior margin almost straight. Anterior part of sulcus projects up as prominent tongue to engage fold of brachial valve. Anterior commisure uniplicate, crenulate. Pedicle valve interior—Dental lamellae not ob- served on Maine specimens, but deeply impressed muscle field present. Muscle field consists of paired diductor impressions, surrounded by narrow ridge of secondary material which originates near sides of delthyrial cavity, enclosing pair of small, elongate, medially divided adductor impressions sunk below level of diductor impressions. Low narrow median septum traverses muscle field, reaches to about mid- length. Impression of external costae marked in peripheral regions of valve. Brachial valve interior—Massive, terminally bi- fid, posteriorly located cardinal process supported by long narrow median septum which reaches almost to midlength. Brachiophores appear to be slender and diverge anterolaterally from base of cardinal process. Muscle field divided by median septum and consists of two pairs adductor impressions, posterior pair small and subcircular in outline, anterior pair elongate and extending to about midlength. Muscle impressions separated by low rounded ridge of sec- ondary material about normal to median septum. Comparison.—Maine specimens resemble E. me— dialls, but their poor state of preservation precludes specific identification. Occurrence.—USGS 10c. 3499—SD, Spencer Lake quadrangle, Somerset County, Maine. Stratigraphic locatiom—Beck Pond Limestone (Lower Devonian). Figured specimens.——USNM 125872, 125882B. Unfigured specimens.——USNM 126387. 125873, Estonia? 3]). Discussion.—A few poorly preserved specimens from Parker Bog Formation have prominent fold and sulcus with Eatom'a type costae. Nothing is known of the interior of this material; exterior is so poorly preserved that assignment to Eatomla can only be speculative. Occurrence.—USGS loc. 3477-SD, Pierce Pond quadrangle, Somerset County, Maine. Stratigraphic location.——Parker Bog Formation (Lower Devonian). Unfigured specimens.—USNM 126388. Family RHYNCHOTREMATIDAE Genus MACHAERARIA Cooper, 1955 Machenria Inninensis Boucot, n. sp. Plate 14, figures 14—21 Exterior.—Unequally bioonvex. shells with brachi- al valve having about 11/2 times convexity of pedicle . valve. Shell outline transversely elliptical with great- est width near midlength. Brachial valve bears rela- tively broad low fold which originates near umbo; pedicle valve has corresponding sulcus. Pedicle valve beak slightly incurved, hides beak of brachial valve. Cardinal margins subterebratulid. Pedicle valve fora- men submegathyrid, deltidial plates appear to be apically conjunct. Lateral and anterior margins rounded. Pedicle valve sulcus prolonged into tongue which fits against fold of brachial valve. Angular costellae about four on fold, three in sulcus, about eight each flank. Anterior commissure uniplicate, crenulate. Pedicle valve interior—Short dental lamellae sup- port relatively stubby hinge teeth. Dental lamellae almost obsolete due to deposition of secondary ma- terial in umbonal cavities. Muscle field well im- pressed, consists of elongate diductor impression medially divided by low myophragm and surround- ing a pit which contains small, elongate adductor impressions. Periphery crenulated by impression of external costellae, but remainder of interior rela- tively smooth due to deposition of secondary ma- terial. Brachial valve interior.—Linear cardinal process anteriorly joined by low, rounded myophragm or ridge which extends to about midlength of valve. Laterally flanking cardinal process are relatively stubby hinge plates supported by crural plates. Na- ture of brachiophores not determined on Maine ma- terial. Muscle field quadripartite, consists of pair medial adductor impressions which are elongate in outline and extend to about midlength from just anterior of cardinalia. Lateral pair adductor impres- sions subtriangular in outline, located just anterior of crural plates. Dental sockets relatively shallow, laterally directed. Base each crural plate medially directed to meet myophragm, and form sessile cru- ralium. Comparison—M. mainensis lacks flaring margins of sulcus as developed in Early Devonian type spe- cies M. formosa. Occurrence.—USGS locs. 3479—SD, 2712—SD, Spencer Lake quadrangle, Somerset County, Maine. 36 Stratigraphic location-Base of Hobbstown For- mation (Upper Silurian). Holotype.—USNM 125869. Figured specimens.—USNM 125867A, 125868, 125869A. Unfigured specimens.—USNM 126376, 126377. Suborder ATRYPOIDEA Snperfamily ATRYPACEA Flmily ATRYPIDAE Subfamily ATRYPINAE Genus ATRYPA Dalman, 1828 Atrypa cf. A. tennesseensis Amsden, 1949 Plate 14, figures 22—27 Exterior—Subsequently biconvex Shells, brachial valve slightly more convex than pedicle valve. Valve outline variably transverse to elongate, but deforma- tion may have partly determined outline because most specimens are more or less crushed. Straight hinge line, shorter than greatest width near mid- length. Anterior commissure rectimarginate, crenu- late. Cardinal margin submegathyrid, delthyrium appears unmodified. Pedicle valve beak probably in- curved over that of brachial valve. Radially disposed costae crossed by very prominent concentric lamel- lae, free at anterior ends as frills. Brachial valve bears about four unbranched costae on each side of median costa while pedicle valve bears five un- branched costae each side of midline. No specimen exceeds about 1 cm length. Lateral and anterior mar- gins rounded. Pedicle valve interior—Very short dental lamellae on either side of delthyrial cavity support stout hinge teeth which project vertically and have sub- circular cross section. Muscle field poorly impressed, appears flabellate, extends anteriorly to near mid- length, about half as wide as valve. Posteriorly located, elongate pair of adductor im- pressions anteriorly and laterally surrounded by flabellate diductor impressions. Costae impression marked on interior of valve except in posteromedial parts. Brachial valve interior—Pair of discrete hinge plates diverge laterally from midline and form an- terior socket walls. Crenulations barely discernible in bottom of dental sockets. Low myophragm ex- tends anteriorly from notothyrial cavity to position one-third of distance from posterior margin. Muscle field very weakly impressed. Costae impression marked except in posteromedial parts of valve. Comparison—Maine specimens resemble small specimens of Atrypa tennesseensis in external form and ornamentation. Secondary costae are. not present except in anteriormost parts of the Maine specimens, whereas in larger specimens from Oklahoma (Ams- den, 1951, pl. 17, fig. 33) numerous, well-defined EARLY PALEOZOIC BRACHIOPODS OF THE MOOSE RIVER SYNCLINORIUM, MAINE costae originate by bifurcation in peripheral regions of shell. Amsden’s figure 32 is a specimen with anomalous ornamentation consisting of costae which are much stronger and bifurcate near umbo and may not belong to A. tennesseensis. Occurrence.—USGS loc. 347 9-SD, Spencer Lake quadrangle, Somerset County, Maine. Stratigraphic location—Base of Hobbstown For- mation (Upper Silurian). Figured specimens.—USNM 126002—126004. Unfigured specimens.—USNM 12‘6396-126401. Atrypa cf. A. arctostriata Foerste, 1903 Plate 14, figures 28—29 Exterior.—Subequally biconvex shells with brachi- al valve having somewhat greater degree of con- vexity than pedicle valve. Shell outline subcircular, greatest width near midlength. Anterior commis- sure rectimarginate, crenulate. Lateral and anterior margins rounded. Frill of spines surrounds periph- ery, costellae increase by bifurcation and are crossed by concentric lamellae having free frills. Pedicle valve interior—Only one fragmentary pedicle valve is available, but it indicates that in small shells the muscle field is weakly impressed, flabellate in form, posteriorly bounded by short dental lamellae. Brachial valve interior—Discrete hinge plates an- teriorly bound dental sockets. Hinge plates and dental sockets anterolaterally directed. Short myo- phragm medially divides weakly impressed, poster- ior, elongate adductor impressions, posteriorly stops at undivided notothyrial cavity, which is probably the site for diductor attachment. Dental sockets very feebly crenulated basally. Anterior of posterior pair of adductor impressions is peripheral impress of external ornamentation. Comparison.—The Maine material is of the same external form and ornamentation as Atrypa arcto- striata, but is too poorly preserved to be specifically assigned without doubt. Occurrence.—USGS 10c. 3479-SD, Spencer Lake quadrangle, Somerset County, Maine. Stratigraphic location—Base of Hobbstown For- mation (Upper Silurian). Figured specimens.—USNM 126014A, 126015B. Atrypa "reticularis" (Linnaeus, 1767) Plate 15, figures 1—6 Atrypa “reticularis” (Linnaeus, 1767), see Boucot, 1959b, p. 741—742, pl. 91, figs. 7—9. Exterior.—Unequally biconvex shells, brachial valve about twice convexity of pedicle valve. Hinge line about two-thirds as wide as maximum width, which is at about one-third of distance to anterior SYSTEMATIC PALEONTOLOGY 37 margin. Cardinal margins submegathyrid. Delthyr- ium open, apparently unmodified. Pedicle valve weakly sulcate. Anterior commissure weakly unipli- cate, crenulate. Shells longitudinally elongate to sub- circular with rounded anterior and lateral margins. Costellae increase by bifurcation, crossed by con- centric growth lines. Pedicle valve beak straight to gently incurved. Pedicle valve interior.—Stout, bladelike hinge teeth project anteromedially from rear of valve, sup— ported by short dental lamellae which are almost entirely obsolete due to deposition of secondary ma- terial in umbonal cavities. Muscle impressions large, flabellate, extends to about midlength, about two- thirds as wide as valve. Muscle impressions elliptical pair small, posteriorly located adductor impressions laterally and anteriorly surrounded by deeply im- pressed diductor field. Diductor field radially divided into number of sectors by short myophragms which begin near anterior margins of delthyrial cavity. Deeply impressed horizontally striate pedicle callist located on posterior wall delthyrial cavity. Crural fossettes relatively deep, restricted to area lateral of hinge teeth. Valve periphery crenulated by im- press of external ornamentation. Areas lateral to muscle impressions deeply pitted. Comparison.—Above description is based on speci- mens from the Tomhegan Formation which resemble A. reticularis s. l. in character of external form and ornamentation. Additional specimens of Atrypa cf. A. reticularis, in a poor state of preservation, have been obtained from other units. Some from the Hardwood Mountain Formation are covered with spines and have a peripheral frill of spines. Occurrence.—USGS locs. 3485—SD. 3150-SD, 3475—SD, 3496—SD, 3469—SD, 3499-SD, 2820-SD, 2750—SD, 3488—SD, 2761—SD, 2840—SD, 2730—SD, Somerset and Piscataquis Counties, Maine. Stratigraphic location—Tomhegan Formation (Lower Devonian), Seboomook Formation (Lower Devonian), Beck Pond Limestone (Lower Devon- ian), Hardwood Mountain Formation (Upper Silur- ian), hornfels on Limestone Hill (Lower Silur- ian), undifferentiated Silurian beds on Deer Island. ‘Figured specimens.——USNM 126008, 126010, 126011, 126068. Unfigured specimens.—USNM 126390—126395. Family LlSSATRYPlDAE Genus LISSATRYPA Twenhofel, 1914 Lissnlrypn sp. Plate 15, figures 7—16 Exterior.—Unequally biconvex shells, pedicle valve about 11/; times convex as relatively gently convex brachial valve. Valve outline subcircular, greatest width slightly posterior of midlength. Lat- eral and anterior margins evenly rounded. Anterior commissure rectimarginate. Pedicle valve delthyrium open, unmodified. Hinge line narrow, about one fifth as wide as greatest width of shell. Cardinal margins terebratulid. Pedicle valve beak suberect, tends to conceal brachial valve. Concentric growth lines. Pedicle valve interior—Stout hinge teeth occur either side delthyrial cavity, laterally bound slightly raised pedicle callist on posterior wall delthyrial cav- ity. Muscle field deeply impressed, consists of nar- row, median adductor impression on raised track and laterally bounded by elongate diductor impres- sions which extend to midlength and are medially bounded by anterolaterally trending, divergent ridges which originate at base of median adductor track. Valve interior smooth. Brachial valve interior.—Discrete, bulbous hinge plates medially join low, broad median septum which is triangular in cross section posteriorly and linear anteriorly. Muscle field two pairs elongate, pos- teriorly situated adductor impressions. Muscle field extends about one-third distance to anterior margin and is subcircular in outline. Spires atrypoid in nature, about three loops to each spire. Comparison.—The material is too poorly pre- served to be specifically identified. The absence of a sulcus clearly indicates this material cannot be as- signed to the genus M eifodia. Occurrence.—USGS locs. 3488—SD, 3473—SD, Spencer Lake quadrangle, locality 4843—SD, Attean quadrangle, Somerset County, Maine. Stratigraphic location—Hardwood Mountain For- mation (Upper Silurian). Figured specimens.——USNM 125905A,B, 126000, 125908A,B, 160130A,B. Unfigured specimens.—USNM 125906, 125909, 126013. Genus NANOSPIRA Amsden, 1951 Nanospin? sp. Plate 15, figure 16 Exterior.—Brachial valve convex, bears promi- nent, broad sulcus originating near umbo. Maximum width located slightly anterior of curved hinge line. Lateral and anterior margins rounded. Anterior commissure sulcate. Exterior smooth except for con- centric growth lines. Brachial valve interior—Pair of laterally directed hinge plates separated by median trough. Interior smooth. Comparison—Single specimen from Maine has external form of Nanospira but not enough evidence is available for generic assignment to be positive. 38 EARLY PALEOZOIC BRACHIOPODS OF THE MOOSE RIVER SYNCLINORIUM, MAINE Occurrence—USGS loc. 3479-SD, Spencer Lake quadrangle, Somerset County, Maine. Stratigraphic location—Base of Hobbstown For- mation (Upper Silurian). Figured specimen.—USNM 126108A. Superfnmily DAYIACEA Family LEPTOCOELIIDAE Genus LEPTOCOELIA Hall, 1859 Leptocoelia flnbellites (Conrad, 1841) Plate 15, figures 17—24 Leptocoelia flabellites (Conrad, 1841), see Boucot, 1959b, p. 741, pl. 91, figs. 1—6. Exterior—Unequally convex shells with almost flat brachial valve, convex relatively naviculate pedi- cle valve. Greatest Width near midlength. Lateral and anterior margins evenly rounded. Brachial valve bears low fold, originating near midlength of valve; pedicle valve has corresponding sulcus. Anterior margin uniplicate and strongly crenulate. Del- thyrium open and unmodified. Pedicle valve beak suberect and incurved so as to conceal beak of brachial valve. Cardinal margins submegathyrid. Hinge line about two-thirds maximum width. Costae crossed by strong, concentric growth lines. Brachial valve fold bears two costae with four costae on each flank; sulcus of pedicle valve contains one costa with four to five on each flank. Costae increase in width anteriorly, do not bifurcate. Interspaces relatively broad, fiat, costae evenly rounded. Pedicle valve interior—Stout hinge teeth, having triangular cross section; lie on either side of del- thyrial cavity. Anterior face crural fossette crenu- lated. Apex of triangular cross section of hinge tooth points anteriorly. Rear of delthyrial cavity occupied by pedicle callist located on slightly raised, striated pad of secondary material. Median face of each hinge tooth indented by step which parallels floor of valve and serves to articulate with distal part of brachio- phore. Muscle field fiabellate in form, reaches anteri- orly to about midlength, is from one-third to one- half maximum width of valve. Muscle field consists of flabellate diductor impression, medially separated by low sharp myophragm which is most pronounced anteriorly. Diductor field encloses deeply impressed, small, elongate, paired adductor impressions which are longitudinally striate and partly roofed over posteriorly by projecting deposit of secondary ma- terial. Umbonal region of valve relatively smooth, remainder of valve strongly crenulated by impress of external ornamentation. Brachial valve interior.—Moundlike, terminally trifid cardinal process anteriorly joined by myo- phragm. Cardinal process laterally fused with socket plates. Crura bladelike, diverge anterolaterally. Dental sockets crenulated, roofed over medially by laterally inclined crura, anterolaterally rimmed by ridge of secondary material in large specimens. Myo— phragm extends to about midlength, narrow, low, and sharp anteriorly, becomes triangular in cross section and more massive posteriorly. Adductor field impression elongate in outline. In large speci- mens, median ridge of terminally trifid cardinal process rises above lateral ridges which are sub— dued by deposition of secondary material. Occarence. — USGS locs. 2691—SD—2705—SD, 2707—SD, 2709—SD—2711—SD, 2717—SD—2722—SD, 2725—SD—2727—SD, 2729—SD, 2731~SD—2735—SD, 2737—SD, 2738—SD, 2740—SD, 2741—SD, 2743-SD, 2745—SD, 2746—SD, 2748—SD, 2749—SD, 2751—SD, 2760—SD, 2761—SD, 2765—SD—2771—SD, 2775—SD, 2776—SD, 2778—SD—2790—SD, 2792—SD—2805—SD, 2807—SD, 2808—SD, 2811—SD—2813-SD, 2819—SD, 2821—SD, 2823—SD, 2824—SD, 2827—SD. 2829—SD, 2830—SD, 2832—SD, 2834—SD, 2837—SD, 2843—SD— 2849—SD, 2853—SD, 2856—SD—2858—SD, 2860—SD, 2862—SD, 2865—SD, 2870—SD, 2872—SD, 2877—SD, 2879-SD, 2880—SD, 2882—SD, 2883—SD. 2890—SD, 3089—SD—3092—SD, 3094—SD, 3225—SD—3229—SD, 3471—SD, 3474—SD, 3481—SD, 3482—SD. 3486—SD, Somerset County; 2861—SD, 3088—SD, Piscataquis County, Maine. Stratigraphic location.——Misery Quartzite Member of the Tarratine Formation (Lower Devonian), up- per and lower sandstones of the Tarratine Forma- tion (Lower Devonian), Seboomook Formation (Lower Devonian), questionably in the Kineo Mem- ber of the Tomhegan Formation (Lower Devonian). Distribution—This species is known from beds of late Helderberg to Schoharie age in eastern North America. Figured specimens.——USNM 126001, 125907, 126009, 126012, 127387. Unfigured specimens.—USNM 126403-126432, 126467, 126468, 126470, 126474—126507. Family ANOPLOTHECIDAE Subfamily COELOSPIRINAE Genus COELOSI’IRA Hall, 1863 Coelospiru 1]). Plate 15, figures 25—32 Discussion—Several localities in Somerset Coun- ty yielded shells with external ornamentation and internal features of Coelospira, but the material is inadequate for specific determination. Figured Hard- wood Mountain specimens are of Silurian rather than Devonian aspect. ' Occurrence.—-USGS locs. 3499—SD, 3469-SD, 3488—SD, 3483—SD, 3496-SD, 2806—SD, Spencer 126005, SYSTEMATIC PALEONTOLOGY 39 Lake quadrangle, loc. 5995—SD, Attean quadrangle, Somerset County, Maine. Stratigraphic locatiou.—Hardwood Mountain For- mation (Upper Silurian), Beck Pond Limestone (Lower Devonian), and McKenney Ponds Member of Tarratine Formation (Lower Devonian). Figured specimens.—USNM 126007,A,AA,C, 160131, 160132A,B, 160133. Uufigured specimeus.—USNM 126402. Suborder SPIRIFEROIDEA Superfamily DELTHYRIDACEA Family EOSPIRIFERIDAE Subfunily EOSPIRIFERINAE Genus HEDEINA Boucot, 1957 Hedeina cf. H. macropleura (Conrad, 1840) Plate 15, figures 33—35 Discussion—Single brachial valve from Beck Pond Limestone and two fragmentary pedicle valves from McKenney Ponds Member of Tarratine Forma- ti0n possess characteristic fine ornamentation of eospiriferids and coarse ornamentation characteris- tic of H edeiua. Specimen from Beck Pond Limestone has cardinalia identical with those of eospiriferids but is too poorly preserved to be specifically assigned without doubt. Material from McKenney Ponds Member certainly belongs to Hedeina as evidenced by broad, flat sulcus and rounded lateral costae, but its specific identity is in doubt. Material from McKenney Ponds Member might have been reworked from preexisting strata of Beck Pond Limestone Which were removed by erosion before deposition of the Tarratine Formation. Occurrence.—USGS locs. 3499—SD, 2810—SD, 2806—SD, Spencer Lake and Pierce Pond quad- rangles, Somerset County, Maine. Stratigraphic location—Beck Pond Limestone (Lower Devonian) and McKenney Ponds Member of the Tarratine Formation (Lower Devonian). Figured specimens—USN M 126006, 126076. Family DELTHYRIDIDAE Subfamily DELTHYRIDINAE Genus DELTHYRIS Dnlmnn, 1828 Delthyris cf. D. kozlowskii Amsden, 1951 Plate 16, figures 1—5 Exterior.—Unequally biconvex shells, brachial valve gently convex, pedicle valve subconical. Shell transversely elongate, outline elliptical. Greatest width located at straight hinge line. Brachial valve bears low rounded fold; pedicle valve has correspond- ing sulcus. Anterior commissure uniplicate, gently crenulate. Lateral margins evenly rounded. Pedicle valve interarea steeply apsacline, slightly concave posteriorly, brachial valve interarea orthocline. Pedi- cle valve interarea about two-thirds as long as pedi- cle valve, brachial valve interarea very short. Del- thyrium narrow, including an angle of about 20°. Deltidial plates narrow, located normal to interarea. Not determined if they are apically conjunct. Two or three low rounded plications separated by broad in- terspaces each flank. Radial striations terminate in fringe of spines on edge of each growth lamella. Pedicle valve interior—Median septum extends about three-quarters distance to anterior margin, bisects delthyrial cavity, and is laterally flanked by dental lamellae. Median septum very high posteri- orly, slopes steeply in anterior direction. Dental lamellae subparallel, extend anteriorly about one- third length of valve. Nature of hinge teeth undeter- mined. Impressions of muscle field not discernible. Interior gently crenulated by impress of external coarse ornamentation. Brachial valve interior.——Medially conjunct hinge plates, postero-median portions excavated and crenu- lated by slits of ctenophoridium. Hinge plates sup- ported laterally by fulcral plates that floor dental sockets. Myophragm bisects area of muscle attach- ment to about midlength. Muscle field itself not discernible. Comparisou.—Maine material is similar to Del- thyris kozlowskii, but lack of adequate material pro- hibits positive identification. Foerste (1903, p. 710; 1909, p. 92, pl. 2, figs. 31A— B)described a species “Reticularia pegramensis,” that possesses a median septum in the pedicle valve as well as dental plates and has all the characteristics of Delthyris s. s. Foerste’s specimens do not, how- ever, possess lateral plications as do most forms as- signed to Delthyris. Amsden’s species D. kozlowskii resembles D. pegramensis in all regards except that the former is strongly plicated. Some specimens of D. kozlowskii from Oklahoma bear very weak lateral plications, but none were observed without a trace. Maine material shows a transition between strongly plicated forms and ones with unplicated flanks. Occurrence—USGS loc. 3479—SD, Spencer Lake quadrangle, Somerset County, Maine. Stratigraphic location—Base of Hobbstown For- mation (Upper Silurian). Figured specimens.——USNM 126038, 126027, 126025. Genus HOWELLELLA Kozlowski, 1946 Howellelln? up. Plate 16, figures 6—7 Pedicle valve interior—Short gently diverging dental lamellae extend anteriorly about one-fifth length of valve. Low myophragm extends slightly anterior of dental lamellae. Shell subcircular in out- line, highly convex. Hinge line straight, both lateral 126036, 4O EARLY PALEOZOIC BRACHIOPODS OF THE MOOSE RIVER SYNCLINORIUM, MAINE and anterior margins rounded. Median sulcus; flanks bear four to five low rounded costae separated by broad interspaces. Muscle field not discernible. Discussion.—Only a few poorly preserved impres- sions of interior of pedicle valve were available. Material is not adequate for specific identification. Ignorance of fine ornamentation and of brachial valve makes generic assignment questionable. Occurrence.—USGS loc. 2950—SD, Spencer Lake quadrangle, and 4841—SD, Attean quadrangle, Som- erset County, Maine. Stratigraphic location—Basal part of the Hard- wood Mountain Formation (Upper Silurian). Figured specimens.—USNM 126020, 160134. Unfigured specimens.——USN M 126021. Howellelll? cf. H. cyclopterus (Hall, 1857) Plate 16, figures 8, 9 Exterior—Evenly convex pedicle valve with transversely elongate, elliptical outline. Greatest width at straight hinge line. Pedicle valve interarea apsacline, slightly incurved. Anterior and lateral margins rounded. Median sulcus; flanks bear about six costae each. Costae rounded in cross section, separated by relatively broad interspaces. Fine orna- mentation consists of striations terminating as tiny spines on edge each growth lamella. Comparison—No information is available regard- ing the interior of the pedicle or brachial valves; therefore, generic assignment is not certain. If shell is a delthyrinid, it is similar to H. cycloptera. Occurrence.—USGS 10c. 3499—SD, Spencer Lake quadrangle, Somerset County, Maine. Stratigraphic looation.—Beck Pond Limestone (Lower Devonian). Figured specimens.——USNM 126257A. "Howellella" tomheglnensis Boucol, n. sp. Plate, 16, figures 10—18 Exterior.—Shells unequally biconvex, pedicle valve slightly more convex than brachial valve. Maxi- mum width slightly anterior of straight hinge line. Shells transversely elliptical in outline. Lateral mar- gins sharply rounded, anterior margin gently rounded. Brachial valve bears relatively narrow fold with subrectangular cross section, may bear indis- tinct medial groove. Sulcus form corresponds to fold, relatively rectangular in cross section. Anterior commissure uniplicate, crenulate. Relatively broad, rounded plications separated by narrow, relatively deep interspaces. Flanks bear seven to ten costae; average about nine. Costae Widen anteriorly, show no evidence of bifurcation or origin by implantation. Pedicle valve interarea apsacline to orthocline, con- cave posteriorly, and slightly incurved. Brachial valve interarea anacline to orthocline. Pedicle valve interarea long, that of brachial valve very short. Delthyrium includes angle of about 45°, is bordered by thin, discrete deltidial plates normal to hinge line. Striae terminate in fringes of spines which overlap each concentric growth lamella. Sulcus from two to three times as wide as first plication lateral to it. Pedicle valve interior.—Dental lamellae relatively short, extend anteriorly from one-fifth to one—third of length. Hinge teeth small, located on medial edge of hinge line. Muscle field deeply impressed in some specimens due to deposition of secondary material, weakly impressed in others; flabellate to elongate in form, bisected by low myophragm, consists of large, anteriorly located diductor impressions and pair of small, posteriorly located adductor impres- sions having rhomboidal outline. Dental lamellae diverge between second and third interspace lateral of sulcus. Umbonal regions of valve pitted. Periph- eral regions of valve crenulated by external orna- mentation; usually one or two fewer plications im- pressed than present on exterior. Dental lamellae indented by steps near edge of delthyrium; these steps serving as seats for deltidial plates. Brachial valve interior.—Posteriorly located cten- ophoridium lateral to which are discrete, baso- medially inclined socket plates whose basal margins extend below bases of sockets. Posterior face each socket plate forms anterior side of socket. Low myo- phragm bisects area of unimpressed muscle attach- ment. Interior crenulated by external ornamentation impress. Measurements.—A plot of pedicle valve length versus width (fig. 15A) shows a seemingly random distribution. This is probably due largely to me- chanical deformation plus the fact that only a small size range of specimens was available. Relation be- tween pedicle valve width and diductor muscle field width is essentially linear and shows a low degree of dispersion which may be due to deformation. Width of pedicle valve sulc-us increases at a constant rate relative to valve width (fig. 150) although a high degree of dispersion is present which may be due to mechanical deformation. The number of plications remains about constant relative to pedicle valve width although there is a slight tendency for in- crease in number with increase in size. Relation between pedicle valve width and sulcus width/first lateral interspace width is random (fig. 15E), prob- ably due to deformation. Relation between pedicle valve width and first lateral interspace width is linear (fig. 15F) with high degree of dispersion pos- sibly due to mechanical deformation. Relation of SYSTEMATIC PALEONTOLOGY 41 sulaus and first lateral interspace widths is linear (fig. 15G), with a high degree of dispersion. The high degree of dispersion shown by the measure- ments illustrates how mechanical deformation can alter characteristics of a sample to make growth rates and a growth curve slope difficult to under- stand. The relation of diductor width and length in the pedicle valve is essentially linear, with a low degree of dispersion. Relation between pedicle valve width and diductor width is essentially linear with a low degree of dispersion (fig. 153) . Plot of half number of plications against width of pedicle valve is normal with mode at nine plications (fig. 15D). Slight tendency for increase in number of plications from four to five in smaller specimens to five and six in larger specimens. Comparison.—H.? nerei (Barrande) has a more angular and proportionately wider fold than “H.” tomheganensis. “H.” gaspensis, “H” angustiplicata var. zaleszczykiensis, and “H.” mckenzica have more lateral plications than does “H.” tomheganensis, whereas “H.” angustiplicata has fewer plications. Occurrence—USGS locs. 2820—SD, 2750—SD, and questionably 2814-SD, Brassua Lake quadrangle, Somerset County, Maine. Stratigraphic location.—Tomhegan (Lower Devonian). Holotype.—USNM 126016. Figured specimens.—USNM 126032, 126017, 126045. Measured specimens.—USNM 125889. Genus ACROSPIRIFER Helmbrecln and chekind, 1923 Acrospirifer murcllisoni (Caslelnau, 1843) Plate 16, figures 19-25 Spim'fer murahisoni (Castelnau, 1843), see Clarke, 1900, p. 46—48, pl. 6, figs. 26—30. Exterior.—Unequally biconvex shells, pedicle valve about 1% times as convex as brachial valve. Shells transversely elliptical, greatest Width at hinge line. Short ears present on well-preserved specimens, but usually broken off. Brachial valve bears median fold with angular cross section, apex rounded. Pedi- cle valve sulcus corresponds to brachial valve fold. Five to seven plications on each flank. Lateral plica- tions have rounded cross section, separated by wide, rounded interspaces. Pedicle valve interarea gently apsacline, concave posteriorly, beak slightly in- curved. Brachial valve interarea gently apsacline to orthocline. Pedicle valve interarea relatively long, that of brachial valve relatively short. Delthyrium open, includes angle of about 90°, bordered by pair narrow deltidial plates inserted normal to interarea. Anterior commissure uniplicate, crenulate. Lateral Formation 125026, margins rounded, anterior margin nearly straight. Well-preserved specimens have fine, radiating striae which terminate as fringe of minute spines over anterior edge of each growth lamella. On most speci- mens fine ornamentation is abraded, shells appear to lack spines or striations between plications. Inter- area smooth except for growth lines which parallel hinge line. Pedicle valve interior.—-Short dental plates border sides of delthyrial cavity but become obsolete in larger specimens due to extensive deposition of sec- ondary material in umbonal and delthyrial cavities. Hinge teeth stout, triangular in cross section, lo- cated medial to hinge line. Muscle field very deeply impressed in deposits of secondary material at pos- terior of valve. Muscle field extends to about mid- length, relatively narrow. Consists of pair of median, narrow adductor impressions bounded by larger, elliptical diductor impressions. Posterior face of del- thyrial cavity bounded by rhomboidal impression of pedicle callist produced anteriorly, by deposition of secondary material, to a point forward of posterior tip of diductor impressions. Secondary material laid down in umbonal cavities is pitted. Valve peripheral regions strongly crenulated by impress of external ornamentation. Short myophragm bisects posterior part of muscle field. Medial face of each dental plate bears step in which to seat a deltidial plate. Brachial valve interior.—-Ctenophoridium situated on posterior wall of notothyrial cavity and laterally bounded by narrow, vertically inclined socket plates which curve to floor dental sockets. Myophragm bi- sects unimpressed muscle attachment area. Sockets shallow, anterolaterally directed. Interior strongly crenulated by external ornamentation impress. Measurements.—Relation between length and width of pedicle valve is essentially linear With a slope of about 60° (fig. 16A) . Degree of dispersion is not high, actual amount of dispersion increases with size, suggesting percentage variation remains about constant. Sample is normally distributed. Relation of pedicle valve width and first lateral interspace width is linear, shows a positive slope of about 85° with low degree of dispersion (fig. 163). Relation of pedicle valve Width to sulcus width is linear; slope about 75° indicates sulcus expands at a much greater rate than does the first lateral inter- space. The degree of dispersion of the sulcus and valve width is not great (fig. 160). Relation between sulcus width and first lateral interspace width is curved; strong tendency for the sulcus to increase in width at a greater rate than the lateral interspace with increase in sulcus width (fig. 16D). Relation between pedicle valve width and the ratio of sulcus '42 EARLY PALEOZOIC BRACHIOPODS OF THE MOOSE RIVER SYNCLINORIUM, MAINE 30 3 A LLI '— Lu 2 .1 :’ 2 Z 20 I. '— (D 2 Lu J LIJ > 10 _I < > Lu J 9 0 Lu “L o 0 1o 20 30 40 PEDICLE VALVE WIDTH, IN MILLIMETEFIS 40 4o 40 D (I) U) [E g n: Lu “J u.I I- }. ' I- ; 30 E 30 g 30 j j j 2 g 2 E z .2. :5 I‘ ;., I‘ E 20 I5 20 E 20 a g ; Lu Lu LLI > 2 3 2 .- > g > . LIJ m LIJ 5’ 10 d 10 § 10 E 5 0 Lu E I; n. o o o o 20 0 1o 20 10 2o DIDUCTOR WIDTH, IN MILLIMETERS SULCUS WIDTH, IN MILLIMETERS ONE-HALF THE NUMBER OF PLICATIONS FIGURE 15.——Measurements of “Howellella” tomheganensis Boucot n. sp. Tomhegan Formation, USGS loc. 2820—SD, Brassua Lake quadrangle, Somerset County, Maine. Width to first lateral interspace width is essentially linear but with a much higher degree of dispersion than the component measurements suggest. probably because the use of ratios tends to multiply the amount of variation (fig. 16E). Relationship between width of fold and sulcus with the respective first lateral interspace and plica- tion is linear, with moderate degree of dispersion. Slight tendency, in pedicle valves in sample from USGS loc. 2720—SD, for the sulcus to expand at greater rate than lateral interspace when sulcus width of about 3 mm reached. Relationship between width of valves and sulcus width/first lateral inter- space width and fold width/first lateral plication width is nearly random, because ratios are used. Relationship between valve width and widths of first lateral plication and interspace is essentially linear; moderate degree dispersion which increases SYSTEMATIC PALEONTOLOGY 43 4o 40 E F (0 (I) E as '— 5 30 ”:1 30 E :. —| _| =' E E E Z - I I E 20 '5 20 g E UJ ”>‘ . > _I o _l ' < < > > W E 10* d 10 17 E E m E m 0 o o 1 2 3 4 o 1 2 3 SULCUS WIDTH/FIRST LATERAL INTERSPACE WIDTH FIRST LATERAL INTERSPACE WIDTH, IN MILLIMETERS FIRST LATERAL INTERSPACE WIDTH, IN MILLIMETERS 0 1 2 3 4 5 6 7 SULCUS WIDTH, IN MILLIMETERS FIGURE 15.—Continued. with increase in size at greater rate than if per- centage variation remained constant. Relationship between pedicle valve width and width of the diductor field essentially linear, steeply sloping, and indicates the muscle field width in- creases more slowly than valve width. (fig. 16G) Dispersion is low. Relationship between diductor width and length is essentially linear with a moder- ate degree of dispersion (fig. 16F). Comparison—Acrospirifer hartleyi (Schuchert, 1913) and A. angulam's (Schuchert, 1913) are proba- bly synonyms of A. murchisomi, but adequate ma- terial is not available for comparison. Occurrence.—USGS 10c. 3482—SD, 3090—SD, 2847— SD, 2827-SD, 2890—SD, 2718—SD, 2727-SD, 2729— SD, 2777—SD, 2705—SD, 2862—SD, 2766—SD, 2834— SD, 3088—SD, 2089—SD, 2767—SD, Somerset and Piscataquis Counties, Main-e. The following USGS localities provided A. cf. A. murchisom': 2774—SD, 2733—SD, 2792—SD, 2857781), 44 EARLY PALEOZOIC BRACHIOPODS OF THE MOOSE RIVER SYNCLINORIUM, MAINE 30 5 w A II LLI I- . ”J . E _| :' E 20 . . .- z ' . I‘ . a . . . " p— .. ' . . O ' ‘. .'- z .. '_ -. m . .1 . ' . LIJ ._ . . 3 1o . . - ' < . . I. - > D I.” . .J I 9 . 0 Lu :1 0 0 1O 20 3O 40 50 PEDICLE VALVE WIDTH, IN MILLIMETERS I. I- 20 Q g B u.| o w E a: m Lu 0: I— a g E :1 1O _, _I < E E z ._ _ < n D I I I. —' .-:'~""'ro'-'.'. ' ._ ’ "“-'.|'"'"a"' a. (I) 0 n E o u. 0 10 20 30 4O 50 PEDICLE VALVE WIDTH, IN MILLIMETEFIS E 20 m C |_ D m E ._I -_J E ' ' 3 10 - - ' I ‘ . . .- ’ '- ' ‘6". I— . _ . ..' e : 1 no I. n ft" g . .I'. r. 'm. . 4“" g :I . u .‘x- I'~ I: U .- _ ' o. _' I D <0 0 O 10 20 3O 40 50 0 10 20 PEDICLE VALVE WIDTH, IN MILLIMETERS FIVE-IFST'KRGhBlnaéfizggCE FIGURE 16.——Measurements of Acrospz'rifer murchisom' (Castelnau, 1843). Lower sandstone of Tarratine Formation, USGS 10c. 2718—SD, Long Pond quadrangle, Somerset County, Maine. SYSTEMATIC PALEONTOLOGY 4 I '— 9 . E 8 < 3 a. I U) C [E . LLI - I— n E . - . _l . ,- - < . - _ - , E . .- .- I- 2 , o‘ . .- .' < . , . . ,. . _I , . . . - E - - k I I— 9 1 E U) D O J 3 U) o o 10 20 30 40 50 PEDICLE VALVE WIDTH, IN MILLIMETERS 20 F :If I-U) (DD: ZLu LUI— “'2‘ (E 23 ‘° SE 5.2-1" SE . ”l‘u’n.’ o o 10 2o DIDUCTOR WIDTH, IN MlLLlMETERS 3’5 20 LIJ '— L“ E E 2 E . f 10 - F- . I. o 9 . . E ..u.-':.q' i'.'..- o :0 n. a ' o I— . - u .. . D E‘ o o o 10 20 30 4o 50 FIGURE 16.——Continued. (Continued on following page). PEDICLE VALVE WIDTH, IN MILLIMETERS 45 46 EARLY PALEOZOIC BRACHIOPODS OF THE MOOSE RIVER SYNCLINORIUM, MAINE 6 H ONE-HALF THE NUMBER OF PLICATIONS 0 1O 2O 30 40 50 PEDICLE VALVE WIDTH, IN MILLIMETEHS FIGURE 16.—Continued. 2725—SD, 2810—SD, 2864—SD, 3474—SD, 2806—SD, 2824—SD, 2797—SD, 2832—SD, 2717—SD, 2691—SD, 2710—SD, 2861—SD, 2735—SD, 2769—SD, 2837—SD, Somerset County, Maine. Stratigraphic location.—Seboomoo~k Formation (Lower Devonian) , McKenney Ponds Member of the Tarratine Formation (Lower Devonian), and the upper and lower sandstones of the Tarratine Forma- tion (Lower Devonian). Figured specimens.—USNM 126047, 126043, 127386. Measured specimens.—USNM 125863—125864. Unfigm'ed specimens.—USNM 126526—126556, 126565—126568. Acrospirifer atlanficus (Clarke, 1907) Plate 17, figures 1—9 Spirifer primaevus Steininger var. atlanticus Clarke, 1907, p. 260—262. Spin'fer gaspensis Williams and Breger, 1916, p. 107—113, pl. 4, fig. 7, non Billings, 1874. Exterior.-—-Unequally bico‘nvex shells, pedicle valve having greater degree convexity. Greatest width at straight hinge line. Shell outline subcircu- lar to alate, deformation altered outlines to a cer- tain extent. Brachial valve bears broad, median fold with angular cross section, pedicle valve has corre- sponding sulcus. Pedicle valve interarea apsacline, concave posteriorly, beak slightly incurved. Brachial valve interarea gently anacline. Pedicle valve inter- 126042, area long, brachial valve interarea short. Pedicle valve delthyrium includes angle of about 60°—-90°, laterally bordered by pair narrow deltidial plates inserted normal to interarea. Six to 14 lateral plica- tions, depending on width of valve, 10 to 12 usual number among specimens studied. Anterior com- missure uniplicate, crenulate. Radial striae terminate as fringe of spines over anterior margins of each growth lamella. Fine orna- mentation variably preserved depending upon de- gree abrasion undergone by specimen; some shells partly smooth, partly striate, and partly fimbriate. Lateral margins gently rounded, anterior margin almost straight. Interarea smooth except for growth lines which parallel hinge line. Pedicle valve interior—Short, bladelike hinge teeth on medial side hinge line, supported by short, dental lamellae which are convex medially. Dental lamellae largely obsolete due to deposition of sec- ondary material in umbonal cavities. Muscle field deeply impressed, restricted to posterior half of valve. Muscle field relatively elliptical to subcircular in outline, consists of narrow, median adductor im- pressions flanked laterally by elongate, anteriorly expanding diductor impressions. Diductor impres- sions striate, extend posteriorly into pair of pointed chambers excavated in secondary material deposited in delthyrial cavity. Pad of secondary material slightly anterior of posterior part of diductor im- SYSTEMATIC PALEONTOLOGY 47 pressions bears relatively small, diamond-shaped pedicle callist. Interior deeply impressed by external ornamentation. One to five fewer lateral plications impressed on interior than are present on valve exterior. Brachial valve interior—Citenophoridium located on posterior face of notothyrial cavity, laterally flanked by pair of socket plates whose lower medial edges do not extend below base of sockets. Myo- phragm bisects unimpressed muscle field and reaches to about the midlength. Measurements—The relationship between length and width of pedicle valve is linear (fig. 170). The high degree of dispersion suggests mechanical de- formation altered the expected distribution. The width of sulcus increases linearly with in- crease in width of pedicle valve (fig. 17A), with a moderate degree of dispersion. The width of the first lateral interspace increases linearly with relation to the pedicle valve width (fig. 173), but the slope is much lower than that for the sulcus width, indicat- ing that the sulcus expands at a greater rate than the first lateral interspace. The relationship between sulcus width and first lateral interspace width is linear, with a moderate degree of dispersion (fig. 170). The low slope indicates sulcus expands at greater rate than interspace. The relationship be- tween pedicle valve width and the ratio of sulcus width to first lateral interspace width is linear, with a very high degree of dispersion (fig. 17E) , induced by the use of the ratio; however, it indicates that the sulcus becomes proportionately wider with increase in width of pedicle valve, that is, it flares anteriorly. The relationship between pedicle valve width and number of lateral plications on one side of sulcus is linear, with a low slope, and a moderate degree of dispersion (fig. 17F). Smaller specimens average about six plications on each side of the sulcus, larger ones about ten. Relation between width and length of diductor field is linear, with an increasing degree of dispersion with increase in size (fig. 17D). Slope of about 45° suggests muscle field expands about equally in both directions. Comparison.—Acrospirifer intermedius (Hall, 1859) has fewer lateral plications than A. atla'nticas. Discussion.—Clarke (1909, p. 83, pl. 19, figs. 5— 12) erroneously cited Tomhegan Point as the locality for specimens having the same aspect as those from adjacent Baker Brook Point. Rocks of Tomhegan Point belong to the Tarratine Formation; those of Baker Brook Point belong to the Tomhegan Formation. The material cited as “Spirifer cf. S. duodenarias (Hall)” by Williams and Breger (1916, p. 104—105, pl. 4, figs. 3—4) is small specimens of A. atlanticus. Williams and Breger’s locality 1061 B’, cited as Little Brassua Lake (1916, p. 109), was not reex- amined; may actually refer to mouth of Stony Brook, as judged from lithology of their specimens. Small specimens of A. atlanticus, less than about 3 cm in width, are difl‘icult to distinguish from A. murchlsoni, owing to similar shape and ornamen- tation. Occurrence.——-USGS locs. 2852—SD, 2820—SD, 2873—SD, 2750—SD, 2752—SD, 2841—SD, 2842-SD, Brassua Lake quadrangle, Somerset County, Maine. The following localities provided A. cf. A. atlantlcus: 2840—SD, 2730-SD, 2723-SD, 2814—SD, 2839—SD, Brassua Lake quadrangle, Somerset County, Maine. Stratigraphic location—Tomhegan Formation. Figured specimens—USNM 126041, 126019, 126031, 126163, 127388. Measured specimens.—USNM 125866, 125865. Unfigured specimens.—USNM 125939, 126557— 126564, 126578. Acrospirifer an. 1 Plate 17, figures 10—14 Exterior—Shells subequally convex. Greatest width at straight hinge line. Shells transverse. Brachial valve bears fold with low rounded to almost flat cross section, pedicle valve has corresponding sulcus. Seven to eight plications each flank. Costae have rounded cross section, separated by rounded interspaces. Lateral margins rounded, anterior mar- gins almost straight. Concentric growth lines crossed by radial striae terminating as fringe of spines over anterior margin each lamella. Anterior commis- sure uniplicate, crenulate. Pedicle valve interarea apsacline, beak slightly incurved. Delthyrium open, includes angle of about 60°. Delthyrium probably bordered by pair narrow deltidial plates. Pedicle valve interarea relatively long, brachial valve inter- area relatively short. Pedicle valve interior.——Short dental lamellae al- most entirely obsolete due to deposition secondary material in umbonal cavities. Medial face each dental lamella indented by step which probably served to seat deltidial plate. Muscle field deeply impressed in secondary material which lines delthyrial cavity and posterior part of valve. Muscle field tripartite, con- sists of elongate, median adductor impression later- ally bounded by pair anteriorly expanding diductor impressions. Peripheral regions crenulated by im- press external ornamentation. Brachial valve interior—Small ctenophoridium located posterior face notothyrial cavity, laterally flanked by discrete socket plates directly attached to shell walls. Muscle field not impressed, myo- 48 EARLY PALEOZOIC BRACHIOPODS OF THE MOOSE RIVER SYNCLINORIUM, MAINE 20 A .m ' u I . u—E 25 . - E2 ' . ' o w- 10 . ‘ .. . ' ' Dj . 35 . . 8; - A ' 0 O 10 20 30 40 50 60 70 80 90 PEDICLE VALVE WIDTH. IN MILLIMETERS UJ o 20 <0) sfi B I LuLLI l-E E3 _J:.J <2 10 $2 I- . n E w —' _l j 9 E ' S 2 3 '1 0 H 1 o 10 20 30 I . I5 PEDICLE VALVE WIDTH, IN MILLIMETEFIS E 3 w 30 9. 2 E C '— Lu 2 _I :’ E 20 Z 1 _ I . I— o I Z l LLI - .J . .a l; 10 .' _ _I 0 < o 10 2o 30 E . BRACHIAL VALVE WIDTH, IN MILLIMETERS 3 ' 5 LLI ‘1 o o 10 20 30 PEDICLE VALVE WIDTH, IN MILLIMETERS 20 _ 20 ‘ I E D I— E o 0 Z 2 U) Lu W Lu a: 4 E x 4 e . I. LU g E .u' .- 0.. 3 g I. . a 2' g 10 . . ; . . < 3 10 .' ' > _' . u o o > ...I a _l :' "- _I E g < E ' '- S 2 :T: 2 I — ' U " O < < 0‘: II co m 0 0o 10 20 30 o 10 20 30 BRACHIAL VALVE WIDTH, IN MILLIMETERS BRACHIAL VALVE WIDTH, IN MILLIMETERS FIGURE 18.—Measurements of Antispirifer harroldi Williams and Breger, 1916. Lower sandstone of Tarratine For- mation, Long Pond quadrangle, Somerset County, Maine. A, brachial valve width versus width of fold; USGS loc. 2721—SD. B, length versus width of pedicle valves; USGS 10c. 2720—SD. C, length versus width of pedicle valves; USGS 10c. 2721—SD. D, length versus width of brachial valves; USGS 10c. 2720—SD. E, length versus width of brachial valves; USGS 10c. 2721—SD. SYSTEMATIC PALEONTOLOGY 3O 20 ‘IO BRACHIAL VALVE WIDTH, IN MILLIMETERS 1 2 3 4 5 6 7 ONE-HALF THE NUMBER OF LATERAL PLICATIONS 30 20 10 BRACHIAL VALVE WIDTH, IN MILLIMETERS 0 1 2 3 4 5 6 ONE-HALF THE NUMBER OF PLICATIONS 20 PEDICLE VALVE WIDTH, IN MILLIMETERS O 1 2 3 4 5 6 7 ONE-HALF THE NUMBER OF PLICATIONS FIGURE 18.——Continued. F, width of brachial valve versus one half the number of lateral plic‘ations; USGS 10c. 2721—SD. G, width of brachial valve versus one half the number of plications; USGS loc. 2720-SD. H, width of pedicle valve versus one half the number of external plications; USGS 10c. 2720—SD. (Continued on following page). EARLY PALEOZOIC BRACHIOPODS OF THE MOOSE RIVER SYNCLINORIUM, MAINE 3O 2O 10 PEDICLE VALVE WIDTH, IN MILLIMETERS 0 1 2 3 4 5 6 7 NUMBER OF PLICATIONS ON LEFT SIDE 30 20 . 3 - 10 SULCUS WIDTH, IN MILLIMETERS 0 1 2 3 4 5 6 7 PEDICLE VALVE WIDTH, IN MILLIMETERS 30 20 PEDICLE VALVE WIDTH, IN MILLIMETEFIS 0 1 2 3 4 5 6 7 SULCUS WIDTH, IN MILLIMETERS FIGURE 18.—Continued. I, pedicle valve width versus number of plications on left side; USGS 10c. 2721—SD. J, pedicle valve width versus sulcus width; USGS loc. 2721—SD. K, pedicle valve width versus sulcus width; USGS 10c. 2720-SD. (Continued on following page). SYSTEMATIC PALEONTOLOGY E’ 30 m L '— LU g : .1 2 . E 20 . . . - I- . ' - Q - . ' E LU . 5 <10 > _J E I 0 < D: m o o 1 2 3 4 5 6 FOLD WIDTH, IN MILLIMETERS 3 M U) m Ll.| ’— Lu 2 2 In ' ' :3 2 .. . E o I I :. U E - '2 : ' 9 1 E z ' . 9 '— < 2 5' o '— (I) n_: 2 3 4 5 6 7 8 LL SULCUS WIDTH, IN MILLIMETERS FIGURE 18.—Continued. L, Width of brachial valve versus width of fold; USGS 10c. 2720—SD. M, width of sulcus versus width of first plication on pedicle valve; USGS loc. 2720—SD. (Continued on following page). 56 EARLY PALEOZOIC BRACHIOPODS OF THE MOOSE RIVER SYNCLINORIUM, MAINE m 3 o: u.I N . . ’— o. ”J I E . - _l :l o . E ' . . g 2 :5 . . . ' '— .. . 9 ' . 5 Lu . o ' ' . E 1 U) m - UJ i— Z '— U) E u. 0 1 2 3 4 5 6 7 SULCUS WIDTH, IN MILLIMETERS 4 0 (O n: [Ll |_. w . g 3 .J c -_-' . E I Z .. 2. . . E . . .. . 9 2 . 3 . ' . z . .. . 9 I O |— a . <1 9 . d 1 , ,4 U) E LL 0 . o I 2 3 4 5 6 7 FOLD WIDTH, IN MILLIMETERS FIGURE 18.—Continued. N, width of sulcus versus first interspace width; USGS loc. 2721—SD. 0, width of fold versus width of first pli- cation on brachial valve; USGS 10c. 2720—SD. (Continued on following page). PEDICLE VALVE WIDTH, IN MILLIMETERS 30 20 1O SYSTEMATIC PALEONTOLOGY 57 3 _I P . < I . - . ”Ii " U, . 1‘. E . . ‘ Z I— . . 9 "g 2 l— .— < -‘ . o :‘ ' I E 2 . . z . 5 —~ I E o - LL 8 1 8 LL 0 I I— l— 9 E 0o 1 2 3 4 6 FOLD WIDTH, IN MILLIMETERS Q ni ‘ R . .9- ’3'. .. I ."'°o 5.: 0‘. ' . a ' 1 2 3 0 1 2 3 4 SULCUS/FIRST LATERAL PLICATION SULCUS WIDTH/FIRST INTERSPACE WIDTH g 30 . w 30 u.| S E T ' u'? E E -° . E . . 2 " . : j -- 2 ' " - , E Z 2° ' . 2 2° . - 1‘ - . . . ; ' v 3 ' 3 . : , u.I . . . uJ ~ 3 1O . 3 1o ' ' < . . > § . .1 _l S S I I o g 3 m 0 g o 1 2 3 0 1 2 3 P, width of fold on brachial valve versus width of first plication lateral to fold; USGS 10c. 2721—SD. Q, pedicle valve width versus ratio of sulcus/first lateral plication width. R, pedicle valve width ver- sus ratio of sulcus width/first interspace width; USGS 10c. 2721—SD. S, brachial valve width versus ratio of fold/first rib; USGS loc. 2720—SD. T, width of valve versus ratio of fold width/first lateral FOLD/FIRST RIB FIGURE 18.—Continued. FOLD WIDTH/FIRST LATERAL PLICATION plication; USGS loc. 2721—SD. (Continued on following page). 01 00 BRACHIAL VALVE WIDTH, IN MILLIMETERS PEDICLE VALVE WIDTH, IN MILLIMETERS PEDICLE VALVE WIDTH, IN MILLIMETEFIS EARLY PALEOZOIC BRACHIOPODS OF THE MOOSE RIVER SYNCLINORIUM, MAINE 30 30 . ' ' E V LLI '— O n ' Lu - g “ u ! : E ' 51 5 I. , g . : 20 :.::.-- 220 . ‘1 " ' z ' . . H J:~ . - - l— o ‘ Q . g 3 ,I . m , 10 . 3 10 < > LIJ _I 9 0 Lu 0 m o 1 2 3 o 1 2 3 FIRST PLICATION WIDTH, IN MILLIMETERS FIRST INTERSPACE WIDTH,‘|N MILLIMETERS 30 30 U) E X '— LU . ' E 3 : ' . . 3 —I . - , a I ' :' ' ' 0 ° E 20 . 5 . E 20 . . I . I~ . . I . I 0 O o '— I I e a o . . E . ; - . g . . 10 ..I 10 ° < > .J S I O < II o m o 1 2 3 o 1 2 3 FIRST PLICATION WIDTH, IN MILLIMETERS WIDTH OF FIRST PLICATION LATERAL T0 FOLD, IN MILLIMETERS 30 0 . Z ,. .II‘, 41-" . 20 ' 5" ‘ ,. I; : 10 ' o o 10 20 1 2 3 4 5 6 7 DIDUCTOR WIDTH, IN MILLIMETEFIS FIGURE 18,—Continued. U, pedicle valve Width versus width of first plication; USGS Ioc. 2720—SD. V, width of pedicle valve versus width of first interspace; USGS 10c. 2721—SD. W, width of brachial valve versus width of first plication; USGS Ioc. 2720-SD. X, width of brachial valve versus width of first plication lateral to fold; USGS 10c. 2721—SD. Y, pedicle valve width versus diductor width; USGS 10c. 2720—SD. Z, Width of pedicle valve versus width of diductor; USGS Ioc. 2721—SD. (Continued on following page). SYSTEMATIC PALEONTOLOGY w a: . l“ 20 FIGURE 18.——Contmued. {3 AA AA, dxiductor length versus diductor width of E pedicle valve; USGS 10c. 2720—SD. BB, (be- :3 low) diductor length versus diductor width on i pedicle valve; USGS 10c. 2721—SD. ;. 10 I— 9 E c: O |_. 8 g 00 1o 20 DIDUCTOR LENGTH, IN MILLIMETERS 8 BB 7 0 6 (I) - . cc LIJ . '— Lu 2 . 3 :' 5 E . g . f . l— 0 . 2 Lu -J 4 u: . O I- . o 3 c E o 3 . 2 1 1 2 3 4 5 6 7 DIDUCTOR WIDTH, IN MILLIMETEFIS 60 EARLY PALEOZOIC BRACHIOPODS OF THE MOOSE RIVER SYNCLINORIUM, MAINE 40 40 A B U) U) 95 35 E 30 E 30 :. E . _l _I g :' 2 E ' ‘. Z ' E I~ g 20 I5 20 fl 3 LU g > _I -‘ . < < ' > > . I.“ ‘3 10 d 10 . — E D 3 g E . o 0o 10 20 30 4o 0 10 20 PEDICLE VALVE WIDTH, IN MILLIMETERS ONE.HALF THE NUMBER OF INTERIOR COSTELLAE U) 0 . s 2 3 2 I: C < V, D . E a. (I . _ (I) LU _I a: I— . :' Lu g E ’5 3 Z _ _I -. 10 —' — 1 I g E 5 . ° w z ' _ ' I; I. D . 5 9 3 ' E 3 3 o u. o w o 1 2 3 o 10 20 30 40 FIRST LATERAL INTERSPACE WIDTH, IN MILLIMETERS PEDICLE VALVE WIDTH, IN MILLIMETERS FIGURE 19.—Measurements of Costellispirifer perimele (Clarke, 1907). Tomhegan Formation, USGS loc. 2750—SD, Brassua Lake quadrangle, Somerset County, Maine. quate comparative material is available to make com- parison possible. “Spirifer” cumberlandiae (Hall, 1857) has a long— er ventral interarea, tends to be orthocline, brachial valve has narrow chilidial plates; these characters are absent in “S.” perimele. Occurrence—USGS loos. 2873—SD, 2750—SD, 2820—SD, 2730—SD, Brassua Lake quadrangle, Som- erset County, Maine. Stratigraphic location.—Tomhegan Formation. Distribution.—Northern Maine, possibly northern New Hampshire (Boucot and Arndt, 1960) and southeastern New York (Boucot, 1959b) . Figured specimens.—USNM 126035, 126037, 126029. Measured specimens.——USNM 125888. Unfigured specimens.—USNM 126580. Subfamily MUCROSPlRll-‘ERINAE Genus MUCROSPIRIFER Gnlnu, 1931 “Mucrospirifer” cf. "M." mun (Hall, 1857) Plate 18, figures 21, 22 Exterior.—Single pedicle valve internal mold available. Transverse, triangular outline, about twice as broad as long. Lateral profile deeply con- vex, with flat catacline interarea, giving shell sub- pyramidal shape. Few poorly preserved radial plica- 126030, SYSTEMATIC PALEONTOLOGY 61 4o 40 40 u: ‘0 m D: 3:; w 3:, E “I: " 30 E 30 E 30 E j j j 2 . E E Z , Z 2 ::~ 21:~ I‘ '5 20 '§ 20 E 20 s E E m LU LLI > .. 5 J 3 < < < > > > LLI W Lu J {j 10 9 10 E’) 10 o . o ' 5 E . E E o 00 1o 20 0 1o 20 0o 10 20 SULCUS WIDTH, IN MILLIMETERS SULCUS WIDTH/FIRST LATERAL PLICATION WIDTH DIDUCTOR WIDTH, IN MILLIMETERS FIGURE 19.—Continued. tions seen impressed on interior. Specimen too poorly preserved to describe further external details. Intermix—Pair thin, platelike dental lamellae are closely spaced and slightly divergent in apex of valve. Occurrene.—USGS loc. County, Maine. Stratigraphic location. —Tomhegan Formation (Lower Devonian). Figured specimen.—USNM 160135. Subfunily COSTISI’IRIFERINAE Temier and Termier, I949 Genus COSTISPIRIFER Cooper, 1942 Costispirifer 51). Plate 18, figures 23—26 Exterior.-—Unequally biconvex shells, pedicle valve having greater degree convexity than brachial valve. Greatest width at straight hinge line. Sub- circular outline. Brachial valve bears low, rounded fold; pedicle valve has corresponding sulcus. Rounded costellae separated by relatively narrow interspaces. Anterior and lateral margins rounded. Anterior commissure uniplicate, crenulate. Pedicle valve interarea gently apsacline, brachial valve in- terarea steeply anacline. Fold and sulcus costellate. Pedicle valve interarea relatively Ion-g, brachial valve interarea steeply anacline. Fold and sulcus costellate. Pedicle valve interarea relatively long, 27 23—SD, Somerset brachial valve interarea short. Delthyrium appears unmodified, includes angle of about 90°. Pedicle valve interior.-—Dental lamellae short, al- most obsolete due to deposition of secondary mate- rial in umbonal cavities. Dental lamellae convex medially, support stubby hinge teeth located on upper margins delthyrium. Muscle field deeply im- pressed, subcircular outline. Muscle field restricted posterior half valve, consists of paired, elongate, median adductor impressions longitudinally striate and laterally bounded by paired, anteriorly expand- ing diductor impressions. Diductor impressions ex- tend posteriorly into pair of pits developed in sec- ondary material lining delthyrial cavity. Anterior and medial to posterior end of diductor impressions is vertical, diamond-shaped, small pedicle callist. Peripheral regions crenulated by impress external ornamentation. Brachial valve interior.—Small ctenophoridium laterally bounded by socket plates which diverge anterolaterally. Basal part of socket plates not sup- ported by crural plates. Interior crenulated by im- press of external ornamentation. Comparison—Maine specimens not specifically identifiable. Much of the material has been badly abraded and the external ornamentation removed. Occurrence.—USGS locs. 3088—SD, 2767—SD, 62 EARLY PALEOZOIC BRACHIOPODS OF THE MOOSE RIVER SYNCLINORIUM, MAINE 3482—SD, Somerset and Piscataquis Counties, Maine. Stratigraphic location.—Lower sandstone of the Tarratine Formation (Lowe-r Devonian) and Seboo- mock Formation (Lower Devonian). Figured specimens.—USNM 126018, 126023. Subhmily KOZLOWSKIELLINAE Genus KOZLOWSKIELLINA Boucot, 1958 Subgenus MEGAKOZLOWSKIELLA Boucot, 1957 Megakozlowskiella :1). Plate 19, figures 1—6 Discassion.-—Poor specimens of Megakozlowski— ella were obtained at several localities in Somerset County; the material is unfit for specific identifica- tion. Specimens from the Beck Pond Limestone have steeply apsacline to catacline pedicle valve interarea and pseudo‘spondylium formed by deposition of sec- ondary material between converging dental lamellae and median septum; in these regards resemble M. cyrtinoides. Occarrence.—USGS Iocs. 3477—SD, 3499—SD, 3601—SD, 3487—SD, 3497—SD, Spencer Lake and Pierce Pond quadrangles, Somerset County, Maine. Stratigraphic location—Beck Pond Limestone (Lower Devonian) and Parker Bog Formation (Lower Devonian). Figured specimens.—USNM 126046, 126044. Family AMBOCOELIIDAE Subfamily AMBOCOELHNAE Genus METAPLASlA Hall and Clarke, 1893 Met-plush cf. M. paucicostatn (Schuchert, 1913) Plate 19, figures 7—11 Exterior.—Unequally convex shells, gently con- vex brachial valve, subconical pedicle valve. Outline subcircular to transversely elliptical, brachial valve more elliptical than pedicle valve. Pedicle valve in- terarea orthocline or gently apsacline or even gently anacline, beak gently incurved. Hinge line straight, greatest width near midlength. Delthyrium appears unmodified, includes angle of about 60° Cardinal margins terebratulid; lateral and anterior margins rounded. Shells appear smooth except for concentric growth lines. Pedicle valve bears very shallow me- dian furrow flanked by pair ill—defined plications; remainder valve smooth. Anterior commissure prob- ably feebly intraplicate. Pedicle valve interior—Inner margins hinge line flanked by stubby hinge teeth. Muscle field deeply impressed into secondary material lining delthyrial cavity. Muscle field longitudinally elongate in form, consists paired, median adductor impressions sepa- rated by myophragm, laterally flanked by pair nar- row diductor impressions. Posterior part delthyrial cavity filled by pad secondary material forming me- dially grooved step, may have served as pedicle callist. Umbonal cavities pitted. Pair subparallel pallial trunks appear to extend from anterior por- tion muscle field almost to anterior margin. Brachial valve interior.—Small bladelike cardinal process laterally flanked by pair elongate crural plates forms pseudocruralium in posterior part of valve. Well-impressed adductor field anterior of car- dinalia is difficult to study on the one specimen avail- able. Valve interior appears smooth. Comparison—Maine specimens appear conspecific with Metaplasia pauciocostata from Shriver Chert; the lack of adequate comparative material makes positive specific assignment ill advised. M. paucicos- tata may be subjective synonym M. pyridata; the lack of verified specimens of latter leaves this point in doubt. Occurrence.—USGS 10c. 2806—SD, Pierce Pond quadrangle, Somerset County, Maine. Stratigraphic location.—McKenney Ponds Mem- ber of Tarratine Formation (Lower Devonian). Figured specimens—USN M 126115A,B,C. Unfigured specimens.—USNM 126258. Metapluin minute Boucol, 1959 Plate 19, figures 12—18 Metaplasia minuta Boucot, 1959a, p. 17—18, pl. 2, figs. 18—24. Exterior.—Unequally biconvex shells, brachial valve gently convex, pedicle valve subconical. Hinge lineline straight, maximum width located about one- third distance anterior from posterior margin. Valves subcircular to transversely elliptical in out- line. Brachial valve bears very low median fold flanked by two pairs very low and distinct costae. Pedicle valve bears low median furrow flanked by one pair prominent costae. Valve surface smooth except for concentric growth lines. Pedicle valve interarea steeply apsacline to catacline, beak in- curved gently. Delthyrium open, includes angle of about 30°. Brachial valve interarea very short, ap- pears orthocline, pedicle valve interarea very long. Anterior commissure faintly intraplicate. Anterior and lateral margins rounded. Pedicle valve interior.—Pair stubby hinge teeth border sides of delthyrium, tracks leave pair of ridges on interior simulating pair short dental lamel— lae. Low myophragm divides weakly impressed mus- cle field into pair elongate impressions. Valve in- terior impressed by plications. Brachial valve interior.—Short, bladelike cardinal process flanked by crural plates reaching floor of valve, form pseudocruralium. Dental sockets floored by socket plates which posteriorly unite with crural plates. Muscle field weakly impressed, consists of SYSTEMATIC PALEONTOLOGY 63 elongate median adductor impression laterally bounded by pair elongate, lateral adductor impres- s10ns. * Comparison—Maine material appears specifically distinct from M. paucioc-ostata which it closely re- sembles. Same form occurs in Camden Chert of Ten- nessee, but collections of US National Museum do not have enough specimens of this species to permit a thorough comparison. Plications in this material are much stronger than in that from the strata of Oriskany age which has been previously assigned to Metaplasia. Occurrence.—USGS 10c. 27 50—SD, 2852—SD, 2820— SD, Brassua Lake quadrangle, Somerset County, Maine. Stratigraphic location.—Tomhegan Formation (Lower Devonian). Figured specimens.—USNM 126086, 126067. Unfigured specimens.—-USNM 126597—126600. Genua PLlCOPLASIA Rental, 1959 Plicoplasia plicala (Weller, 1903) Plate 19, figures 19—29 Plieoplasia plicata (Weller, 1903), see Boucot, 1959a, p. 22, pl. 1, figs. 1-9. Exterior.—Unequally biconvex shells. brachial valve gently convex, pedicle valve subconical. Great- est width at straight hinge line, or at position slightly anterior. Transversely elliptical outline. Brachial Valve bears broad, shallow medial medial sulcus, pedicle valve bears broad, low median fold. Anterior commissure weakly sulcate, crenulate. Pos- terior part lateral margin straight, remainder and anterior margin, rounded. Plications rounded cross section, separated by wide, U-shaped interspaces. One plication situated in sulcus, each flank bears three to four plications including one bounding fold or sulcus laterally. Plications bounding fold larger than remainder present on shell. Pedicle valve inter- area orthocline to gently apsacline, beak straight to gently incurved. Pedicle valve interarea relatively long, brachial valve interarea short. Brachial valve interarea orthocline to gently anacline. Concentric growth lines relatively lamellose anteriorly. Del- thyrium open, includes angle of about 30°. Pedicle valve interior—Stout hinge teeth situated either side hinge line’s median edge. Hinge tooth track buried in secondary material deposited against posterior valve wall. Myophragm bisects muscle field, extends anteriorly to about midlength. Muscle field weakly discernible. Interior crenulated by plica- tion impress. Brachial valve interior.-—Prostrate, posteriorly di- rected, medially grooved, terminally bifid cardinal 126110, process laterally flanked by basomedially converg- ing crural plates whose lower edges are parallel. Dental sockets formed medially by crural plates, an- terolaterally by socket plates which bottom them, posteriorly by interarea. Dental sockets triangular in cross section. Myophragm bisects valve, muscle field unimpressed. Valve interior crenulated by im- press external ornamentation. Comparison.—P. plicata has a shallower sulcus and lower fold than P. cooperi. P. plicata tends to have more lateral costae than P. cooperi. Occurrence.—USGS locs. 2813—SD. 2811—SD, 2798—SD, 2797—SD, 3089—SD, and questionably 3481—SD, 2735—SD, 2731—SD, 3482—SD, 2884—SD, Somerset and Piscataquis Counties, Maine. Stratigraphic location.-Lower sandstone of the Tarratine Formation (Lower Devonian), and the Seboomook Formation (Lower Devonian). Distribution—Eastern North America. Figured specimens.—USNM 126114, 126106, 126256, 126259, 126260. Superfamily CYRTINACEA Family CYRTINIDAE Subfalnily CYRTININAE Genus CYRTINA Davidson. 1858 Cytfina? cf. C. rash-ta (Hall, 1857) Plate 19, figures 30-32 Exterior—Single badly deformed internal mold available. Transverse, subtriangular outline, deeply unequally biconvex lateral profile. Pedicle valve very deep, subpyramidal, with high flat catacline ventral interarea. Narrow median pedicle valve sulcus, cor- responding low brachial valve fold. Numerous rela- tively small radial plications; eight or nine discerni- ble on one valve flank. Pedicle valve interior.—Specimen too poorly pre- served to be certain of internal features; appears to be median groove or line of demarcation that may have accommodated ventral median septum. Brachial valve interior.—Internal structures too poorly preserved to afford description. Interior cor- rugated by impress of radial plications. Occurrence.—USGS loc. 2883-SD. Spencer Lake quadrangle, Somerset County, Maine. Stratigraphic location. — Seboomook Formation (Lower Devonian). Figured specimen.—USNM 160136. Suborder ATHYRIDOIDEA Superfamily ATHYRIDACEA Family MERISTELLIDAE Subfamily MERISTINAE Genus MERISTA Sueas, 1851 Merista cf. M. tennesseensia Hall and Clarke, 1895 Plate 20, figures 1-—6 Exterior.——Subequally convex shells with longi- tudinally elongate, elliptical outline. Maximum 126087, 64 EARLY PALEOZOIC BRACHIOPODS OF THE MOOSE RIVER SYNCLINORIUM, MAINE width near midlength. Cardinal margins pedicle valve terebratulid, beak incurved over posterior end brachial valve. Lateral and anterior margins rounded. Brachial valve bears low broad fold, origi- nating about midlength, pedicle valve bears broad shallow sulcus originating in anterior quarter of shell. Anterior commissure uniplicate. Concentric growth lines. Pedicle valve interior.—Well-developed shoe-lifter process. Short dental lamellae bound delthyrial cavity. Brachial valve interior.——Short septalium sup- ported by median septum which extends anteriorly to about midlength. Comparison—Maine material resembles M. ten- nesseensis in external form but is inadequate for a positive specific identification. Occurrence.—USGS loc. 3488—SD, Spencer Lake quadrangle, Somerset County, Maine. Stratigraphic location—Hardwood Mountain For- mation (Upper Silurian). Figured specimens.——-USN M 126105, 126085. subfamily MERISTELLINAE Genus MERISTELLA Hall, 1860 Merislelln Int: (Hall, 1859) Plate 20, figures 7—13 Meristella lata (Hall, 1859'), see Clarke, 1900 p. 45. Exterior.—Unequally convex shells, brachial valve having greater degree convexity than pedicle valve. Shell outline subcircular to longitudinally elliptical. Pedicle valve cardinal margins terebratuloid. Lateral, anterior margins rounded. Brachial valve bears low fold originating near midlength. The greatest width located slightly posterior to midlength. Pedicle valve bears broad shallow sulcus. Anterior commissure uniplicate. Concentric growth lines. Pedicle valve interior—Short, obsolete dental lamellae border delthyrial cavity, support stout hinge teeth. Umbonal cavities filled with secondary mate- rial. Muscle field deeply impressed into secondary material lining posterior part of valve. Muscle field flabellate, outline subtriangular, extends anteriorly to about three-quarters distance to anterior margin, about half as Wide as valve. Muscle field longitudi- nally striate. Posterior part delthyrial cavity filled with secondary material into which is indented a spherical chamber serving as pedicle callist. Brachial valve interior—Cardinal plate appears formed from medially conjunct hinge plates, basally supported by medial septum which extends anteri- orly three-quarters distance to anterior margin. Median part cardinal plate depressed. Spires later- ally directed, nature of jugum not ascertained. Spire contains about 13 whorls. Dental sockets anteriorly directed, partly roofed over by projecting edges of cardinal plate. Muscle field unimpressed. Occurrence.—USGS locs. 2810—SD, 2806—SD, 3089—SD, 2732—SD, 2731—SD, 2722—SD, 2813—SD, 2776—SD, 3482—SD, 2720—SD, 2777—SD, 2797—SD, 2719—SD, 2847—SD, 2760—SD, 2798—SD, 2767—SD, 2701—SD, 2884—SD, 2770—SD, 2729-SD, 2861—SD, 2721—SD, 2771—SD, Somerset and Piscataquis Coun- ties, Maine. Stratigraphic location.—-McKevnney Ponds Mem- ber and upper and lower sandstones of the Tarra- tine Formation (Lower Devonian), Seboomook For- mation (Lower Devonian). Figured specimens.—USNM 126103, 126116A, 126079, 126082, 126261, 127390. Unfigared specimens.—USNM 126104, 126469, 126581—126596. Meristellal sp. Discussion—Inadequate material of a rostro- spiroid that resembles Meristella was obtained at several localities in strata of pre-Oriskany age. Occarrence.—Localities 3499—SD, 3487—SD, 3477— SD, Pierce Pond and Spencer Lake quadrangles, Somerset County, Maine. Stratigraphic location—Parker Bog Formation (Lower Devonian) and Beck Pond Limestone (Lower Devonian). Unfigured specimens.—USNM 126569—126577. Genus CHARIONOIDES Boucol, Johnson and Staton, 1964 Charionoides doris (Hall, 1860) Plate 20, figures 14-22 Charionoides doris (Hall, 1860), see Boucot, Johnson, and Staton, 1964, p. 817—818, pl. 127, figs. 14-20. Exterior.—S-ubequally biconvex shells, pedicle valve more convex than gently convex brachial valve. Almond-shaped shell outline, maximum width about two-thirds distance from posterior. Cardinal mar- gins terebratulid; palintrope elongate, slightly con- cave. Ventral beak very attenuate. Anterior and lateral margins rounded. Concentric growth lines. Neither fold nor sulcus present. Delthyrium appears closed by pair small plates, except for small meso- thyrid foramen. Pedicle valve interior.—Short dental lamellae laterally bound delthyrial cavity. Muscle field weakly impressed in small specimens, in large speci- mens deeply impressed and fiabellate. Muscle field subtriangular impression deeply impressed posteri- orly, probably represents diductor impressions, pos- terior of which are small elongate adductor impres- sions. Pedicle callist located in posterior part delthyrial cavity. Brachial valve interior.—-Sessile cruralium formed SYSTEMATIC PALEONTOLOGY 65 from medially conjunct hinge plates supported by low median septum. Brachiophores extend from upper parts hinge plates. Median septum bisects cruralium base as low ridge. Occurrence.—USGS locs. 2750—SD, 2820—SD, Brassua Lake quadrangle, Somerset County, Maine. Stratigraphic location. ——Tomhegan Formation (Lower Devonian). Figured specimens. — USNM 126111B, 126092, 126102, 126081, 126080A, 126117. Family NUCLEOSPIRIDAE Genus NUCLEOSPIRA Hall, 1859 Nucleospin up. Plate 20, figures 23—27 Exterior.—Unequally convex shells, pedicle valve having greater degree convexity than brachial valve. Shell outline subcircular, greatest width near mid- length. Cardinal margins pedicle valve terebratulid. Pedicle valve bears shallow, narrow sulcus. Concen- tric growth lines. Lateral and anterior margins rounded. Peripheral region bears series of fine spines. Delthyrium unmodified, includes angle of about 90°. Pedicle valve interior.—Short tooth tracks border delthyrial cavity, support small, stubby hinge teeth. Low median septum extends anteriorly to about valve midlength. Muscle field unimpressed. Anterior of valve smooth. Brachial valve interior.—Platelike cardinal proc- ess posteriorly reflexed, supported anteriorly by short median septum. Narrow adductor impressions either side median septum. Anterior of valve is smooth. Comparison—Material is inadequate for specific identification. Occurrence.—USGS loc. 3479—SD, Spencer Lake quadrangle, Somerset County, Maine. Stratigraphic location.—Base of Hobb-stown For- mation (Upper Silurian). Figured specimens.—USNM 126091, 126262. Family ATHYRIDIDAE Subhmily ATHYRIDINAE Genus PROTATHYRIS Kozlowski, 1929 Prolathyris up. Plate 20, figures 28—37; figure 20 Exterior.—Shells biconvex, pedicle valve having greater degree convexity. Shell outline subcircular transversely elliptical. Hinge line curved, much shorter than maximum width located near mid- length. Pedicle valve beak gently incurved, inter- area steeply apsacline to catacline. Pedicle valve interarea relatively long, brachial valve interarea short. Brachial valve bears broad low fold, pedicle valve has corresponding sulcus. Anterior commis- sure uniplicate. Concentric growth lines. Pedicle valve interior.——Short dental lamellae lat- erally bound delthyrial cavity (fig. 20), posteriorly support bladelike, medially directed hinge teeth. Muscle field weakly impressed, but appears to con- sist of elongate muscle field medially divided by low broad median septum. Delthyrium not modified. Brachial valve interior.—Cardinal plate formed from medially conjunct hinge plates, medial trough. Brachiophores attached to lateral portions cardinal plate. Cardinal plate laterally overlaps anteriorly directed dental sockets. Spire present. Comparison.—Adequate material is unavailable for making specific assignment. The chief differ— ences (with the possible exception of the spiralia and jugum, which were not studied) between the related genera Greenfieldia and Protathyris are the relative degree of valve convexity and the inclina- tion of the ventral interarea. Protathyris has a steep- ly inclined interarea which may be catacline; Green- fieldia has a gently apsacline to orthocline ventral interarea. Protathyris has a more inflated and gib- bous aspect than does Greenfieldia. Occurrence.—USGS 10c. 2822—SD, Attean quad- rangle, Somerset County, Maine. Stratigraphic location. — Hardwood Mountain Formation (Upper Silurian). Figured specimens.—USNM 126070, 126262. Unfigured specimens.—USNM 126059, 126071. Suborder TEREBRATULOIDEA Snperfamily TEREBRATULACEA Family CENTRONELLIDAE Subfamily RENSSELAERIINAE Genus NANOTHYRIS Cloud, 1942 Nanolllyris hodgei Boucol, u. up. Plate 21, figures 1—6 Exterior—Shell biconvex, brachial valve gently convex, pedicle valve more strongly convex, slightly subcarinate form. Shells small. Valves longitudinally elongate, subcircular to elliptical outline. Greatest width located near midlength. Maximum thickness located about one-third to one-half distance anterior from beak. Anterior margin rectimarginate, faintly crenulate. Cardinal margins subterebratulid. Hinge line straight. Pedicle valve beak erect to suberect. Planarea one-half to two-thirds as wide as maximum width. Umbonal regions smooth, peripheral regions faintly costellate. Costellae low, rounded cross sec- tion separated by rounded interspaces. Shell thin. Pedicle valve interior—Dental lamellae thin, ex- tend anteriorly about one-fifth maximum length. Muscle field feebly impressed. Muscle field appears to consist of elongate diductor field medially enclos- ing small, elliptical adductor impression. Brachial valve interior.—Concave cardinal plate posteriorly perforate, free, supported basally by dis- 66 EARLY PALEOZOIC BRACHIOPODS OF THE MOOSE RIVER SYNCLINORIUM, MAINE crete crural plates. Dental sockets narrow, short, make angle of about 45° with hinge line. Muscle field weakly impressed, medially divided by low myo- phragm. Narrow, elongate, paired lateral adductors slightly impressed posteriorly, enclose medial ad- ductor. Measurements. — Relationships among length, width, and thickness are linear (fig. 21) with a rela- tively high degree of dispersion, reflecting the lim- ited size range of the sample. Comparison—N. hodgei is distinct from N. mu- tabilis. The former has a weaker impress of the costellae, less terebratulid outline of cardinal margin, and less gibbous lateral aspect. N. hodgei differs from N. subglobosa in its less gibbous lateral aspect much less strongly impressed ornamentation, less terebratulid cardinal margins, and more erect beak. N. reesidei has more strongly impressed ornamenta- tion, more terebratulid cardinal margins, and a slightly more gibbous profile. N. hodgei differs from N. subglobosa crassa and N. subglobosa auus in its less well impressed radial ornamentation; its out- line less circular than that of N. subglobosa crassa. Occurrence. — USGS locs. 2832-SD, 2890—SD, 2721—SD, 3090—SD, Somerset County, Maine. Stratigraphic location—Tarratine Formation. Distribution—Somerset County, Maine. Holotype.—-USNM 126194. Measured paratypes.—USNM 125940. Figured paratypes. — USNM 126191, 126183, 125943. Unfigured paratypes.—USNM 125941, 125942, 125944, 125945. Nanolliyris cf. N. subglobosa (Weller, 1903) Plate 21, figures 7—13 Exterior—Shell biconvex, gently convex brachial valve, more highly convex, carinate pedicle valve. Pedicle valve about 11/2 times as deep as brachial valve. Shells elongate, elliptical outline. Maximum width situated near midlength or slightly posterior. Lateral and anterior margins rounded. Anterior mar- gin rectimarginate, weakly crenulate. Weak costellae distributed about periphery, absent in umbonal re- gions. Cardinal margins subterebratulid. Hinge line short, rounded. Pedicle valve beak appears suberect. Pedicle valve interior—Dental lamellae thin, short. Muscle field very weakly impressed. Brachial valve interior—Serial sections demon- strate presence of cardinal plate supported by crural plates, posteriorly perforate. 126220, Comparison—These specimens most closely re- semble N. subglobosa but the deformed state of avail- able material precludes positive specific identifica- tion. Occurrence—USGS locs. 3601—SD, 3600-SD, Spencer Lake quadrangle, Somerset County, Maine. Stratigraphic location—Beck Pond Formation (Lower Devonian). Figured specimens—USNM 126187, 126192 A, B. can... RENSSELAERIA Hall, 1859 Renssellerin sp. Plate 21, figures 14—17 Exterior.—Fragmentary exteriors present are badly abraded. Pedicle foramen appears submeso— thyrid- Cardinal margins terebratulid. Pedicle valve beak nearly straight. Pedicle valve more strongly convex than brachial valve. Lateral margins rounded, anterior margins appear rounded. Pedicle valve interior.—Dental lamellae short, al- most entirely obsolete due to deposition secondary material in umbonal cavities. Hinge teeth bladelike, inclined basally towards midline. Hinge teeth fused basally with dental lamellae. Muscle field deeply im- pressed, consists of elongate diductor field with subparallel sides, posteriorly divided by myophragm along which were probably very narrow adductor impressions. Secondary material almost fills pos- terior part of delthyrial cavity, except for deeply impressed muscle field. Upper surface of secondary material bears pedicle callist. Brachial valve interior. —— Thickened cardinal plate, posteriorly perforate, supported by pair crural plates. Dental sockets narrow, laterally directed. Muscle field weakly impressed, consists of two pairs elongate subparallel adductor impressions. Low myo- phragm medially divides muscle field. Comparisou.——Maine material too poor to be spe- cifically identified. Discussion.—In regard to the stratigraphic range of the genus, Cloud (1942, p. 47) assigned “Atrypa” aequiradiata Conrad, 1842, from the Becraft Lime- stone, to Nanothyris. Reexamination shows it be- longs to Rensselaeria s. s. because of its radially ornamented umbones and internal morphology which differ in no important respect from those of the other species assigned to Rensselaeria by Cloud (1942), p. 55, 56) . Material assigned by Boucot (in Woodard, 1951, p. 76) to “Beachia cf. thum’i” belongs to Rensselaeria sp. FIGURE 20.—Serial sections of specimen of Protathyris sp. (X 4.5). Hardwood Mountain Formation. USGS 10c. 2822—SD, Attean quadrangle, Somerset County, Maine. Numbers are measurements in millimeters from posterior end of shell. 68 WIDTH, IN MILLIMETERS LENGTH, IN MILLIMETERS FIGURE 21.—Measurements of pedicle valves of Nanothyris hodgei Boucot, n. sp. Lower sandstone of Tar- ratine Formation. USGS 10c. 3090—SD, Long Pond quadrangle, Somerset, Maine. 20 _| O 20 10 EARLY PALEOZOIC BRACHIOPODS OF THE MOOSE RIVER SYNCLINORIUM, MAINE LENGTH, IN MILLIMETERS 20 10 3 THICKNESS, IN MILLIMETERS 10 20 WIDTH, IN MILLIMETEFIS 30 3 THICKNESS, IN MILLIMETERS SYSTEMATIC PALEONTOLOGY 69 Occurrence.——USGS locs. Somerset County, Maine. Stratigraphic lo-cation.—Lower sandstone of the Tarratine Formation (Lower Devonian). Figured specimens.——USNM 126216, 126214. Unfigured specimens—USNM 126215, 126213. 2767—SD, 2729—SD, Subfnmily EURYTHYRIDINAE Genus BEACHIA Hall and Clarke, 1893 Benchin lhunii (Clarke, 1907) Plate 21, figures 18—30; plate 22, figures 1—5 Megalanteris thunei (Clarke, 1907), p. 168, pl. 26, figs. 8—13; pl. 27, figs. 7—15. Exterior.—Shell subequally biconvex, subcarinate pedicle valve having somewhat greater degree con- vexity than gently convex brachial valve. Shell sub- circular. Pedicle foramen mesothyrid. Small delti- dial plates medially conjunct. Anterior commissure rectimarginate, feebly crenulate in some specimens. Lateral and anterior margins rounded. Cardinal mar- gin subterebratulid. Concentric growth lines, some- times faintly impressed costellae, particularly pe- ripherally. Pedicle valve beak suberect. Maximum width located at midlength. Lateral margins may be introverted, but on most specimens this feature not evident. Pedicle valve interior.—Dental lamellae short, he- come almost entirely obsolete with increase in speci- men size, due to deposition secondary material in umbonal cavities. Anterior part dental lamellae bears bladelike hinge teeth. Inner face each hinge tooth longitudinally grooved to receive lateral margin of cardinal plate. Posterior part delthyrial cavity filled with secondary material except for site of pedicle callist. Diductor field deeply impressed, narrow and straight-sided in form, divided medially in posterior part of valve by low myophragm. Myophragm bor- dered by pair narrow adductor impressions. Anterior of myophragm is elliptical anterior diductor im- pression. Interior of valve smooth. Brachial valve interior.——Cardinalia of small shells consist of discrete hinge plates united antero- medially by narrow band (posterior of band is large, triangular foramen). Under hinge plates are sup- porting crural plates. Muscle field small shells bare- ly impressed, although myophragm prominent. This stage brachial valve similar to Nanothyris. In large specimens cardinalia consist of cardinal plate bear- ing ponderous, moundlike thickening which serves as cardinal process. Foramen in cardinal plate closed off as size of cardinal process increases, at same time crural plates become almost entirely submerged in secondary material. Dental sockets deep, antero- laterally directed. Muscle field consists of posteriorly deeply impressed adductor impression divided me- dially by myophragm. Posteromedial pair adductor impressions can be distinguished from pair of elon- gate lateral impressions in some shells. Pallial marks extend toward anterior margin from anterior mar- gins of muscle field, subparallel to midline. Brachio- phores bladelike. Comparison.—B. thunii does not possess the prominent, introverted lateral margins of B. sues- sana, as judged from material studied, in addition the latter species seems to possess more consistent- ly costellate ornamentation. Cardinal process of B. thunii is commonly swollen and enlarged whereas in B. suessana the cardinal plate is not commonly great- ly thickened. Specimen ques-tionably referred to Meganteris by Cloud (1942, p. 109-110) belongs to Beachia thum'i. Occurrence.—USGS locs. 2813—SD, 2700—SD, 2701—SD, 2731—SD, 2729—SD, 2732—SD, 2751—SD, 2792—SD, 2769—SD, 2767—SD, 2776—SD, 2798—SD, 2803-SD, 2812-SD, 2771—SD, 2777-SD, 2725—SD, 2830—SD, 2761—SD, 2810—SD, 3482—SD, 3486—SD, 2806—SD, 2872—SD, 2760-SD, 2870—SD, 3474—SD, 2864—SD, 2796—SD, 2884—SD, Somerset and Pisca- taquis Counties, Maine. Stratigraphic l0cation.—-Seboomook Formation (Lower Devonian), and McKenney Ponds Member and lower sandstone of the Tarratine Formation (Lower Devonian). Distribution.—-—Eastern North America. Figured specimens. — USNM 126198-126200, 126202—126207, 126209, 127217, and New York State Museum 8436. Unfigured specimens—USNM 126177, 126605- 126608, 126610—126634. Genus CLOUDOTHYRIS Boucot and Johnson, 1968 Cloudotllyri: poslovalis Boucot and Johnson, 1968 Plate 22, figures 6—13 Cloudothyris postovalis Boucot and Johnson, 1968, p. B19— B20, pl. 7, figs. 26—42. Exterior.—Shell subequally biconvex, pedicle valve slightly more convex than brachial valve. Shell out- line elongate, elliptical. Lateral margins introverted, defining long lateral reentrants which merge pos- teriorly with planareas. Anterior commissure recti- marginate. Pedicle valve beak short, suberect. Pedi- cle valve foramen mesothyrid. Deltidial plates, prob- ably present, appear' conjunct, small. Beak ridges sharp, define prominent planareas which extend width of posterior margin. Maximum width located about one-third distance anterior from beak. Con- centric growth lines. Pedicle valve interiorw—Dental lamellae almost 7O EARLY PALEOZOIC BRACHIOPODS OF THE MOOSE RIVER SYNCLINORIUM, MAINE obsolete in large specimens due to deposition sec- ondary material in umbonal cavities. Dental lamellae short. Hinge teeth bladelike, supported by dental lamellae. Muscle field elongate, one-quarter to one- fifth as wide, and half as long as valve. Long, nar- row diductors posteriorly and laterally surround small, elongate, median adductor scar; diductors laterally bounded by elongate, pedicle valve adjustor impressions. Myophragm bisects posterior part mus- cle field, extends anteriorly to adductor impression. Pedicle callist situated against posterior wall del- thyrial cavity. Brachial valve interior—Cardinal plate thick, completely sessile, imperforate, supported by crural plates almost entirely submerged in secondary ma- terial which fills in space under cardinal plate. An- terior portion each crural plate recognizable in small specimens. Rising vertically from cardinal plate is elongate, terminally bifid cardinal process. Two limbs of cardinal process taper to anteriorly directed tips. Posterior face each limb indented by slits which diverge laterally from base of cardinal process. Lat- eral sides of slits vertically grooved. Cardinal proc- ess of small specimens consists of low, bifid, stump- like structure. Cardinal process covers median half cardinal plate, leaving only lateral parts exposed. Dental sockets located at posterior base cardinal plate, open posteriorly in large specimens, closed posteromedially by convergent outer socket ridges of small specimens. Dental sockets shallow anteri- orly, extend medially well under posterior edge of cardinal plate. Muscle field elongate, one-fifth as wide and one-quarter to one-fifth as long as valve. Posterolateral adductor scars sublunate to elongate, deeply impressed posteriorly. Anterior extension posterolateral adductor scars bounds ovate, antero- medial adductor scars. Myophragm well developed in large specimens, extends posteriorly from antero— medial adductor impression to anterior face of car- dinal plate. Valve interior smooth. Occurrence.——USGS locs. 2750—SD, 2814—SD, 2820—SD, 3238-SD, Brassua Lake quadrangle, Som- erset County, Maine. Stratigraphic location.—Tomhegan Formation. Distribution—Somerset County, Maine; western New York; and possibly northern New Hampshire (Boucot and Arndt, 1960). Holotype.—USNM 147299 (Boucot and Johnson, 1968, pl. 7, figs. 28—33) is herein designated the holotype. Figured specimens.——USNM 126195, 126210A, 126208, 126218. Unfigured specimens.—USNM 126635—126637. Subfunily AMPHlGENllNAE Genus AMPHIGENIA Hall, 1867 Amphigeuia parva Clarke, 1907 Plate 22, figures 14—24 Exterior—Shell unequally biconvex, naviculate pedicle valve about twice as deep as gently convex brachial valve. Shell outline subcircular to elongate, elliptical. Maximum width located one-third to one- half distance anterior from beak. Anterior commis- sure rectimarginate. Lateral margins straight to rounded, anterior margin rounded. Hinge line short, rounded. Concentric growth lines, some specimens, faint costellae. Beak region characters similar those described by Cloud (1942, p. 78, 79) for A. elongata. Pedicle valve interior—Dental lamellae short, medially convergent to form spondylium basally sup- ported by median septum. Median septum extends to about midlength. Stubby hinge teeth located on posterior margins spondylium. Posteriorly spondyl- ium is laterally buttressed by mystrochial plates. Excavation for pedicle callist located at posterior end delthyrial cavity. Myophragm bisects bottom of spondylium. Brachial valve interior—Cardinal plate possesses foramen posteriorly, basally supported by pair crural plates. Cloud (1942, p. 78) mentioned pres- ence of crural plates converging medially about myophragm to form structure resembling cruralium. A. curta shows this structure actually formed by deposition of secondary material between crural plates to raise tube connecting foramen from base of shell. This character is one associated with in- crease in size rather than taxonomic position. Suite of specimens of A. parva shows stages having every gradation from ones with no secondary material on valve floor to those with tube elevated and cruralium appearing to be present. Narrow pair posterolateral adductor impressions laterally border elongate me- dial adductor impressions, bisected by myophragm which extends under cardinal plate. Dental sockets laterally directed, posteriorly border cardinal plate. Measurements—Relations between length and width of both pedicle and brachial valves are al- most random due to mechanical deformation (figs. 22A, D), plus the fact that the anterior portions of some specimens may have been broken off, plus the indication that this species is exception-ally variable, some specimens having a subcircular outline and others being very elongate. Relation between length of median septum and width and length of pedicle valve is linear with a high degree of dispersion (figs. 228, C) , which suggests that length and width meas- urements are very variable due to both biologic and mechanical factors. 71 SYSTEMATIC PALEONTOLOGY .3550 uwmuufiow O? 65.3% 63:23:”. 33A 553an .lecmrm .03 wwwb .uoBaEuoh saga—non. .roafi ~3230 ease.“ S§n3a§vq no madmfivhamaosllfim "EDGE mmwFMEZIEZ Z_ ~IFOZm4 m>n_<> n_<_IO.._<> MADE—ma O? on ON or on O? on wmwkm§.J.:_>_ Z_ .IPOZMJ EDmem ON 0.. wmwkm_>:n_u:_2 Z_ .IFOZMJ w>4<> mu.0_n_wn_ Om ON O_. or ON on SHEILEINH'IIW NI 'HLCIIM EAWVA 3'10ICIEd OF ON SHSLEWI'I‘IIW NI 'HLDNEI‘I wnidEIS on Or ON ON SHELBWI'IWIW NI 'HLCIIM EA'IVA BWOICIEId 72 EARLY PALEOZOIC BRACHIOPODS OF THE MOOSE RIVER SYNCLINORIUM, MAINE Comparison.—Amphigenia parva is most similar to A. elongata, but the greater size reached by many specimens of the latter indicates that size may be a valid criterion for discriminating between members of the two species. From a collection of several hundred valves of A. parva, none was observed to reach the dimensions common to specimens of A. elongata, that is, as much as 10 cm in length. A. carta shares with A. parva the habit of not reaching more than about 3 cm maximum length, but A. carta has an almost equally biconvex form; brachial valve almost as convex as pedicle valve, whereas in A. parva brachial valve is relatively flat as contrasted with deeply earinate pedicle valve. A. chickasawensis has relatively flat valves in contrast to other species of the genus. Occurrence.—USGS loos. 2750—SD, 2852—SD, 2820—SD, 2873—SD, 2839—SD, 2814—SD, 2752—SD, 2723 ?-SD, Brassua Lake quadrangle, Somerset County, Maine. Stratigraphic location. — Tomhegan Formation (Lower Devonian). Distribution.—Somerset County, Maine, and northwestern New Hampshire (Boucot and Arndt, 1960). Figured specimens. — USNM 126197, 126196, 126193, 126188, 126219, 126190, 127389. Measured specimens—USNM 126603A—126603AI, 126604A—126604AP. Unmeasured specimens—USNM 125995—125999, 126601. Family RHIPIDOTHYRIDIDAE Subfamily GLOBITHYRIDINAE Genus GLOBITHYRIS Cloud, 1942 Globitllyrls cullidu (Clarke, 1907) Plate 23, figures 1—5 Globithyris calli-da (Clark, 1907), see Cloud, 1942, p. 84, pl. 11, figs. 5—8, 10—14. Exterior.—Shell biconvex, deeply inflated pedicle valve more convex than brachial valve. Valve out- lines subcircular; brachial valve tending to be trans- versely elliptical, pedicle valve longitudinally ellipti- cal. Anterior and lateral margins rounded. Anterior commissure rectimarginate, crenulate. Pedicle valve foramen hypothyrid, deltidial plates medially con- junct. Cardinal margins terebratulid. Pedicle valve beak incurved. Costae having round cross section; interspaces relatively narrow, V-shaped form. Faint growth lines cross concentric ornamentation. Twenty-six to forty-three costae, thirty to thirty- eight more common. Maximum width located near midlength. Pedicle valve interior—Short, subparallel dental lamellae border delthyrial cavity. Muscle field not discernible. Brachial valve interior.—Discrete hinge plates border cruralium formed from basally fused crural plates in manner analogous to Amphigenia, forming median septum. Cruralium very small, septum ex- tends to about mid-length. Dental sockets shallow, posteriorly border hinge plates. Brachiopvhores ex- tend out from anterolateral edges of hinge plates. Measurements.—The relation between length and 30 30 A B m E. (D u: E E 2 [LI 1 E 20 -..I i 20 :1 .- E :l I'\ Z 2 ‘. I 5 5 I 5 10 “Z, 10 Z _l “J :1 J Lu I w 0 0 0 10 20 30 0 WIDTH, IN MlLLIMETEFlS 10 20 3O 40 50 NUMBER OF PLICATIONS FIGURE 23.—Measurements of Globithyris calliala (Clarke, 1907). Tomhegan Formation, USGS 10c. 2764—SD, Brassua Lake quadrangle, Somerset County, Maine. A, width versus length of shells. B, shell length versus number of plica— tions. SYSTEMATIC PALEONTOLOGY 73 width for both Globithym's callida and G. diania is essentially linear, G. callida shows a much higher degree of dispersion due to mechanical deformation of the specimens (figs. 23A and 24A). G. callida has a larger number of plications, relatively independent of size, than G. diania. G. callida has 30—40 (fig. 23B) with mode about 35, whereas G. diania has about 15—30 (fig. 243) with mode about 23. There is an overlap in the number of plications, but the modes are very distinct from each other. Cornparison.——Globithyris callida is distinguished from G. diania by latter’s coarse ornamentation. Ornamentation types overlap so that intermediate specimens cannot be assigned specifically unless they are part of a population. Remarks.—It is notable that both Globithyris and the closely related genus Rhenorensselaeria have both a coarse-ribbed and a fine-ribbed species oc- curing at the same stratigraphic horizon. It is possi- ble that these differences in ornamentation may reflect some sexual or environmental rather than taxonomic control. Occurrence.—USGS 2736—SD, 2716—SD, 2714—SD, locs. 2764—SD, 2866—SD, 2791—SD, 2757—SD, 2772-SD, 2690—SD, 2828—SD, 3086—SD, 2859-SD, 2835-SD, 2854—SD, 2727—SD, 2719—SD, 2747-SD, 2744—SD, 2838—SD, 2713—SD, 2714-SD, 3087—SD, Somerset and Pisca- taquis Counties, Maine; Globithy’r‘is cf. G. callida, USGS loos. 2850—SD, 2833—SD, 2863—SD, Somerset County, Maine. Stratigraphic location. — Tarratine Formation (Lower Devonian), and Tomhegan Formation (Lower Devonian). Distribution—Northern Maine. Figured specimens. — USNM 126165, 67778, 126180. Measured specimens. — USNM 125946A— 125946BP. Unfigared 125977. 2840—SD, 2874—SD, 3085—SD, specimens.—USNM 125947—125972, Globitllyris diania (Clarke, 1907) Plate 23, figures 6—13 Globithyris diania (Clarke, 1907). See Cloud, 1942, p. 83. Remarks.—Globithyris diania is identical to G. callida in all regards except number of costae, as discussed under G. callida. Occurrence—USGS loos. 2869—SD, 2713-SD, 2855—SD, 2867—SD, 2755—SD, 2868—SD, 2851—SD, 2758—SD, 2759—SD, 2715—SD, 2756—SD, 2763—SD, and G. cf. G. diania, 2762—SD, 2755—SD, Somerset County, Maine. Stratigraphic location. —Tarratine Formation (Lower Devonian) and Tomhegan Formation (Lower Devonian) . Distribution—Northern Maine. Figured specimens—USN M 126164, 126182. Unfignred specimens—USNM 125978—125993. Measured specimens.—USNM 125994A— 125994AK. 126167, Globithyria up. Remarks—A few localities have yielded glo- bithyrids too poorly preserved to be assigned specifically. Occurrence—USGS locs. 2878-SD, 2820—SD, 2754—SD, 2892—SD, 2750—SD, 2752—SD, 2724—SD, 2818—SD, 2836—SD, 2816—SD, 2817—SD, Somerset County, Maine. Stratigraphic (Lower Devonian), (Lower Devonian). Unfigured specimens—USNM 125973-125976. location. — Tarratine Formation and Tomhegan Formation Family MUTATIONELLIDAE Cloud, 1942 Subfamily MUTATIONELLINAE Cloud, 1942 Genus MUTATIONELLA Kozlowski, 1929 Mutationella parlinensis Boucot, Caster, lves, and Talent, 1963 Plate 23, figures 14-26 Mutationella parlinensis, Boucot, Caster, Ives, and Talent, 1963, p. 110—113, pl. 34, figs. 6—16; pl. 35, figs. 1—2. Exterior.—Unequally biconvex, brachial valve gently convex, pedicle valve more inflated. Outline subcircular to longitudinally elliptical. Greatest width and thickness usually located near midlength. Anterior commissure rectimarginiate, crenulate. An- terior and lateral margins rounded. Cardinal mar- gins su'bmegathyrid. Pedicle foramen submesothyrid. Large shells possess short planareas. Deltidial plates discrete. Pedicle valve beak erect to suberect in large specimens, tends to become slightly incurved over brachial valve. Shell punctate. Costellae in- crease in width peripherally. Costellae have broad, rounded cross sections separated by narrow, V- shaped interspaces. Usually 40-60 costellae on each valve. Pedicle valve interior.—Shell thin in umbonal region, no thickening except in delthyrial cavity of large specimens. Dental lamellae short, thin. Muscle field indistinct, elongate in form, extends about one- third distance to anterior margin. Consists of nar- row, median diductor laterally bounded by elongate lateral diductor impressions. Posterior portion del— thyrial cavity occupied by pedicle callist. Pedicle foramen much narrower than chamber occupied by pedicle callist. Hinge teeth stubby, basally supported by dental lamellae. Interior crenulated by impress 74 EARLY PALEOZOIC BRACHIOPODS OF THE MOOSE RIVER SYNCLINORIUM, MAINE 30 20 10 WIDTH, IN MILLIMETERS o 10 20 30 4o LENGTH, IN MILLIMETERS 30 2O 10 NUMBER OF PLICATIONS O 10 20 30 SHELL WIDTH, IN MILLIMETERS external ornamentation except in umbonal region, which tends to be smooth. Brachial valve interview—Discrete hinge plates. A few specimens possess narrow band joining anterior portions of hinge plates. In majority of specimens narrow band not present, because it broke during burial or because it was never formed. Dental sock- ets elongate, laterally directed; closed over postero— medially by outer socket ridges but widen laterally and are free. Muscle field weakly impressed in most specimens, ellipsoidal outline, one-fifth as wide and one-third as long as valve. Large, ovate, lateral ad- ductors enclose pair narrow, median adductors. Median adductors widen rapidly at anterior end of muscle field. Muscle field bisected by myophragm. Measurements—The relation between length and width of brachial valves is linear, with a low degree 30 M O NUMBER OF PLICATIONS ES O ,, __ 0 1 0 2O 30 40 SHELL LENGTH, IN MILLIMETERS FIGURE 24.—-—Measurements of Globithyris diam'a (Clarke, 1907). Tomhegan Formation, USGS 10c. 2869—SD, Bras- sua Lake quadrangle, Somerset County, Maine. A, length versus width of shells. B, length of shell versus number of plications. 0, width of shell versus number of plica- tions. of dispersion (fig. 25A). Number of plications in- creases with increase in size, with a range of 35-56 recorded (figs. 253, C) . Increase in number of plica- tions is linear, with a high degree of dispersion. Comparison—M. parlmensis is closely allied morphologically to the type species in all regards except that the former attains dimensions about three times those in the latter. Large specimens of M. parlinensis have a more carinate pedicle valve than do large specimens of M. podolica, but this may be a function of difference in size. Curvature of beak is similar to that in M. podolica and is not strongly incurved as in Cloudella stewarti. This species possesses short dental lamellae, as does the type species. Cardinalia in all respects are similar to those of the type species, even in evanescent char- acter of the band connecting the hinge plates. BRACHIAL VALVE WIDTH, IN MILLIMETERS BRACHIAL VALVE LENGTH, IN MILLIMETERS 30 20 10 30 20 10 BRACHIAL VALVE LENGTH, IN MILLIMETERS SYSTEMATIC PALEONTOLOGY 30 20 10 0 BRACHIAL VALVE WIDTH, IN MILLIMETERS 0 1O 20 3O 10 20 30 40 60 NUMBER OF PLICATIONS 10 20 30 4O 60 NUMBER OF PLICATIONS FIGURE 25.—Measurements of Mutationella parlinensis Boucot, Caster, Ives, and Talent, 1963. Lower sandstone of Tarratine Formation, USGS 10c. 2718—SD, Long Pond quadrangle, Somerset County, Maine. 75 76 EARLY PALEOZOIC BRACHIOPODS OF THE MOOSE RIVER SYNCLINORIUM, MAINE Weakly impressed musculature is similar to that of the type species, as is the thin shell in umbonal region of the pedicle valve. M. parlinensis closely resembles “Trigeria” gau- dryi Hall and Clarke, 1893 (non Oehlert, 1877) from Ridgeley Sandstone near Cumberland, Md. No specimens of “T.” gaudryi Hall and Clarke (non Oehlert) were available for study, but plaster casts of figured specimens were studied. M. barroisi has fewer costellae than M. parlinensis. M. parlinensis is similar internally and externally to Mutationella falklandica (Clarke, 1913) except that the latter has relatively coarser costae. “Rensselaeria” circularis Schuchert, 1913 may be identical (as suggested by Cloud, 1942, p. 83) with “Trigeria” gaudryi Hall and Clark (non Oehlert) but lack of well-preserved material does not permit solution of the problem. “R.” circularis was assigned by Cloud (1942, p. 83) to Globithyris, but this as- signment is untenable owing to the absence of crural plates or a median septum in the brachial valve of the former. A myophragm is present in the brachial valve of “R.” circularis, as in Muta- tionella, and hinge plates appear to be discrete. It is suggested that “R.” circularis be assigned ques- tionably to Mutationella pending discovery of more adequate specimens. Occarence and stratigraphic location—USGS locs. 2718—SD, 2691—SD, 2719—SD, 2717—SD, 2721—SD, 2731—SD, 2735—SD, 2720—SD, 2765-SD, 2749—SD, 2777—SD, 2795—SD, 2793—SD, 2776—SD, 2767—SD, 2722—SD, 2861—SD, 2727—SD, 2732—SD, 2831—SD, 2890—SD, 2875—SD, 2824—SD, 2865—SD, 2823—SD, 2783—SD, 2821—SD, 2834—SD, 2847—SD, 2862—SD, 3088—SD, 3090—SD, 3225—SD, 2832—SD; Mutation- ella cf. M. parlinensis fro-m USGS locs. 2819—SD, 2825-SD, 2747—SD, Somerset County, Maine, in the Tarratine Formation (Lower Devonian) and ques- tionably the Kineo member of the Tomhegan Forma- tion (Lower Devonian). Distribution—Somerset County, Maine, and pos- sibly Cumberland, Md., if “Rensselaeria” circularis and “Trigeria” gaadryi belong to this genus and species. A single poorly preserved specimen from the Percé limestone at Percé, Quebec, in the col- lections of the Museum of Comparative Zoology, Harvard Univ., Cambridge, Mass, resembles this species. Strata of similar age in Piscataquis and Penobscot Counrties, Maine, have not, as yet, pro- duced specimens of Mutationella, but Cloudella (a closely related genius) occurs in the area of Mataga- mon Dam, Penobscot County, and Mendathyris (an- other closely related genus) occurs in the P‘resque Isle area, Aroostock County, Maine. Holotype.—UNSM 126178. Figured specimens.—USNM 126168, 126186, 126166, 126173, 126171. Measured specimens.——USNM 126671 A—V. Unmeasured specimens.—USNM 126638, 126674. REFERENCES CITED Albee, A. L., and Boudette, E. L., 1972, Geology of the Attean quadrangle, Somerset County, Maine: U.S. Geol. Survey Bull. 1297, 110 p. Amsden, T. W., 1951, Brachiopods of the Henryhouse Forma- tion (Silurian) of Oklahoma: Jour. Paleontology, v. 25, no. 1, p. 69—96. Berdan, J. M., 1963, Eccentricosta, a new Upper Silurian brachiopod genus: Jour. Paleontology, v. 37, no. 1, p. 245—256. Berry, W. B. N., and Boucot, A. J.. 1970, Correlation of the North American Silurian rocks: Geol. Soc. America Spec. Paper 102, 289 p. Billings, Elkanah, 1874, Paleozoic fossils, v. 2, pt. 1: Canada Geol. Survey, 144 p., pls. 1—10. Boucot, A. J., 19593, Early Devonian Ambocoeliinae (Brachi- opoda): Jour. Paleontology, v. 33, no. 1, p. 16—24, pls. 1, 2. 1959b, Brachiopods of the Lower Devonian rocks at Highland Mills, New York: Jour. Paleontology, v. 33, no. 5, p. 727—769. 1961, Stratigraphy of the Moose River synclinorium, Maine: U.S. Geol. Survey Bull. 1111—E, p. 153—188. 1969, Geology of the Moose River and Roach River Synclinoria, northwestern Maine; with contributions by E. W. Heath: Maine Geol. Survey Bull. no. 21, 117 p. Boucot, A. J., and Amsden, T. W., 1964, Chonostrophiella, a new genus of chonostrophid brachiopod: Jour. Paleon- tology, v. 38, no. 5, p. 881—884, pl. 141. Boucot, A. J ., and Arndt, Robert, 1960, Fossils of the Little- ton Formation (Lower Devonian) of New Hampshire: U.S. Geol. Surv. Prof. Paper 334—B, p, 41—51. Boucot, A. J., Caster, K. E., Ives, David, and Talent, J. A., 1963, Relationships of a new Lower Devonian terebra— tuloid (Brachiopoda) from Antarctica: Bull. Am. Pale- ontology, v. 46, no. 207, p. 77—151, pls. 16-41. Boucot, A. J., and Harper, C. W., 1968, Silurian to lower Middle Devonian Chonetacea: Jour. Paleontology, v. 42, no. 1, p. 143—176, pls. 27—30. Boucot, A. J., and Johnson, J. G., 1967a, Species and distri- bution of Coelospira (Brachiopoda): Jour. Paleontology, v. 41, no. 5, p. 1226—1241, pls. 163—166. 1967b, Paleogeography and correlation of Appa- lachian Province Lower Devonian sedimentary rocks: Tulsa Geol. Soc. Digest, v. 35, p. 35—87, 2 pls. 1968, Brachiopods of the Bois Blanc Formation in New York: U.S. Geol. Survey Prof. Paper 584-B, 27 p. Boucot, A. J., Johnson, J. G., and Staton, R. D., 1964, On some atrypoid, retzioid and athyridoid Brachiopoda: Jour. Paleontology, v. 38, no. 5, p. 805—822. Boucot, A. J., Johnson, J. G., and Talent, J. A., 1969, Early Devonian brachiopod zoogeography: Geol. Soc. America Spec. Paper 119, 106 p. Boucot, A. J., and Yochelson, E. L., 1966, Paleozoic Gastro- poda from the Moose River synclinorium, northern Maine: U.S. Geol. Survey Prof. Paper 503-A, p. A1— A20, pls. 1—3. 126185, REFERENCES CITED 77 Caster, K. E., 1939, A Devonian fauna from Colombia: Bull. Am. Paleontology, v. 24, no. 83, p. 1—218. Clarke, J. M., 1900, The Oriskany fauna of Becraft Moun- tain, Columbia County, N.Y.: New York State Mus. Mem. 3, v. 3, 101 p. 1907, Some new Devonic fossils: New York State Mus. Bull. 107, p. 153—291. 1909, Early Devonic history of New York and east- ern North America: New York State Mus. Mem. 9, pt. 2, 250 p., 34 pls. Cloud, P. E., Jr., 1942, Terebratuloid Brachiopoda of the Silurian and Devonian: Geol. Soc. America Spec. Paper 38, 182 p., 26 pls. Cooper, G. A., 1956, Chazyan and related brachiopods: Smith- sonian Misc. 0011., v. 127, 1024 p. Dunbar, C. 0., 1919, Stratigraphy and correlation of the Devonian of western Tennessee: Tennessee State Geol. Survey, Bull. 21, 127 p. Foerste, A. F., 1903, Silurian and Devonian limestones of western Tennessee: Jour. Geology, v. 11, p. 554-583, 679—715. 1909, Fossils from the Silurian formations of Ten- nessee, Indiana, and Kentucky: Denison Univ. Sci. Lab. Bull. 14, p. 61—116. Goldring, Winifred, 1933, A new species of crinoid from the Devonian (Oriskany) of Maine: Portland Soc. Nat. History Proc., v. 4, p. 153—155, pls. 3, 4. Hall, James, 1863, Contributions to palaeontology: New York State Cabinet of Natural History 16th annual report, p. 3226. Hall, James, and Clarke, J. M., 1894, An introduction to the study of the genera of Palaeozoic Brachiopoda: New York Geol. Survey, Palaeontology, v. 8, pt. 2; 1893, prepr., p. 1—317; 1894, 394 p., pls. 21—84 [1895]. Lespérance, P. J., 1968, Ordovician and Silurian trilobite faunas of the White Head Formation, Percé region, Québec: Jour. Paleontology, v. 42, no. 3, p. 811—826. Neuman, R. B., 1968, Paleogeographic implications of Ordo- vician shelly fossils in the Magog belt of the northern Appalachian region, Chap. 3 in Zen, E-an, and others, eds., Studies of Appalachian geology, northern and maritime: New York and London, Interscience Pub- lishers, p. 35—48. Oliver, W. A., Jr., 1960, Devonian rugose corals from north- ern Maine: U.S. Geol. Survey Bull. 1111—A, p. 1—23, pls. 1—5. 1962, Silurian rugose corals from the Lake Témiscou- ata area, Quebec: U.S. Geol. Survey Prof. Paper 430—B, p. 11—19, pls. 5—8 [1963]. 1967, Succession of rugose coral faunas in the Lower and Middle Devonian of eastern North America, in Internat. Symposium on the Devonian System, Calgary, Alberta, 1967 [Proc.] V. 2: Calgary, Alberta, Alberta Soc. Petroleum Geologists, p. 733-744 [1968]. Pirsson, L. V., and Schuchert, Charles, 1914, Note on the occurrence of the Oriskany Formation on Parlin Stream, Maine: Am. Jour. Sci., Ser. 4, v. 37, p. 221-224. Schuchert, Charles, and Cooper, G. A., 1931, Synopsis of the brachiopod genera of the suborders Orthoidea and Pentameroidea, with notes on the Telotremata: Am. Jour. Sci., Ser. 5, v. 22, p. 241—251. Stumm, E. C., 1962, Silurian corals from the Moose River synclinorium, Maine: U.S. Geol. Survey Prof. Paper 430—A, p. 1—9, pls. 1—4. Walmsley, V. G., Boucot, A. J., and Harper, C. W., 1969, Silurian and lower Devonian salopinid brachiopods: Jour. Paleontology, v. 43, no. 2, p. 492-516. Weller, S., 1903, The Paleozoic faunas: New Jersey Geol. Survey, Pal. 3, 462 p. Whittington, H. B., and Campbell, K. S. W., 1967, Silicified Silurian trilobites from Maine: Harvard Univ., Mus. Comp. Zoology Bull., v. 135, p. 447—483, 19 pls. Williams, Alwyn, 1951, Llandovery brachiopods from Wales with special reference to the Llandovery District: Geol. Soc. London Quart. Jour., v. 107, pt. 1, no. 425, p. 85- 136. 1953, North American and European stropheodontids: their morphology and systematics: Geol. Soc. America Mem. 56, 67 p. Williams, Alwyn, and others, 1965, Treatise on invertebrate paleontology, Part H, Brachiopoda, V. 1-2: New York, Geol. Soc. America (and Univ. Kansas Press), 927 p. Williams, H. S., and Breger, C. L., 1916, The fauna of the Chapman sandstone of Maine: U.S. Geol. Survey Prof. Paper 89, 347 p., 27 pls. Woodard, H. H., 1951, Report on the geology of a portion of the Spencer Lake area, Maine: Maine Geol. Survey Rept. State Geologist 1949—50, p. 68—77. p 4 " My}: m ‘ . I V 22k; ; r... - .( ‘ * jk Page A acinum, Rhynchonella _________________ 30 acinus, Diabolirhynchia ________________ 30 Rhychonella _______________ 30 Acrosm'rifer _______________________ 41 angularis ____________ 43 utlanticus _______ _ 3, 46; pl. 17 hartleyi ____________ 43 intermedius ___ _ ..... 47 murchisom’ ___ 41, 43, 47; pl. 16 olssom _- ___________ 51 sp. 1 - ______ 47; pl. 17 sp. 2 ___ ___________ 48:1)1. 18 aequiradiata, Atrypa. _ _____________ 66 Age, faunas __________________________ 1 Hardwood Mountain Formation ___ 2 Little Big Wood Pond beds _______ 2 alsa, Discomyorthis ___________________ 13 Rhipidamella ____________________ 13 altisulcata, Ancillotoechia ___________ 30; pl. 12 Ambocoeliidae _________________________ 62 Ambocoeliinae ________________________ 62 Amphigenia ________________________ 3, 70, 72 chickasawensis ___________ 72 curta. _____________________ 72 elongata __________________ 70 puma ___________ _ 70; pl. 22 Amphigeniinae _________________ - 70 Amphistrophz‘a ______________ _ 1 funiculata ___ __ l, 2 amplu, St'rophonella ___- _ 23 Anaillotoechia __________________ 30 altisulcata ___- 30; pl. 12 biulveata ___ __________ 30 haraganensis 30 sp ___________________ 30; pl. 12 angularis, Acrospi'rifer ________________ 43 ungustiplicam zaleszczykiensis, H owellella 41 Anopliidae ____________________________ 25 Anoplothecidae _______________________ 38 Anopliinae ___________________________ 25 Anthyrididae _________________________ 65 Anthyridinae _________________________ 65 Antispirifer ___________________________ 50 harroldi _______________ 50; pl. 18 Appalachian forms ____________________ 3 Appalachian ostracode faunda _________ 3 arctostriata, Atrypa ________________ 86; pl 14 arcuaria, Ism‘this _____________________ 10 a'rcuuta, Eadevonaria _____ _ 3, 27; pl 11 Aroostook County __________ ___- 3 atlanticus, Acrospirifer _ ; pl 17 At'rypa _____________________ 36 aequirudiata ______ 66 arctoat'riata. _______ 36; pl. 14 reticuluris ___. ___- 2, 36'; pl. 15 reticularis 5.1 _ 37 tennesseenis _ . 14 Atrypidae _-_ _____________________ 36 Atrypinae ____________________________ 36 11011.8, Nanothyris subglobosa ___________ 66 B Baltic-type fauna _____________________ 3 barroisi, Mutationella __________________ 76 INDEX [Italic page numbers indicate major references] Page Beachia .............................. 3, 69 suessana 69 thum'i _____________ 66, 69; pls. 21, 22 Bear Pond Limestone Member _______ 3 Beck Pond ___________________ ___- 3 Beck Pond Limestone _- 3 beckensis, Sieberella ___- _____ 20; pl. 6 becmftensis, Schuchertella ___________ 24; pl. 9 Berdan, J. M., cited __________________ 2 bialveata, Ancillotoechia _______________ 30 Rhynchonella ________________ 30 bidentata, Rhynchonella _______________ 30 blainvillei, Protoleptoatrophia _______ 22; pl. 8 brownsportensis, Orthostrophia _______ 4; pl. 1 C callida, Globithmis ________ _ 72. 73; pl. 23 Camaratoechia ____________________ 30 canadensis, Dawsonelloides __ - 26‘; pls. 10, 11 Centronellidae _________________ 65 Charionoidea _________ 64 doris ________ ___ 3, 64; pl. 20 chickasuwensis, Amphigenia ___________ 72 Chonetidae ___________________________ 25 Chonetinae ___________________ 25 Chonostrophiella ______________________ 1, 28 camplanata ___- 28; pls. 11, 12 Chonostrophiellinae ___________________ 28 circularis, Rensselaeria ________________ 76 cliftonensis, Dalejina __________________ 16 Cloud, Preston E., Jr., cited __________ 1 Cloudella _____________________________ 76 stewarti __________ _ 74 Cloudothyris ________________ ___ 69 pastwalia ___ _ 69; pl. 22 Coelospim _____________ ___ 1, 2, 38 sp _________ - 38; pl 15 Coelospirinae _______________ 38 coeymanensis, Gyptdula ___________ 19 complanata, Chonost'rophidla. ___ 28; pls. 11, 12 Conodonts, Little Big Wood Pond ______ 2 coopen’, Plicoplasia ____________________ 63 Coral analysis, Oliver, W. A., Jr. ______ 4 Corals, Little Big Wood Pond __________ 2 Costellirostra _________________________ 34 sp ___________________ .94; pl. 14 Costellispirifer ________________________ 51 perimele ____________ 51; pl. 18 Costispirifer __________________________ 3, 61 511 .................... 61: pl. 18 Costispiriferinae ______________________ 61 crassa, Nanothyris subglobasa ___ 66 cumberlandiae, Spirifer _________ --_ 60 Cupularostrum ______________ _ _ _ _ 29 macrocosta, ___ - 2.9, pl. 12 recticostatum _ ___- 30 sp ___- _- 30:191. 12 curtu, Amphigenia. ___ _ 72 cycloptera, Howellella __ _ 40 cyclopterus, H owellella 40; pl. 16 Cyrtina ___________ __-_ 63 rostrum _____ 63, pl. 19 Cyrtinidae _____________ --__ 63 Cyrtininae _________________ ___- 63 cyrtinoides, Megakozlawskiella _ ______ 62 Cyrtoniscus ____________________ ___- 1, 25 nectus ________________ _ 25; pl. 9 Page D Dalejina _________________________ __ 13, 16 cliftonensis ___________ - 16 newsomenais _________ - 16 subtriangularis .- -- 16 sp. 1 __________ __ 16; pl. 3 sp 2 _______ _ _ - _ 1 6 sp ___________________ 1 6 Dalmanellidae ____________ 9 Dawsonelloides ___________________ 1, 3, 26 Deltnyrididae __ ___________________ Delthyridinae _______________ 39 Delthyris 39 (Delthyris) __________________ 2 kozlowskii ________________ 39; pl. 16 pegramensis _________________ 39 demisaa, St'ropheodonta ______________ 21; pl. 6 Devonian, LOWer ________________ _ 3 Diabolirhynchia ________________ _ 30 acinus _________ _ _ _ 30 dianiu, Globithyris _________ 75’; pl. 23 Dicaelosia _________________ - ___ 1, 11 sp __________ 1, 11; pl. 3 Dicaelosiidae ________________ 11 Dinorthidae ___________ 8 Discomyorthis _______________ 12, 13, 16 13 13 musculosa ______________ 13 musculosa solaris--- 12; pls. 3. 4 oblata __________________ 12, 13 sp ______________________ 18. 16 Dolerorthinae _________________________ 5 Dolero’rthis ___________________________ 2, 5 hami ______________________ 7; pl. 2 hobbstownensis __________ 5, 6; pl. 1 rustica asiliensis ____________ 7 sp ________________________ 7 ; pl. 2 5.3. ........................ 6 doris, Charitmoides _________ 3. 64; pl. 20 Draboviine _________________ _ 18 dryope, Rhynchonella ___ _ 29 duodenan‘us, Spirifer _____________ _ 47 E Early Devonian fauna, Moose River synclinorium _____________ Early Silurian fossils, Limestone Hill __ 1 Early Wenlock brachiopods, Limestone Hill Eatonm _______ medialis sp ___________________________ anniidae ____________________________ 34 Eccentricosta ________________________ 2, 3, 25 Sp ___________________ 25; pl. 10 Eitelian age __________________________ 4 elongata, Amphigenia _________________ 70 engelmanni, Spirife-r __________________ 51 Eodevonaria __________________________ 1, 27 arcuata ______________ 3, 27,- pl. 11 anpirifer ____________________________ 2 Eospiriferidae 39 Eospiriferinae 39 eryna, Discomyorthis __________________ 13 ‘79 80 Page Esopus age ___________________________ 3 European forms ___- 3 Eurythyridinae _____________ 69 excellens, Rhynchonella ________________ 29 F falklandica. Mutationella ______________ 76 Famine Limestone ____________________ 4 Faunas, age __________________________ 1 flabellitcs, Leptocoelia. ______________ .98; pl. 15 formosa, Machaeraria _________________ 35 funiculata, Amphistrophia _____________ l, 2 G Gaspé ________________________________ 3 gaspensis, Howellella. __ _ 41 gaudryi, Trigeria __ _ 76 Globithyridinae __ ..... 72 Globithyris ______ __ 3, 72, 73, 76 calltda _ 72, 73; pl. 23 dianm _ __ 73; pl. 23 sp ________________________ 73 Glyptorthinae _________________________ ‘7 Greenfieldia ___________________________ 65 Gym‘dula _____________________________ 2, 19 coeymanensis ________________ l9 sp. 1 _____________________ 19; pl. 5 sp. 2 ___________________ 1.9; pls. 5, 6 Gypidulidae ___________________________ 19 Gypidulinae ___________________________ 19 H hami, Dole-ro-rthis ................... 7; pl. 2 haruganensis, Ancillotoevhia ______ 30 Hardwood Mountain Formation -_ 2 harroldi, Antispirz'fer __ - 50; pl. 18 hartleyi, Acrospirifer _ _ 43 Hedeina ____________ 39 macropleura _ 39; pl. 15 Helderberg age _______________________ 3 Helderberg faunas ____________________ 3 Hesperotbidae ________________________ 5 H csperorthis __________________________ 5 H ipparionya: __________________________ 3. 24 5p _____________________ 24; pl. 9 hitchcocki, Salopina __________________ 18; pl. 5 Hohbstown Formation, lower conglomerate member _____ 2 hobbstownensia, Dolerorthis _______ 5; 6; pl. 1 hodgei, Nanothyris _________________ 65; pl. 21 Howellella ____________________________ 2, 39 angustiplicata Zaleszczykiensis 41 cycloptera __________________ 40 cyclopterus - __________ 1.0; pl. 16 gaspensis _- _____________ 41 mckenzica __ _ 41 nerei ________ __ 41 tomheganensiz 40; pl. 16 sp ______________________ 39; pl. 16 I intermedius, Acrospirifer ______________ 47 Introduction - ___ 1 Isorthinae _ _ _ 10 Isorthis ________ - 10 arcuaria _ ___ 10 pe’relegans _ _ 10, 11; pl. 3 81). 1 _ 10, 11; pl. 2 3p. 2 __ _ 10; pl. 2 sp 3 _______________________ 10; pl 3 K Kennebec Formation ___________________ 3 Kineo Rhyolite _______________________ 3 INDEX Page Kirkidium knighti ___ - 2 knighti, Kirkidium ___ _ 2 Kazlowskiellina __ _ 62 Kozlowskiellinae _____ __ 62 kozlowskii, Delthy'ris _______________ 39; pl. 16 L lata, Meristella ____________________ 64; pl. 20 Late Llandovery brachiopods, Limestone Hill ............ 1 Leptaena _____________________________ 20 rhomboidalis ______________ 20, pl 6 Leptaenidae __________________________ 20 Leptaenisca __________________________ 25 sp ______________________ 25:1)1 9 Leptaenoideinae ______________________ 25 Leptocoelia ___________________________ 38 flabellitea _______________ .78; pl. .15 Leptocoeliidae ________________________ 38 Leptostrophia ________________________ 21, 22 magnifica _____________ .21; pl. 7 sp _____ _ 22; pl. 7 Levenea ___________ _ 11 subcarinata - _-_ 11 sp ________ _ 11; pl. 3 Limbinaria muricatu. __________________ 2 Limestone Hill, Early Silurian fossils__ 1,3 early Wenlock brachiopods _______________ 1 late Llandovery brachiopods _______________ 1 Liasatrypa ___________________________ 2, 37 sp _____________________ 37'; pl. 15 Lissatrypidae _________________________ 37 Little Big Wood Pond _________________ 2 Lobster Lake Formation ______________ 2 lucia, Orthis _______________ 17 2 M Machaera’ria __________________________ 35 formosa _- ___ 35 mainensis __ .95; pl. 14 macro, Mucrospirifer _______________ 60; pl. 18 macrocosta. Cupularostrum _________ 29; pl. 12 macropleura, Hadeina. ______________ 39; pl. 15 magnifica, Leptostrophia ___________ 21; pl. 7 mainensis, Machaeraris _____________ 35; pl. 14 mckenzica, H owellella _________________ 41 medialis, Eatonia ___________________ 84; pl. 14 Megakozlowslciella. _____________________ 62 cyrtinoides _________ 62 sp _______________ 62; pl. 19 Megalanteris thunei - 69 Meganteris _ _ 69 Meifodia _____ _ 37 Mendathyris ___ - 76 Merista ___________ _ 63 tennesseensis __________ 63; pl. 20 Meristella _____________________ ___ 64 lata _____________________ 64; pl 20 sp __________________________ 64 Meristellidae __________________________ 63 Meristellinae __________________________ 64 Meristinae ____________________________ 63 Metaplasia _________________________ 1, 62, 63 minuta _________________ 62; pl. 19 paucicostata _____________ 62; pl. 19 pyxidata __________________ 62 minuta, Metaplasiu ________________ 62; pl. 29 Mont Wissick Formation _____________ 2 Moose River synclinorium, purpose geologic study ____________ 1 Mucrospirifer _________________________ 60 macra ______________ 60; pl. 18 Mucrospiriferinae _____________________ 60 murahisani, Acrospirijer _____ 1,1, 43, 47; pl. 16 Page muricata, Limbinaria _-_ _ 2 musculosa, Discomyo‘rthis _ _ l3 Rhipidomella __ __- _ 13 solaris, Discomym‘this ".12; pls. 3, 4 mutabalis, Nanothyris ________________ 66 Mutationella _________________________ l, 3, 73 barroisi _________________ 76 community ______________ 3 falklandica _______________ 76 parlinensis ________ 73, 76; pl. 23 podolica _________________ 74 Mutationellidae _______________________ 73 M‘utationlellinae ______________________ 73 N Nunospi'ra. Nanothyris hodgei _________________ 65; pl. 21 mutabilis _________________ 66 reesidei ___________________ 66 aubglobosa. _____________ 66; pl. 21 subglobosa avus ___________ 66 subglobosa crassa _________ 66 nectus, Cyrtoniscus _________________ 25: pl. 9 nerei, Howellella ______________________ 41 New Brunswick ______________________ 3 ncwsomensis, Dulejina ________________ 16 Nucleospira ___________________________ 65 sp _ 65; pl. 20 Nucleospiridae ________________________ 65 0 oblata, Discomyorthia ________________ 12, 13 Old World genus __--.. _______________ 3 Old World province __________________ 3, 4 Oliver, W. A., Jr., coral analysis _____ 4 olssoni, Acrospirifer __________________ 51 Onondaga age ___________________ 4 Ordovician age __________________ 3 Late _- -_ 3 Middle ______ _-__ 3 Ordovician trilobites, Late _ 3 Oriskany age _____________ _ 3, 4 Orthambonites _____ _ 5 sp _________________ l, 5; pl. 1 Orthidae _____________________________ 4 Orthinae _____________________________ 5 Orchis lucia __________________________ 17 Orthoid, unidentified _______________ 11; pl. 3 Orthostrophia __. ______________________ 4 browmportenais _______ 4; pl. 1 strophomcnoides ______ 4: pl. 1 Orthostrophinae ______________________ 4 asiliensis, Dolerorthis rustica __________ 7 Ostrocodes, Little Big Wood Pond _____ 2 P Paleontology, systematic _- _________ 4 Parker Bog Formation -_ _______ 3 parlinensis, Mutatio'nella _ _ 7.9, 76; pl. 23 parva, Amphigenia. ______ ___- 70, pl. 22 paucicostata, Metaplasia _ __ 62; pl. 19 pegramensis, Delthy'ris __ __ 39 Reticularia _ _ _ 39 Penobscot County _______ _ 3 Pentameridae --_- ________ _ 20 Pentamerinae ___- - 20 Pentameroides - _ _ 20 Pentamerus sp ______ - 1, 20; pl. 6 perelegans, lam-this _ 10, 11; pl. 3 perimels, Costellispirifer __ 51; pl. 18 Spirifer ___ _ 51 pirssonae, Spirifer ___- - 51 planoconvexa, Platyorthis __________ 16; pl. 4 Page Platyorthinae _________________________ 1 6 Platym'thia ___ planacm’wexa ___________ 16'; pl. 4 Plectambonitidae _______________________ 23 Plectambonitinae _____________________ 20 plicata, Plicoplasia _______________ 63; pl. 19 Plicoplasia ____________________________ 1, 63 cooperi ___________________ 63 plicata ________________ 63; pl. 19 podolica, Mutationella _________________ 74 postovalis, Cloudothyris ____________ 69; pl. 22 Pridoli age ____________________ _- 2, 3 Protathyris ____________________ _ _ 65 Protathyris sp _____________ 65; pl. 20 Protoleptostrophia __________ _ _ 22 22; pl. 8 3p __- 22, 23; pls. 7, 8 Provinciality __________________ 3 Ptychopleurellv. ___ _______ 2, 7, 8 sp _ ______ 7; pl. 2 5.5 ________________ 8 punctulifem, St-rophonella __________ 2.9; pl. 8 Purpose, geologic study Moose River synclinorium _____________ 1 pymidata, Metaplasia __________________ 62 Q. R Quebec _______________________________ 3 recticostatum, Cupularostrum _________ 30 reeaidei, Nanothy'ris __________________ 66 References cited ______________________ 76 Rensselae'ria __________________________ 3, 66 circularis ________________ 7 6 sp ___________________ 66; pl. 21 5.5 ...................... 66 Rensselaeriinae _______________________ 65 Resserella ____________________________ 2, 9 5p ____________________ .9, 10; pl. 2 Reticutan’a pegramensis _______________ 39 reticula'ris, At'rypa _____________ 2, 36; pl. 15 Atrypa, 5.1 _________________ 37 Rhenish community __________________ 3 Rhenm‘ensselaeria _____________________ 73 Rhipidomella also. _____________________ 13 musculosa ________________ 13 Rhipidomellidae _____________________ 12, 16 Rhipidomellinae _______________________ 12 Rhipidothyrididae _____________________ 72 rhomboidalia, Leptaena ______________ 20; pl. 6 Rhynchanella acinum _________________ 30‘ INDEX Page Rhynchtmella acinus __________________ 30 bialveata ________________ 30 bidentatz ______________ 30 dryope _________________ 29 excellens ________________ 29 Rhynchotrematidae ____________________ 35 roeme‘ri, Sieberella ____________________ 20 rostmta, Cyrtina ___________________ 63; p]. 19 rustica oailie‘nsis, Dale'ro’rthis __________ 7 S Saint George, Quebec _________________ 4 Sally Mountain ______________________ 2 Salopina _____________________________ 1, 18 hitchcocki _________________ 18; pl. 5 Schizophmia striatula _________________ 13 sp ______________________ 13 Schizophoriidae _______________________ 18 Schizara/mma ________________________ 4: pl. 1 subplicata _______________ 4 Schoharie age _______________________ 3, 4 Schuche’rtella _________________________ 24 becmftensis ____________ 24: pl. 9 sp ____________________ 24; pl. 9 Schuchertellidae ................. 24 Schuchertellinae _________________ 24 Seboomook Formation ___ 3 Siebc’rella. ___________________ -- 20 beckensis -- 20; pl. 6 Toemeri ______ 20 sieberi _- - 20 sieberi, Siebe-rella _ 20 Silurian age ________________________ 2, 3 solaris, Discomyo’rthis musculasa“ 12; pls. 3, 4 Somerset County _____________________ 3 Sowe'rb'yites sp _________ 1, 20; pl. 6 Spencer Mountain outlier ____________ 2 Sphaerirhynchia ______________________ 31 sp. 1 _________ 31; p15. 12, 13 sp. 2 _____________ 32; pl. 13 sp. 3 ______________ 33; pl. 13 sp ________________ 83; pl. 13 Spinoplasia ___________________________ 3 Spirife-r cumberlandiae _______________ 60 duodenarius _________________ 47 engelmanni .................. 51 perimele _____________________ 51 pirsszmae .................... 51 worthenanus _________________ 51 atewa’rti, Cloudella ___________________ 74 81 Page Stratton quadrangle __________________ 1 striatula, Schizopho'ria ________________ 13 Stropheodonta ________________________ 21 demissa _____________ 21; pl. 6 sp _________________ 21; pl. 6 s. s ____________________ 21 Stropheodontidae _____________________ 21 Stropheodontinae _____________________ 21 Strophomenidae ______________________ 25 strophomenoides, Orthostrophia ______ 4; pl. 1 Strophonella _______________________ _ 23 ampla ___________________ 23 punctulifem ___________ 23; pl. 8 subcarinuta, Levenea _________________ 11 subglobosa, Nanothy’ris ____________ 66; pl. 21 twus, Nanothyris _________ 66 crasaa, Nanothyris ________ 66 subplicata, Schizoramma __________ _ 4 subtriangularis, Dalejina. _________ _ 16 suessana, Beachia _______________ _ 69 sulcata, Sulcatinu _______________ _ 34 Sulcutina ____________________________ 2, 33 sulcata _____________ _ 34 tennessecnis ___ _____ 34 sp _____________ _ 33; pl. 13 Systematic Paleontology ______________ 4 T Tarratine Formation _________________ 3 tennesseenis, Atrzma ______ _- 86; pl. 14 Mariam __ .68; pl. 20 Sulcatimz _ 34 thunei, Megalunte’ris __ __________ 69 thunii. Beachia ______ _ 66, 6‘9; pls. 21, 22 Tomhegan Formation ________________ 3 tomheganensis, Howellella __ 40; pl. 16 Tonoloway Limestone _____ __ 2 T’rigeria gaudryi __ -_ 76 Trigonirhynchiidlae ____________________ 29 U, V, W, Z Uncinulidae __________________________ 31 Valcourea ____________________________ 8 sp _____________________ 1, 8: pl. 2 Wenlock age, late ................... 2 worthenanus, Spirifer ________________ 51 zaleszczukiensis, Howellclla ___- angustiplicata. ____________ 41 PLATES 1-23 [Contact photographs of the plates in this report are available, at cost, from 11.8. Geological Survey Library, Federal Center, Denver, Colorado 80225] PLATE 1 FIGURES 1—5. Orthostrophia cf. 0. brownsportensis Amsden. 1949 (p. 4). Hardwood Mountain Formation. USGS loc. 4843—SD, Attean quadrangle, Somerset County, Maine, X 3. 1. Impression of interior of pedicle valve. Note tripartite form of muscle field with concentric striae, obsolescent nature of dental lamellae, presence of umbonal fold and anterior sulcus. Note low groove in bottom of delthyrial cavity which may have delimited pedicle adjustor area. USNM 160109A. 2. Impression of exterior of pedicle valve. Note presence of fold in umbonal region, costae originating by both bifurcation and implantation. Concentric filae also prominent, together with concentric growth lines. Counterpart to specimen in figure 1. USNM 1601093. 3. Impression of interior of brachial» valve. Note quadripartite form of muscle field, linear cardinal proc- ess laterally bounded by parallel ridges, and broad myophragm extending anteriorly from cardinal process to bisect muscle field. USNM 160110A. 4. Impression of exterior of brachial valve. Note presence of prominent sulcus in umbonal region. Coun- terpart to specimen in figure 3. USNM 1601103. 5. Impression of interior of pedicle valve. Note the form of the muscle field and the presence of a fold in the umbonal region which changes into a sulcus anteriorly. USNM 160111. 6—12. Orthostrophia cf. 0. strophomenoides (Hall, 1857) (p. 4). Beck Pond Limestone. USGS 10c. 3499—SD, Spencer Lake quadrangle, Somerset County, Maine. X 1. 6. Posterior view. Note relatively greater length of pedicle interarea and posteriorly sulcate brachial valve. USNM 125786B. 7. Anterior view. Note uniplicate commiSSure with fold on brachial valve. USNM 1257863. 8. Side view. Note somewhat greater convexity of pedicle valve. USNM 1257863. 9. View of pedicle valve. Note sulcus present anteriorly and absent posteriorly. USNM 1257863. 10. View of brachial valve. Note sulcus present posteriorly and fold present anteriorly. USNM 1257863. 11. Internal impression of pedicle valve. Note short dental lamellae, tripartite muscle field and impres- sion of costellae anteromedially. USNM 125786A. 12. Internal impression of brachial valve. Note quadripartite adductor impressions, short median septum, bladelike cardinal process, and stubby brachiophores. USNM 1257853. 13—14. Schizommma fissistriata Foerste, 1909 (p. 4). Interior and exterior of brachial valve (X 2). Note pair of ridges lateral to cardinal process ridge. USNM 87017. Osgood Formation. New Marion, Indiana 15—17. Ortha/mbonites? sp‘. (p. 5). Latex replica of exterior: posterior, anterior, and brachial views. (X 3). USNM 125753. Kennebec For- mation. USGS 10c. 4317—00, Brassua Lake quadrangle, Somerset County, Maine. 18—26. Dolerorthis hobbstownensis Boucot, n. sp. (p.5). Base of Hobbstown Formation. USGS 10c. 3479—SD, Spencer Lake quadrangle, Somerset County, Maine. Holotype, USNM 125796A, B. 18, 22. Latex replica and impression of interior of brachial valve (X 2). USNM 1257963. 19. Posterior view of latex replica of exterior of pedicle valve (X 4). USNM 125903A. 20, 25. Impression and latex replica of exterior of brachial valve (X 2). USNM 125796A. 21, 23. Impression and latex replica of exterior of pedicle valve (X 2). USNM 125903A. 24, 26. Impression and latex replica of interior of pedicle valve (X 2). USNM 1257973. GEOLOGICAL SURVEY PROFESSIONAL PAPER 784 PLATE 1 24 ORTHOSTROPHIA, SCHIZORAMMA, 0RT11AMBONITES‘?, AND DOLERORTHIS FIGURES 1—5. 6—8. 9—17. 18—22. 23—27. 28~32. 33—35. PLATE 2 Dolerorthis cf D. hami (Amsden, 1951) (p. 7). 1—4. Base of Hobbstown Formation. USGS loc. 3479—SD, Spencer Lake quadrangle, Somerset County, Maine. X 2. 1,2 Impressions of interior and exterior of pedicle valve. US'NM 1257953, A. 3, 4. Impressions of interior and exterior of brachial valve. USNM 125794A, B. 5. Impression of interior of pedicle valve (X 3). USNM 160112. Harwood Mountain Formation. USGS loc. 5995—SD, Attean quadrangle, Somerset County, Maine. Dole'rorthis? sp. (p. 7). Base of Hobbstown Formation. USGS 10c. 3479—SD, Spencer Lake quadrangle, Somerset County, Maine. X 2. 6. Impression of interior of pedicle valve. USNM 1258003. 7,8. Latex replica and impression of exterior of pedicle valve. USNM 1258000. Ptychopleurella sp. (p. 7). Base of Hobbstown Formation. USGS loc. 3479—SD, Spencer Lake quadrangle, Somerset County, Maine. 9, 11. Latex replica and impression of exterior of brachial valve (X 2). USNM 125801A. 10. Latex replica of exterior of pedicle valve (posterior View X 3). USNM 125804. 12, 13. Impression and latex replica of interior of pedicle valve (X 2). USNM 125802A. 14, 16. Impression and latex replica of exterior of pedicle valve (X 2). USNM 1258033. 15. Latex replica and impression of interior of brachial valve (X 2). USNM 1258013. Valcourea sp. (p. 8). Kennebec Formation. USGS 10c. 4317—00, Brassua Lake quadrangle, Somerset County, Maine. 18. Latex replica of exterior (posterior view X 2). USNM 125751. 19, 21. Impression of interior of brachial and pedicle valves (X 2). USNM 1257520. 20, 22. Impression of exterior of pedicle and brachial valves (X 1). USNM 1257503, A. Resserella sp. (p. 9). Base of Hobbstown Formation. USGS 10c. 3479—SD, Spencer Lake quadrangle, Somerset County, Maine. X 2. 23. Impression of exterior of pedicle valve. USNM 125772A. 24. Latex replica of exterior of pedicle valve. USNM 125774A. 25. Impressions of interior of pedicle valve. USNM 1257743. 26, 27. Impressions of exterior and interior of brachial valve. USNM 1257733, A. Iso'rthis sp. 1 (p. 10). Base of Hobbstown Formation. USGS 3479—SD, Spencer Lake quadrangle, Somerset County, Maine. X 2. 28, 29. Impression and latex replica of interior of brachial valve. USNM 125828. 30, 31. Impressions of interior and exterior of brachial valve. USNM 125829A, 3. 32. Impression of interior of pedicle valve. USNM 125827. Isorthis sp. 2 (p. 10). 33. Impression of interior of brachial valve (X 3). USNM 125831A. Hardwood Mountain Formation. USGS 10c. 3495—SD, Spencer Lake quadrangle, Somerset County, Maine. 34, 35. Impressions of exterior and interior of pedicle valve (X 3). USNM 1258263, A. Parker Bog For- mation. USGS 10c. 3487—SD, Spencer Lake quadrangle, Somerset County, Maine. GEOLOGICAL SURVEY PROFESSIONAL PAPER 784 PLATE 2 29 3O 32 DOLERORTHIS, DOLERORTHIS?, PTYCHOPLEURELLA, VALCOUREA, RESSERELLA, AND ISORTHIS PLATE 3 FIGURES 1—16. Isorthis sp. 3 (p. 10). Hardwood Mountain Formation. Attean quadrangle, Somerset County, Maine. X 2. 1. Impression of interior of pedicle valve. USNM 160113. USGS 10c. 5587—SD. 2. Impression of interior of brachial valve. USNM 160114. USGS loc. 5583—SD. 3. Impression of interior of brachial valve. USNM 160115. USGS loc. 5586—SD. 4. Impression of interior of pedicle valve. USNM 160116. USGS. loc. 5586—SD. 5—8, 10. Impression of interior: anteriorvliew, side view, pedicle view, brachial view, and posterior view (pedicle valve up). USNM 160117. USGS 10c. 5583—SD. 9. Impression of interior of pedicle valve. USNM 160118. USGS loc. 5583—SD. 11—16. Impression of interior: brachial view, pedicle view, side view (pedicle valve to right), side view (pedicle valve to left), posterior view (pedicle valve up), and anterior view (pedicle valve up). USNM 160119. USGS 10c. 5583—SD. 17. Isorthis cf. I. perelega‘rw (Hall, 1859) (p. 11). Impression of interior of brachial vaIVe (X 2), USNM 125813. Beck Bond Limestone. USGS loc. 3499—S-D, Spencer Lake quad- rangle, Somerset County, Maine. 18—20. Levenea sp. (p. 11). Beck Pond Limestone. USGS 10c. 3499—SD, Spencer Lake quad- rangle, Somerset County, Maine. X 2. 18, 19. Impression of interior of pedicle valve. USNM 125812, 125811B. 20. Impression of interior of brachial valve. USNM 125814. 21. Unidentified orthoid (p. 11). Impression of interior of pedicle valve (X 4), USNM 126473. Lobster Mountain volcanics. USGS loc. 3280—SD, North East Carry quadrangle, Piscataquis County, Maine. 22. Dicaelasia sp. (p. 11). Exterior of brachial valve (X 3). USNM 125810. Beck Pond Limestone. USGS 10c. 3499~SD, Spencer Lake quadrangle, Som- erset County, Maine. 23—25. Discomyorthis musculosa solaris (Clarke, 1907) (p. 12). Tomhegan Formation. USGS loc. 2750—SD, Brassua Lake quad- rangle, Somerset, County, Maine. X 20. 23. Latex replica of exterior of brachial valve. USNM 125808A. 24. Latex replica of interior of pedicle valve. USNM 125806. 25. Impression of interior of pedicle valve. USNM 125806. GEOLOGICAL SURVEY PROFESSIONAL PAPER 784 PLATE 3 [SORTHIS, LEVENEA, UNIDENTIFIED ORTHOID, DICAELOSIA, AND DISCOMYORTHIS FIGURES 1—3. 10. 11—18. PLATE 4 Disaomyorthis musculosa solo/r138 (Clarke, 1907) (p. 12). Tomhegan Formation. USGS 10c. 2750—SD, Brassua Lake quad~ rangle, Somerset County, Maine. X 2. 1, 3. Latex replica and impression of interior of brachial valve. USNM 125809A. 2. Impression of interior of pedicle valve. USNM 125807B. Dalejma sp. 2 (p. 16). Impression of interior and exterior of brachial valve (X 3). USNM 125824B, A. Hardwood Mountain Formation. USGS loc. 3469—SD, Spencer Lake quadrangle, Somerset County, Maine. Dalejina sp. 1 (p. 16). Base of Hobbstown Formation. USGS 10c. 3479—SD, Spencer Lake quadrangle, Somerset County, Maine. X 3. 6, 8. Impression of exterior and interior of pedicle valve. USNM 125821B, A. 7, 9. Impression of interior and exterior of brachial valve. USNM 125822B, A. Discomyorthis sp. (p. 13). Impression of interior of pedicle valve (X 3). USNM 125823. Beck Pond Limestone. USGS loc. 3499—SD, Spencer Lake quadrangle, Somerset County, Maine. Platyorthis planoconvexa (Hall, 1859) (p. 16). Lower sandstone of Tarratine Formation, Somerset. County, Maine. 11, 12. Impression and latex replica of interior of pedicle valve (X 2). USNM 125819. USGS 10c. 3474—SD, Pierce Pond quadrangle. 13, 15. Latex replica of interior and exterior of brachial valve (x 2, X 3). USNM 1258203, A. USGS 10c. 2771—SD, Moosehead Lake quadrangle. 14, 17. Impression and latex replica of exterior of pedicle valve (X 2). USNM 125825A. USGS 10c. 3474—SD, Pierce Pond quadrangle. 16, 18. Impressions of interior and exterior of brachial valve (X 2). USNM 12582013, A. USGS loc. 2771—SD, Moosehead Lake quadrangle. GEOLOGICAL SURVEY PROFESSIONAL PAPER 784 PLATE 4 DISCOMYORTHIS, DALEJINA, AND PLATYORTHIS PLATE 5 FXGURES 1—11. Salopina hitchcocki Walmsley, Boucot, and Harper, 1969 (p. 18). 1, 2. Latex replica and impression of interior of brachial valve (x 3). Lower sandstone of Tarrati'ne Formation. USGS 10c. 2872—SD, Pierce Pond quadrangle, Somerset County, Maine. USNM 125783A. 3, 4. Latex replica and impression of exterior of brachial valve (X 3). Lower sandstone of Tarratine Formation. USGS loc. 2872—SD, Pierce Pond quadrangle, Somerset County, Maine. USNM 125783B. 5. Latex replica of interior of pedicle valve (X 3). Lower sandstone of Tarratine Formation. USGS 10c. 2743—SD, Moosehead Lake quadrangle, Somerset County, Maine. USNM 125798A. 6. Impression of interior of pedicle valve (X 3). Seboomook Formation. USGS 10c. 2884—SD, Chesuncook quadrangle, Piscataquis County, Maine. USNM 125793A. 7, 8. Latex replica and impression of interior of brachial valve (X 3). Tarratine Formation. USGS 10c. 2861—SD, North East Carry quadrangle, Somerset County, Maine. Holo- type, USNM 125782B. 9. Impression of interior of pedicle valve (X 3). Lower sand- stone of Tarratine Formation. USGS 10c. 2705—SD, Bras- sua Lake quadrangle, Somerset County, Maine. USNM 125781B. 10, 11. Impression and latex replica of exterior of brachial valve (X 3). Tarratine Formation. USGS loc. 2861—SD, North East Carry quadrangle, Somerset County, Maine. Holo- type, USNM 125782A. 12—14. Gypidula sp. 1 (p. 19). Hardwood Mountain Formation, USGS loc. 3488-SD, Spencer Lake quadrangle, Somerset County, Maine. 12. View of inner crural plates and hinge plate (X 3). USNM 126508. 13, 14. Side view of spondylium (X 3) and exterior of pedicle valve (X 2). USNM 126509A. 15—23. Gypidula sp. 2 (p. 19). 15—19. Exterior: side, brachial, pedicle, posterior, and anterior views (X 1). Hardwood Mountain Formation. USGS 10c. 2728—SD, Spencer Lake quadrangle, Somerset County, Maine. USNM 125840. 20—23. Anterior (brachial valve above), opposite side, and pos- terior views (X 2). Hardwood Mountain Formation. USGS 10c. 5587—SD, Attean quadrangle, Somerset County, Maine. USNM 160120. GEOLOGICAL SURVEY ‘ PROFESSIONAL PAPER 784 PLATE 5 22 SALOPINA AND GYPIDULA PLATE 6 FIGURES 1, 2. Gypidula sp. 2? (p. 19). Views of silicified cardinalia (X 1). USNM 125842. Hardwood Mountain Formation. USGC loc. 3485-S‘D, Spencer Lake quad- rangle, Somerset County, Maine. 3—8. Sieberella. beckensis Boucot, n. sp. (p. 20). Beck Pond Limestone. USGS loc. 3499—SD, Spencer Lake quad- rangle, Somerset County, Maine. 3, 6, 8. Exterior (X 1), side (X 2), and posterior (X 2) views of pedicle valve. Holotype, USNM 125841A. 4. Exterior of brachial valve (X 1). USNM 125838B. 5. Side view of spondylium (X 3). USNM 125837. 7. Cross section of brachial valve (X 3). Note lyre-shaped cross section of brachial lamellae. USNM 125839. 9, 10. Pentamerus? sp. (p. 20). Lime-silicate hornfels of Early Silurian age. USGS loc. 3475—SD, Limestone Hill, Stratton quadrangle, Somerset County, Maine. 9. Impression of interior of pedicle valve (X 2). USNM 125835. 10. Partly exfoliated pedicle valve (X 1). USNM 125836B. 11. Sowerbyites? sp. (p. 20). Latex replica of exterior of brachial valve (X 3). USNM 126152A. Kennebec Formation. USGS 10c. 4317—00, Brassua Lake quadrangle, Somerset County, Maine. 12—16. Leptaena “rhomboédalis” (Wilckens, 1769) (p. 20). Spencer Lake quadrangle, Somerset County, Maine. 12. Exterior of pedicle valve (X 2). USNM 126141B. Beck Pond Limestone. USGS 10c. 3499—SD. 13, 15. Impression (X 1) and latex replica (X 2) of interior of brachial valve. USNM 126146A. Beck Pond Limestone. USGS 10c. 3497—SD. 14, 16. Impressions of interior and exterior of pedicle valve (X 1). USNM 1261273, A. Base of Hobbstown Formation. USGS 10c. 3479-SD. 17—19. Stropheodonta cf. S. demissa (Conrad, 1842) (p. 21). Tomhegan Formation. Brassua Lake quadrangle, Somerset County, Maine. 17. Latex replica of exterior of brachial valve (X 3). USNM 126145B. USGS 10c. 2750—SD. 18, 19. Latex replica of exterior and impression of interior of brachial valve (X 2). USNM 126144A. USGS 10c. 2820—SD. 20—23. afi. Stropheodonta sp. (p. 21). Tomhegan Formation. Brassua Lake quadrangle, Somerset County, Maine. 10. Latex replica of exterior of pedicle valve (X 2). USNM 126153A. USGS 10c. 2820—SD. 21. Impression of interior of pedicle valve (X 1). USNM 126126A. USGS loc. 2750—SD. 22. Latex replica of interior of pedicle valve (X 3). USNM 126125. USGS 10c. 2820—SD. 23. Latex replica of exterior of brachial valve (posterior view X 3). USNM 126126B. USGS loc. 2820—SD. GEOLOGICAL SURVEY PROFESSIONAL PAPER 784 PLATE 6 21 22 19 GYPIDULA, SIEBERELLA, PENTAMERUS?, SOWERBYITES? LEPTAENA, AND STROPHEODONTA PLATE 7 FIGURES 1-9. Leptostrophia cf. L. magnified (Hall, 1857) (p. 21). Lower sandstone of Tarratine Formation. Somerset County, Maine. 10, 11. 12, 13. 1, 3. 4, 8. 7. 9. Impression and latex replicas of interior of pedicle valve (x 1). USNM 126121. USGS 10c. 2813—SD, Pierce Pond quad- rangle. Latex replica and impression of interior of brachial valve (X 4). USNM 126158B. USGS 10c. 2813—SD, Pierce Pond quadrangle. Latex replica of exterior (pOsterior brachial view x 1). USNM 126159. USGS 10c. 2705—SD, Brassua Lake quad- rangle. Latex replica of interior of brachial valve (X 2). USNM 126142. USGS loc. 2813—SD, Pierce Pond quadrangle. Impression of interior of pedicle valve (X 1). USNM 126122. USGS loc. 2813—SD, Pierce Pond quadrangle. Impression of interior of pedicle valves (X 1). USNM 126123. USGS 10c. 2705-SD, Brassua Lake quadrangle. Leptostrophia? sp. (p. 22). Parker Bog Formation. USGS 10c. 3477—SD, Pierce Pond quad- 10. 11. rangle, Somerset County, Maine. Impression of interior of brachial valve (X 3). USNM 126120. Impression of interior of pedicle valve (X 2). USNM 126147B. Leptost’rophia sp. or Protoleptostrophia sp. (p. 22). Latex replica and impression of interior of pedicle valve (X 1). USNM 126118A. Seboomook Formation. USGS 10c. 2857—SD, Brassua Lake quadrangle, Somerset County, Maine. GEOLOGICAL SURVEY PROFESSIONAL PAPER 784 PLATE 7 9 LEPTOSTROPHIA, LEPTOSTROPHIA?, AND PROTOLEPTOSTROPHIA PLATE 8 FIGURES 1—8. Protoleptostrophia cf. P. blainvillei (Billings, 1874) (p. 22). Tomhegan Formation. Brassua Lake quadrangle, Somerset County, Maine. 1. Latex replica of exterior of pedicle valve (X 2). USNM 126154A. USGS loc. 2820—SD. 2, 3. Latex replica and impression of interior of brachial valve (X 2). USNM 126140. USGS 10c. 2750—SD. 4. Impression of exterior of brachial valve (x 1). USNM 126137B. USGS loc. 2820—SD. 5, 6. Impression and latex replica of interior of brachial valve (X 1). USNM 126137A. USGS 2820—SD. 7, 8. Latex replica of exterior and impression of interior of pedi- cle valve (x 1). USNM 126136B, A. USGS loc. 2820—SD. 9—13. Protoleptostrophia sp. (p. 23). Lower sandstone of Tarratine Formation. Somerset County, Maine. (X 2). 9, 10. Impression of interior and latex replica of exterior of pedicle valve. USNM 126135B, A. USGS loc. 2719—SD, Long Pond quadrangle. 11. Impression of interior of brachial valve. USNM 126134A. USGS 10c. 2719—SD, Long Pond quadrangle. 12, 13. Latex replicas of interior of brachial valve. USNM 126133, 126132. USGS loc. 2701~SD, Brassua Lake quadrangle. 14—18. Strophonella cf. S. punctul’ifera (Conrad, 1838) (p. 23). Parker Bog Formation. USGS loc. 3477—SD, Pierce Pond quad- rangle, Somerset County, Maine. X 1. 14. Latex replica of exterior of pedicle valve from slab. USNM 126119. 15. Latex replica of exterior of pedicle valve from slab. USNM 126119. 16, 17. Latex replica and impressiOn of interior of pedicle valve. USNM 126143. 18. Latex replica of interior of brachial valve, USNM 126157A. GEOLOGICAL SURVEY PROFESSIONAL PAPER 784 PLATE 8 16 17 PROTOL EPTOSTR OPHIA AND STR OPHONELLA PLATE 9 FIGURES 1—11. “Schuchertella” becmftensis (Clarke, 1900) (p.24). 1—6. Lower sandstone of Tarratine Formation. Long Pond quadrangle, Somerset County, Maine. 1, 3. Impression and latex replica of interior of brachial valve (X 3). USNM 126138. USGS 10c. 2719-SD. 2, 4. Impression and latex replica of brachial valve (X 3). USNM 1261513, A. USGS 10c. 2777—SD. 5. Latex replica of interior of brachial valve (X 2). USNM 126160B. USGS 10c. 2719—SD. 6. Latex replica of exterior of pedicle valve (posterior view X 4). USNM 126162A. USGS loc. 2719—SD. 7. Impression of exterior of brachial valve (X 1). USGS 126139B. Seboomook Formation. USGS 10c. 2761—SD, Brassua Lake quadrangle, Somerset County, Maine. 8, 11. Impression and latex replica of interior of pedicle valve (X 1). USNM 126156B. Seboomook Formation. USGS 3091—SD, Long Pond quadrangle, Somerset County, Maine. 9, 10. Impression and latex replica of interior of pedicle valve (X 3). USNM 126124. Lower sandstone of Tarratine Formation. USGS loc. 2720—SD, Long Pond quadrangle, Somerset County, Maine. 12—14. “Schuche'rtella” sp. (p. 24). Tomhegan Formation. USGS loc. 2820—SD, Brassua Lake quadrangle, Somerset County, Maine. 12. Latex replica of interior of brachial valve (posterior view X 2). USNM 126148A. 13, 14. Latex replicas of interior and exterior of brachial valve (X 1). USNM 126148A, B. 15—17. Hipparionyx sp. (p. 24). McKenney Pond-s member of Tarratine Formation. USGS 10c. 2810—SD, Pierce Pond quadrangle, Som- erset County, Maine. X 1. 15. Impression of interior of pedicle valve. USNM 126149A. 16. Interior of pedicle valve. USNM 126149B. 17. Impression of interior of brachial valve. USNM 1261500. 18—20. Leptaem'sca sp. (p. 25). Beck Pond Limestone. USGS 10c. 3499—SD, Spencer Lake quadrangle, Somerset County, Maine. X 2. 18. Exterior of pedicle valve (posterior view). USNM 126129. 19, 20. Partly exfoliated pedicle valves. USNM 126129, 1261288. 21—37. Cyrtomfiscus nectus (Clarke, 1907) (p. 25). Tomhegan Formation. USGS 10c. 2852—SD, Brassua Lake quadrangle, Somerset County, Maine. 21. Latex replica of interior of pedicle valve (X 3). Specimen misplaced. 22. Impression of interior of pedicle valve (X 4). USNM 127392B. 23, 24. Latex replicas of exterior and intr of brachial valve (X 4). USNM 126223B, A. ' 25, 26. Impression of interior and latex replica of exterior of pedicle valve (X 4). USNM 126229. 27, 35. Impressions of exterior and interior of brachial valve (X 4). USNM 126223B, A. 28. Impression of exterior of brachial valve (X 3). USNM 126248. 29, 30. Impressions of interior and exterior of brachial valve (X 4). USNM 126228. 31. Latex replica of interior of pedicle valve (X 3). USNM 126247. 32—34. Latex replicas of interior of brachial valve (X 4). USNM 126227, 126222, 126223A. 36. Latex replica of exterior of pedicle valve (X 4). USNM 126231A. 37. Impression of interior of pedicle valve (X 4). USNM 127391A. GEOLOGICAL SURVEY PROFESSIONAL PAPER 784 PLATE 9 34 35 “SCHUCHERTELLA”, HIPPARIONYX, LEPTAENISCA, AND CYRTONISCUS PLATE 10 FIGURES 1—7. Eccentricosta sp. (p. 25). Hardwood Mountain Formation. Attean quadrangle, Somerset County, Maine. 1. Impression of exterior of brachial valve (X 2). USNM 160121. USGS 10c. 5583—SD. 2. Impression of exterior of pedicle valve (X 3). Note impression of spines and undulating costellae. USNM 160122A. USGS 10c. 5995—SD. 3. Impression of interior of pedicle valve (X 3). Note the short myophragm at the posterior of the valve. USNM 1601223. USGS loc. 5995—SD. 4. Impression of exterior of pedicle valve (X 2). Note impression of spines and undulating costellae. USNM 160123. USGS loc. 4841—SD. 5. Impression of exterior of pedicle valve (X 2). Note undulating costellae. USNM 160124. USGS 10c. 5583—SD. 6. Impression of interior of pedicle valve (X 3). USNM 160125. USGS loc. 4841—SD. 7. Impression of exterior of pedicle valve (X 2). Note impressions of spine bases and undulating costellae. USNM 160126. USGS 10c. 4841—SD. 8—12. Dawsonelloicles canadensis (Billings, 1874) (p 26). Lower sandstone of Tarratine Formation, Somerset County, Maine. 8. Impression of interior of pedicle valve (X 1). USNM 126244. USGS 10c. 2813—SD, Pierce Pond quadrangle. 9. Impression of interior of brachial valve (x 2). USNM 126246. USGS 10c. 2813—SD, Pierce Pond quadrangle. 10. Impression of interior of pedicle valve (X 1). USNM 126233. USGS loc. 2701—SD, Brassua Lake quadrangle. 11. Impression of interior of pedicle valve (X 1). USNM 126251B. USGS 10c. 2701—SD, Brassua Lake quadrangle. 12. Latex replica of exterior of brachial valve (posterior view X 2). USNM 126252. USGS 10c. 2701—SD, Brassua Lake quad- rangle. GEOLOGICAL SURVEY PROFESSIONAL PAPER 784 PLATE 10 11 ECCENTRICOSTA AND DA WSONELLOIDES PLATE 11 FIGURE 1—7. Dawsonelloides canadensis (Billings, 1874) (p. 26). Lower sandstone of Tarratine Formation. Somerset County, Maine. 1, 2. Latex replicas of interior of pedicle and brachial valves (X 1). USNM 126250, 126245. USGS loc. 2701—SD, Brassua Lake quadrangle. 3, 4. Latex replicas of interior of pedicle valve (X 1, X 4). USNM 126233, 126253B. USGS 10c. 2813—SD, Brassua Lake quad- rangle. 5. Latex replica of interior of brachial valve (X 1). USNM 126239A, USGS loc 2701—SD, Brassua Lake quadrangle. 6. Latex replica of exterior of pedicle valve (X 1). USNM 1262530. USGS loc. 2813—SD, Pierce Pond quadrangle. 7. Impression of interior of pedicle valve (X 1). USNM 126250. USGS 10c. 2701—SD, Brassua Lake quadrangle. 8—16. Eodevonaria arcuata (Hall, 1857) (p. 27). Tomhegan Formation. USGS 10c. 2820—SD, Brassua Lake quad- rangle, Somerset County, Maine. 8, 9. Impressions of interior of pedicle valve (X 2). USNM 126235A, 126230A. 10. Latex replica of exterior of pedicle valve (X 2). USNM 126230B. 11, 14. Latex replica and impression of interior of brachial valve (X 2). USNM 126238A. 12, 16. Impression and latex replica of exterior of brachial valve (X 2, X 4). USNM 126234B, A. 13. Impression of interior of pedicle valve (X 2). USNM 126232A. 15. Impression of exterior of brachial valve (X 2). USNM 126226A. 17—25. Chonostrophiella complamata. (Hall, 1857) (p. 28). Lower sandstone of Tarratine Formation. USGS loc. 2718—SD, Long Pond quadrangle, Somerset County, Maine. ‘ 17, 23. Impression and latex replica of interior of brachial valve (X 2). USNM 126236A. 18. Impression of interior of pedicle valve (X 2). USNM 126254B. 19. Impression of interior of brachial valve (X 2). USNM 126241. 20. Impression of interior of brachial valve (X 1). USNM 1262370. 21, 22. Impression and latex replica of interior of pedicle valve (X 1). USNM 126225. 24, 25. Latex replicas of interior and exterior of pedicle valve (X 2). USNM 126240, 126243. GEOLOGICAL SURVEY PROFESSIONAL PAPER 784 PLATE 11 DA WSONELLOIDES, EODEVONARIA, AND CHONOSTROPHIELLA PLATE 12 FIGURES 1, 2. Chonostrophiella complaxnata, (Hall, 1857) (p. 28). Latex replica of interior (X 2) and exterior (X 1) of pedicle valve. USNM 126242A, B. Lower sandstone of Tarratine For- mation. USGS loc. 2718—SD, Long Pond quadrangle, Somerset County, Maine. 3—11. Cupularostrum macroeosta, Boucot, n. sp. (p. 29). Tomhegan Formation, Brassua Lake quadrangle, Somerset County, Maine. . 3. Latex replica of exterior of pedicle valve (X 1). USNM 125856. USGS loc. 2750—SD. 4. Impression of the interior of pedicle valve (X 1). USNM 125857. USGS 10c. 2842—SD. 5. Impression of interior of brachial valve (posterior view x 3). USNM 125855A. USGS loc. 2750—SD. 6, 7. Impressions of interior of brachial valve (X 3, x 1). USNM 125904B. USGS loc. 2750—SD. 8, 10. Impression (posterior view x 3) and latex replica (X 2) of exterior of brachial valve. USNM 125852. USGS loc. 2842—SD. 9. Impression of interior of pedicle valve (X 1). Holotype, USNM 125851A. USGS 10c. 2750—SD. 11. Latex replica of interior of pedicle valve (X 1). USNM 125852A. USGS loc. 2842—SD. 12—21. Cupularostrum? sp. (p. 30). Beck Pond Limestone. USGS 10c. 3601—SD, Spencer Lake quad- rangle, Somerset County, Maine. X 3. 12, 13, 16, 19, 21. Exterior: brachial valve, pedicle valve, pos» terior, side, and anterior views. USNM 125859. 14, 15, 17, 18, 20. Exterior: side, anterior, pedicle valve, brachial valve, and anterior views. USNM 125862. 22—26. Ancillotoechia sp. (p. 30). Exterior: side, posterior, anterior, and brachial valve views (X 3); pedicle valve (X 4). USNM 125902. Beck Pond Limestone. USGS loc. 3499—SD, Spencer Lake quadrangle, Somerset County, Maine. 27—31. “Ancillotoechia” cf. A. altisulcata (Amsden, 1951) (p. 31). Exterior: posterior, anterior, side, brachial valve, and pedicle valve views (X 2). USNM 125854. Hardwood Mountain Forma- tion. USGS loc. 3488—SD, Spencer Lake quadrangle, Somerset, County, Maine. 32—36. Sphaerirhynchia sp. 1 (p. 31). Exterior: side, pedicle valve, brachial valve, posterior, and an- terior views (X 2). USNM 125874. Hardwood Mountain For- mation. USGS loc. 3470—SD, Spencer Lake quadrangle, Somer- set, County, Maine. GEOLOGICAL SURVEY PROFESSIONAL PAPER 784 PLATE 12 17 21 23 .25 26 CHONOSTROPHIELLA, CUPULAROSTR UM, CUPULAROSTR UM?, ANCILLOTOECHIA, “ANCILLOTOECHIA”, AND SPHAERIRHYNCHIA PLATE 13 FIGURES 1—6. Sphaen’rhynchia sp. 2 (p. 32‘). Beck Pond Limestone. USGS 10c. 3499-SD, Spencer Lake quad- rangle, Somerset County, Maine. 1, 3—6. Exterior: pedicle valve, posterior, brachial valve, anterior, and side views (X 3). USNM 125876. 2. Impression of interior of brachial valve (X 1). USNM 125877. 7—12. Sphaerirhynchia sp. 3 (p. 33). Beck Pond Limestone. USGS 10c. 3601—SD, Spencer Lake quad- rangle, Somerset County, Maine. X 1. 7. Impression of interior of pedicle valve. USNM 125871. 8—12. Exterior: brachial valve, pedicle valve, posterior, anterior, and side views. USNM 125870. 13—18. Sphaerirhynchia? sp. (p. 33). Hardwood Mountain Formation. USGS loc. 3488—SD, Spencer Lake quadrangle, Somerset, County, Maine. 13—15. Impression of interior of pedicle valve, brachial valve, and posterior view (X 2). USNM 127385A. 16, 17. Exterior: side and anterior views (X 2). USNM 127385A. 18. Section of brachial and pedicle valves (X 5). USNM 127385B. 19—25. Sulcatina sp. (p. 33). Hardwood Mountain Formation. USGS 10c. 3488—SD, Spencer Lake quadrangle, Somerset County, Maine. X 1. 19. Latex replica of exterior (posterior View). USNM 125853A. 20, 25. Partly exfoliated exterior (side and posterior views). USNM 125847B. 21. Exterior of brachial valve. USNM 125848. 22. Impression of interior (anterior view). USNM 125847A. 23, 24. Impression of interior of pedicle and brachial valves. USNM 125850. 26—28. Sphaem'rhynchia sp. 1 (p. 31). Hardwood Mountain Formation. USGS loc. 5586—SD, Attean quad- rangle, Somerset County, Maine. X 2. 26, 27. Impressions of interior (posterior view). USNM 160127, 160128. 28. Impression of interior of pedicle valve. Note raised impres- sions for adductor atttachment surrounded by diductor field. USNM 160129. PROFESSIONAL PAPER 784 PLATE 13 GEOLOGICAL SURVEY 26 SPHAERIRHYNCHIA , SPHA ERIRHYNCHIA ?, AND SULCA TINA PLATE 14 FIGURES 1—5. Costellirostra sp. (p. 34). Exterior: brachial valve, pedicle valve, posterior, anterior, and side views (X 3). USNM 125885. McKenney Ponds Member of Tarratine Formation. USGS 10c. 2806—SD, Pierce Pond quad- rangle, Somerset County, Maine. 6-13. Eatom'a cf E. medialis (Vanuxem, 1842) (p. 34). Beck Pond Limestone. USGS 10c. 3499—SD, Spencer Lake quad- rangle, Somerset County, Maine. 6, 8—11. Exterior: side (X 2), brachial valve (X 1), anterior (X 2), posterior (X 2), and pedicle valve (X 2) views. USNM 125882B. 7, 12. Impressions of interior of pedicle (X 1) and brachial (X 3) valves. USNM 125873. 13. Impression of interior of brachial valve (X 1). USNM 125872. 14—21. Machaeraria mainensis Boucot, n. sp. (p. 35). Base of Hobbstown Formation. USGS loc. 3479-SD, Spencer Lake quadrangle, Somerset County, Maine. 14, 15. Impression and latex replica of interior of pedicle valve (X 2). USNM 125867A. 16. Latex replica of exterior of pedicle valve (X 2). USNM 125867A. 17, 19. Impression (X 3) and latex replica (X 4) of interior of brachial valve. Holotype, USNM 125869A. 18. Latex replica of interior (posterior view X 4). Holotype, USNM 125869A. 20. Latex replica of exterior of pedicle valve (X 3). USNM 125868. 21. Latex replica of exterior (posterior view X 3). USNM 125868. 22—27. Atrypa cf. A. termesseensis Amsden, 1949 (p. 36). Base of Hobbstown Formation. USGS loc. 3479—SD, Spencer Lake quadrangle, Somerset, County, Maine. 22, 23. Impressions of exterior of pedicle and brachial valves (X 2). USNM 126003A, B. 24. Impression of interior of brachial valve (X 2). USNM 1260030. 25, 27. Latex replica and impression of interior of pedicle valve (X 3). USNM 126002. 26. Latex replica of exterior of pedicle valve (X 3). USNM 126004. 28, 29. Atrypa cf. A. arctostriata Foerste, 1903 (p. 36). Impressions of exterior and interior of brachial valve (X 3). USNM 126014A, 126015B. Base of Hobbstown Formation. USGS Ioc. 3479—SD, Spencer Lake quadrangle, Somerset County, Maine. GEOLOGICAL SURVEY PROFESSIONAL PAPER 784 PLATE [4 26 27 28 COSTELLIROSTRA, EATONIA, MACHAERARIA, AND ATR YPA PLATE 15 FIGURES 1—6. Atrypa “reticulam’s” (Linnaeus, 1767) (p. 36). 1—3. Tomhegan Formation. USGS 10c. 2750—SD, Brassua Lake quadrangle, Somerset County, Maine. X 1. 1, 3. Impression and latex replica of interior of pedicle valve. USNM 126011. 2. Latex replica of exterior of pedicle valve. USNM 126010A. 4. Exterior of brachial valve (X 1). USNM 126008A. Hardwood Mountain Formation. USGS loc. 3488—SD, Spencer Lake quadrangle, Somerset County, Maine. 5, 6. Latex replica and impression of interior of brachial valve (X 1). USNM 126068B. Tomhegan For— mation. USGS 10c. 2820—SD, Brassua Lake quadrangle, Somerset County, Maine. 7—15. Lissatrypa sp. (p. 37). 7—13. Hardwood Mountain Formation. Spencer Lake quadrangle, Somerset County, Maine. 7, 8. Latex replicas of interior of pedicle (X 2) and brachial (X 3) valves. USNM 125908B, A. USGS loc. 3473—SD. 9. Partly exposed spire (X3). USNM 126000. USGS loc. 3488—SD. 10, 11. Impression and latex replica of interior of brachial valve (X 3). USNM 1259058, A. USGS 10c. 3473—SD. 12. Impression of interior of pedicle valve (X 2). USNM 125908B. USGS loc. 3473—SD. 13. Impression of interior of brachial valve (X 3). USNM 125905A. USGS loc. 3473—SD. 14, 15. Impressions of interior and exterior of pedicle valve (X 3). Note the trapezoidal impression of the muscle field (fig. 14) and faint, concentric growth lines (fig. 15). USNM 160130A, B. Hard- wood Mountain Formation. USGS loc. 4843—SD, Attean quadrangle, Somerset County, Maine. 16. Nanospim? sp. (p. 37). Partly exfoliated exterior of brachial valve (X 5). USNM 126108A. Base of Hobbstown Formation. USGS loc. 3479—SD, Spencer Lake quadrangle, Somerset County, Maine. 17—24. Leptocoelia flabellites (Conrad, 1841) (p. 38). 17—19, 22—24. Lower sandstone of Tarratine Formation. USGS loc. 2718—SD, Long Pond quadrangle, Somerset County, Maine. 17. Impression of interior of brachial valve (X 2). USNM 126005B. 18, 22. Impression and latex replica of interior of pedicle valve (X 2). USNM 126001. 19. Impression of exterior of pedicle valve (X 2). USNM 1260‘05A. 23. Latex replica of interior of brachial valve (X 2). USNM 127387. 24. Latex replica of exterior of brachial valve (X 3). USNM 126012. 20. Exterior of brachial valve (X 2). USNM 126009. Seboomook Formation. USGS 10c. 3094—SD, Long Pond quadrangle, Somerset County, Maine. 21. Exterior of brachial valve (X 2). USNM 125907. Seboomook Formation. USGS 10c. 2880—SD, Attean quadrangle, Somerset County, Maine. 25—32. Coelospim sp. (p. 38). 25—28. Hardwood Mountain Formation. USGS loc. 3488—SD, Spencer Lake quadrangle, Somerset County, Maine. X 4. 25, 26. Impressions of interior of pedicle and brachial valves. USNM 1260070. 27. Latex replica of exterior of brachial valve. USNM 126007A. 28. Latex replica of exterior of brachial valve. USNM 126007AA. 29—32. Hardwood Mountain Formation. USGS loc. 5995—SD, Attean quadrangle, Somerset County, Maine. X 3. 29. Impression of interior of pedicle valve. USNM 160131. 30. Impression of interior of pedicle valve (turned to side) USNM 160132A. 31, 32. Impressions of exterior and interior of pedicle valve. USNM 160132B. 160133. 33—35. Hedeina, cf. H. macropleura (Conrad, 1840) (p. 39). 33. Partly exfoliated exterior of brachial valve (X 1). USNM 126076A. Beck Pond Limestone. USGS loc. 3499—SD, Spencer Lake quadrangle, Somerset County, Maine. 34. Exterior of pedicle valve (X 1). USNM 126006. McKenney Ponds Member of Tarratine Forma- tion. USGS loc. 2806—SD, Pierce Pond quadrangle, Somerset County, Maine. 35. Impression of interior of brachial valve (posterior view X 2). USNM 126076A. Beck Pond Lime- stone. USGS 10c. 3499—SD, Spencer Lake quadrangle, Somerset County, Maine. PROFESSIONAL PAPER 784 PLATE 15 GEOLOGICAL SURVEY 1 - 16 | 20 25 27 28 33 ATR YPA, LISSATR YPA, NANOSPIRA?, LEPTOCOELIA, COELOSPIRA, AND HEDEINA 35 FIGURES 1—5. 10—18. 19-25. PLATE 16 Delthyris cf. D. kozlowskii Amsden, 1951 (p. 39). Base of Hobbstown Formation. USGS loc. 3479—SD, Spencer Lake quadrangle, Somerset County, Maine. Impression of interior of pedicle valve (X 2). USNM 126038. Impression of interior of brachial valve (X 3. USNM 126027A. Latex replica of exterior of pedicle valve (X 2). USNM 126036B. Impression of exterior of pedicle valve (X 6). USNM 126036B. . Impression of interior of pedicle valve (X 2). USNM 126025B. Howellella? sp. (p. 39). 6. Cross section (X 3), showing presence of short dental lamellae and absence of a median septum. USNM 160134. Hardwood Mountain Formation. USGS 10c. 4841—SD, Attean quadrangle, Somerset County, Maine. 7. Impression of interior of pedicle valve (X 3). USNM 126020. Hardwood Mountain Formation. USGS 10c. 2950—SD, Spencer Lake quadrangle, Somerset County, Maine. Howellella? cf. H. cyclopte'rus (Hall, 1857) (p. 40). Latex replica and impression of exterior of pedicle valve (X 1). USNM 126257A. Beck Pond Limestone. USGS loc. 3499—SD, Spencer Lake quadrangle, Somerset County, Maine. “Howellella” tomheganensis Boucot, n. sp. (p. 40). Tomhegan Formation. USGS 10c. 2750—SD, Brassua Lake quad- rangle, Somerset County, Maine. 10. Latex replica of interior of pedicle valve (X 1). USNM 126017B. 11. Latex replica of interior of brachial valve (X 2). USNM 126032B. 12. Latex replica of exterior of brachial valve (X 1). Holotype, USNM 126016B. 13. Impression of interior of brachial valve, posterior view (X 2). Holotype, USNM 126016A. 14. Impression of interior of brachial valve (X 1). Holotype, USNM 126016A. 15. Latex replica of exterior of pedicle valve (X 2). USNM 126026A. 16. Impression of interior of pedicle valve (X 3). USNM 125026A. 17, 18. Latex replica and impression of interior of pedicle valve (X 2). USNM 126045B, A. Acrospim‘fer murahisoni (Castelnau, 1843) (p. 41). Lower sandstone of Tarratine Formation. USGS loc. 2718—SD, Long Pond quadrangle, Somerset County, Maine. X 1. 19. Latex replica of exterior of pedicle valve. USNM 126047A. 20, 23. Latex replica and impression of interior of pedicle valve. USNM 126047D. 21, 24. Latex replica and impression of interior of brachial valve. USNM 126043. 22. Latex replica of exterior of brachial valve. USNM 127386B. 25. Impression of interior of pedicle valve. USNM 126042A. mewwr GEOLOGICAL SURVEY PROFESSIONAL PAPER 784 PLATIi l6 23 DELTHYRIS, HOWELLELLA?, “HOWELLELLA”, AND ACROSPIRIFER PLATE 17 FIGURES 1—9. Acrospi'rifer atlanticus (Clarke, 1907) (p. 46). Tomhegan Formation. Brassua Lake quadrangle, Somerset County, Maine. 1, 2. Impression and latex replica of interior of brachial valve (x 1). USNM 126041B. USGS 10c. 2750—SD. 3. Latex replica of exterior of brachial valve (X 1). USNM 126041A. USGS 10c. 2750—SD. 4, 9. Latex replica of exterior of brachial valve (X 1, X 4). USNM 1261638. USGS 10c. 2750—SD. 5, 6. Impression and latex replica of interior of pedicle valve (x 1). USNM 126019B. USGS 10c. 28-20—SD. 7. Latex replica of interior of pedicle valve (X 1). USNM ' 127388B. USGS 10c. 2750—SD. 8. Impression of interior of pedicle valve (x 1). USNM 126031B. USGS 10c. 2820—SD. 10—14. Acrospim'fer sp. 1 (p. 47). Lower sandstone of Tarratine Formation. USGS loc. 2796—SD, Pierce Pond quadrangle, Somerset County, Maine. X 1. 10, 13. Impression and latex replica of interior of brachial valve. USNM 126090. 11, 12. Latex replica and impression of interior of pedicle valve. USNM 126075. 14. Latex replica of exterior of brachial valve. USNM 126090. PROFESSIONAL PAPER 784 PLATE I7 GEOLOGICAL SURVEY ACROSPIRIFER PLATE 18 FIGURES 1—5. Acrospirifer sp. 2 (p. 48). Lower sandstone of Tarratine Formation. USGS loc. 2872—SD, Pierce Pond quadrangle, Somerset County, Maine. 1, 2. Latex replica (X 3) and impression (X 1) of interior of pedicle valve. USNM 126113. 3. Latex replica of exterior of brachial valve (X 1). USNM 126089B. 4, 5. Latex replica and impression of interior of brachial valve (X 1). USNM 126089A. 6—14. Antispirifer harroldi Williams and Breger, 1916 (p. 50). Lower sandstone of Tarratine Formation. Long Pond quadrangle, Somerset County, Maine. 6. Impression of interior of brachial valve (X 1). USNM 126039. USGS 10c. 2720—SD. 7. Impression of interior of pedicle valve (X 1). USNN 126040. USGS loc. 2721—SD. 8, 10. Latex replica (X 1) and impression (X 3) of exterior of brachial valve. USNM 126024B. USGS 10c. 2721—SD. 9. Latex replica of exterior of pedicle valve (X 2). USNM 126028A. USGS 10c. 2722—SD. 11. Impression of interior of brachial valve (X 2). USNM 126034B. USGS loc. 2720—SD. 12, 14. Latex replica and impression of interior of brachial valve (X 2). USNM 126034B. USGS 10c. 2720—SD. 13. Latex replica of exterior of pedicle valve (X 2). USNM 126033B. USGS loc. 2720—SD. 15—20. Costellispirifer perimele (Clarke, 1907) (p. 51). Tomhegan Formation. USGS loc. 2750—SD, Brassua Lake quad- rangle, Somerset County, Maine. 15. Impression of interior of pedicle valve (X 1). USNM 126029B. 16. Latex replica of exterior of brachial valve (X 2). USNM 126037A. 17. Latex replica of exterior of pedicle valve (X 1). USNM 126035A. 18, 19. Latex replica and impression of interior of brachial valve (X 1). USNM 126030B. . 20. Impression of interior of pedicle valve (X 1). USNM 126035B. 21, 22. “Mucrospirifer”c’f. “M.” macro (Hall, 1857) (p. 60). Ventral and posterior views of internal mold (X 2). USNM 160135. Tomhegan Formation. USGS loc. 2723—SD, Brassua Lake quadrangle, Somerset County, Maine. 23—26. Costispim’fer sp. (p. 61). 23, 25. Latex replica of exterior and impression of interior of pedicle valve (X 1). USNM 126023B, A. Seboomook For- mation. USGS 10c. 3482—SD, Spencer Lake quadrangle, Somerset County, Maine. 24, 26. Impression of interior and latex replica of exterior of brachial valve (X 1). USNM 126018B, A. Lower sand- stone of Tarratine Formation. USGS loc. 2767—SD, Moosehead Lake quadrangle, Somerset County, Maine. GEOLOGICAL SURVEY PROFESSIONAL PAPER 784 PLATE 18 ACROSPIRIFER, ANTISPIRIFER, COSTELLISPIRIFER, “MUCROSPIRIFER”, AND COSTISPIRIFER PLAT-'E 19 FIGURES 1—6. Megakozlowskiella sp. (p. 62). Beck Pond limestone. Spencer Lake quadrangle, Somerset County, Maine. X 2. 1. Impression of interior of pedicle valve. USNM 126044B. USGS 10c. 3497—SD. 2—6. Exterior: brachial valve, pedicle valve, side view, posterior View, and anterior view. USNM 126046. USGS loc. 3601—SD. 7—11. Metaplasia cf. M. paucicostata (Schuchert, 1913) (p. 62). McKenney Ponds Member of Tarratine Formation. USGS loc. 2806—SD, Pierce Pond quadrangle, Somerset County, Maine. 7. Latex replica of exterior of pedicle valve (X 3). USNM 126115A. 8, 9. Impression of interior of brachial and pedicle valves (X 4). USNM 126115B. 10, 11. Latex replica (x 3) and impression of interior of pedicle valve. USNM 1261150. 12—18. Metaplasia minute Boucot, 1959 (p. 62). Tomhegan Formation. USGS loc. 2750—SD, Brassua Lake quad- rangle, Somerset County, Maine. 12. Latex replica of exterior of pedicle valve (posterior view X 4). USNM 126110B. 13, 17. Impression and latex replica of interior of brachial valve (X 4). USNM 126086A. 14. Latex replica of exterior of pedicle valve (X 3). USNM 126067A. 15. Impression of interior of pedicle valve (X 3). USNM 126067B. 16. Impression of interior of pedicle valve (posterior view (X 4). USNM 1261100. 18. Latex replica of exterior of brachial valve (X 4). USNM 126086B. 19—29. Plicoplasia plicata (Weller, 1903) (p. 63). 19—28. Lower sandstone of Tarratine Formation USGS 10c. 2813— SD, Pierce Pond quadrangle, Somerset County, Maine. 19. Latex replica of exterior of brachial valve (X 2). USNM 126114A. 20. Latex replica of exterior of pedicle valve (X 2). USNM 126087A. 21. Impression of interior of pedicle valve (X 2). USNM 126087B. 22. Impression of interior of brachial valve (X 2). USNM 126114B. 23. Latex replica of interior of brachial valve (X 3). USNM 126106. 24, 25. Latex replica and impression of interior of pedicle valve (X 3). USNM 126260. 26, 27. Latex replica and impression of interior of pedicle valve (X 3). USNM 126259. 28. Latex replica of exterior of pedicle valve (X 3). USNM 126087A. 29. Impression of interior of brachial valve (X 3). Seboomook Formation. USGS loc. 2884—SD, Chesuncook quadrangle, Picataquis County, Maine. USNM 126256B. 30—32. Cyrtina? cf. C. rostrata (Hall, 1857), (p. 63). Brachial, posterior, and side views of internal mold (X 2). USNM 160136. Seboomook Formation. USGS 10c. 2883—SD, Spencer Lake quadrangle, Somerset County, Maine. GEOLOGICAL SURVEY PROFESSIONAL PAPER 784 PLATE 19 32 MEGAKOZLOWSKIELLA, METAPLASIA, PLICOPLASIA, AND CYRTINA? PLATE 20 FIGURES 1—6. Merista cf. M. tennesseensis Hall and Clarke, 1895 (p. 63). Hardwood Mountain Formation. USGS loc, 3488—SD, Spencer Lake quadrangle, Somerset County, Maine. 1—4. Exterior: side, anterior, posterior, and pedicle valve valve (partly exfoliated) views (X 1). USNM 126105B. 5. Impression of interior of pedicle valve (X 2). USNM 126085. 6. Exterior of brachial valve (X 1). USNM 126105B. 7-13. Meristclla lata (Hall, 1859), (p. 64). 7, 8, 10—12. Lower sandstone of Tarratine Formation. Long Pond quadrangle, Somerset County, Maine. 7. Latex replica of exterior of pedicle valve (X 1). USNM 126116A. USGS 10c. 2721—SD. 8. Impression of spire (X 1). USNM 126079A. USGS 10c. 2777—SD. 10. Latex replica of interior of pedicle valve (X 2). USNM 127390. USGS 10c. 2777r—SD. 11, 12. Latex replica and impression of interior of brachial valve (x 2). USNM 126261A.. USGS loc. 2721—SD. 9. Impression of interior of brachial valve (X 1). USNM 126103. Seboomook Formation. USGS 10c. 3482—SD, Spencer Lake quadrangle, Somerset County, Maine. 13. Impression of interior of brachial valve (x 2). USNM 126082B. Seboomook Formation. USGS loc. 2884—SD, Chesuncook quadrangle, Piscataquis County, Maine. 14—22. Charionoides dom's (Hall, 1860), (p. 64). Tomhegan Formation. Brassua Lake quadrangle, Somerset County, Maine. 14—16. Impression of interior of pedicle and brachial valves and side View (X 2). USNM 1261118. USGS loc. 2820—SD. 17. Latex replica of beak of pedicle valve (X 3). USNM 126117B. USGS loc. 2750—SD. 18. Impression of interior of brachial valve (X 1). USNM 126102. USGS loc. 2750—SD. 19, 20. Latex replica and impression of interior of brachial valve (X 2). USNM 126081. USGS loc. 2750—SD. 21. Impression of interior of pedicle valve (X 2). USNM 126080A. USGS 10c. 2750—SD. 22. Impression of interior of pedicle valve (X 2). USNM 126092. USGS loc. 2750—SD. 23—27. Nucleospim sp. (p. 65). Base of Hobbstown Formation. USGS loc. 3479—SD, Spencer Lake quadrangle, Somerset County, Maine. 23, 24. Impression and latex replica of interior of pedicle valve (X 3). USNM 1260918. 25, 27. Latex replica and impression of interior of brachial valve (x 4). USNM 126262. 26. Latex replica of exterior of pedicle valve (X 4). USNM 126091A. 28—37. Protathyris sp. (p. 65). Hardwood Mountain Formation. USGS 10c. 2822—SD, Attean quadrangle, Somerset County, Maine. X 2. 28—31, 33. Exterior: brachial valve, pedicle valve, posterior, an- terior, and side views. USNM 126262. 32, 34—37. Exterior: posterior, pedicle valve, side, brachial valve, and anterior views. USNM 126070. GEOLOGICAL SURVEY PROFESSIONAL PAPER 784 PLATE 20 .4 3O 13 MERISTA, MERISTELLA, CHARIONOIDES, NUCLEOSPIRA, AND PROTA THYRIS PLATE 21 FIGURES 1—6. Nanothyn's hodgei Boucot, n. sp. (p. 65). Lower sandstone of Tarratine Formation. USGS loc. 2832—SD, Brassua Lake quadrangle, Somerset County, Maine. 1. Latex replica of interior of brachial valve (X 3). Holotype, USNM 126194. 2. Impression of exterior (posterior view X 2). Paratype, USN M 125943. 3, 4. Impression of interior of brachial and pedicle valves (X 2). Paratype, USNM 126191. 5. Latex replica of exterior of pedicle valve (X 3). Paratype, USNM 126183A. 6. Latex replica of exterior of brachial valve (X 3). Paratype, USNM 126220A. 7—13. Nanothyris cf. N. subglobosa (Weller, 1903) (p. 66). Beck Pond Limestone. USGS Ice. 3601— SD, Spencer Lake quad- rangle, Somerset County, Maine. 7—11. Exterior: pedicle valve, brachial valve, side, posterior, and anterior views (X 2). USNM 126192A. 12. Exterior of pedicle valve (X 1). USNM 126192B. 13. Cross section of brachial valve (X 3). USNM 126187. 14—17. Rensselaeriu sp. (p. 66). Lower sandstone of Tarratine Formation. USGS loc. 2767—SD, Moosehead Lake quadrangle, Somerset County, Maine. X 2. 14, 15. Impressions of interior of brachial valves. USNM 126216C, D. 16, 17. Impressions of interior of pedicle valves. USNM 126216B, 126214. 18—30. Beach'ia thum'i (Clarke, 1907) (p. 69). Lower sandstone of Tarratine Formation. Somerset County, Maine. 18, 23—25, 29. Exterior: brachial valve, anterior, side, posterior, and pedicle valve views (X 1). USNM 126198. USGS 10c. 2813—SD, Pierce Pond quadrangle. 19. Impression of interior of pedicle valve (X 1). USNM 126207B. USGS loc. 2796—SD, Pierce Pond quad- rangle. 20. Latex replica of interior of pedicle valve (X 1). USNM 1262170. USGS 10c. 2813—SD, Pierce Pond quadrangle. 21. Impression of interior of brachial valve (X 1). USNM 126206A. USGS 10c. 2796—SD, Pierce Pond quadrangle. 22. Impression of interior of brachial valve (X 1). USNM 126202. USGS loc. 2813—SD, Pierce Pond quadrangle. 26, 30. Impression and latex replica of interior of pedicle valve (X 1). USNM 126200. USGS 10c. 2731—SD, Long Pond quadrangle. 27. Latex replica of interior of brachial valve (X 1). USNM 126199. USGS 10c. 2813—SD, Pierce Pond quadrangle. 28. Impression of interior of brachial valve (X 3). USNM 126203. USGS loc. 2813—SD, Pierce Pond quadrangle. GEOLOGICAL SURVEY PROFESSIONAL PAPER 784 PLATE 21 29 NANOTHYRIS, RENSSELAERIA, AND BEACHIA PLATE 22 FIGURES 1—5. Beachia thum‘i (Clarke, 1907) (p. 69). 1. Impression of interior of brachial valve (X 1). NYSM 8436. Unnamed sandstone. Telos Lake Dam, Telos quadrangle, Pis- cataquis County, Maine. 2. Impression of interior of brachial valve (X 1). USNM 126199. Lower sandstone of Tarratine Formation. USGS 10c. 2813— SD, Pierce Pond quadrangle, Somerset County, Maine. 3. Impression of interior of brachial valve (X 1). USNM 126204. Seboomook Formation. USGS loc. 2884—SD, Chesuncook quad- rangle, Piscataquis County, Maine. 4. Latex replica of interior of pedicle valve (X 2). USNM 126205. Lower sandstone of Tarratine Formation. USGS Ice. 2813—— SD, Pierce Pond quadrangle, Somerset County, Maine. 5. Impression of exterior of pedicle valve (X 2). USNM 126209. Lower sandstone of Tarratine Formation. USGS Ice. 2777— SD, Long Pond quadrangle, Somerset County, Maine. 6—13. Cloudothyris postovalis Boucot and Johnson, 1968 (p. 69). Tomhegan Formation. Brassua Lake quadrangle, Somerset County, Maine. 6. Impression of interior of brachial valve (X 2). USNM 126210A. USGS loc. 2750—SD. 7, 12. Impression and latex replica of interior of brachial valve (X 2). USNM 126218. USGS loc. 3238—SD. 8, 9. Latex replica and impression of interior of pedicle valve (X 1). USNM 126208A. USGS loc. 2750—SD. 10. Impression of cardinal process (X 5). USNM 126195. USGS 10c. 2814—SD. 11. Latex replica of interior of pedicle valve (posterior view X 3). USNM 126208A. USGS 10c. 2750—SD. 13. Latex replica of exterior of pedicle valve (X 1). USNM 126208B. USGS loc. 2750—SD. 14—24. Amphigenia parva Clarke, 1907 (p. 70). Tomhegan Formation. USGS 10c. 2750—SD, Brassua Lake quad- rangle, Somerset County, Maine. 14. Impression of interior of pedicle valve (posterior view X 2). USNM 126193B. 15. Impression of interior of pedicle valve (X 1). USNM 126190A. 16. Impression of interior of pedicle valve (posterior view X 2). USNM 126196A. 17. Impression of interior of pedicle valve (X 1). USNM 1262190. 18. Impression of interior of brachial valve (X 2). USNM 126197B. 19. Latex replica of exterior of brachial valve (X 1). USNM 126219B. 20, 21. Latex replica and impression of interior of brachial valve (X 1). USNM 126219A. 22. Impression of interior of pedicle valve (X 2). USNM 126196A. 23. Impression of interior of brachial valve (X 2). USNM 126188A. 24. Latex replica of interior of pedicle valve (X 2). USNM 127389. GEOLOGICAL SURVEY PROFESSIONAL PAPER 784 PLATE 22 21 22 23 BEACHIA, CLOUDOTHYRIS, AND AMPHIGENIA PLATE 23 FIGURES 1—5. Globithyris callida (Clarke, 1907) (p. 72). Tomhegan Formation. Brassua Lake quadrangle, Somerset County, Maine. ' 1. Exterior of pedicle valve (X 1). USNM 126165A. USGS 10c. 2713—SD. 2. Anterior View of interior of shell (X 5). USNM 126180. USGS loc. 2713—SD. 3. Exterior of brachial valve (X 2). USNM 126165B. USGS 10c. 2713—SD. 4. Interior view of brachial valve (X 5). USNM 126180. USGS 10c. 2713—SD. 5. Impression of interior of brachial valve (X 2). USNM 67778. East side of Brassua Lake opposite inlet of Moose River. 6—13. Globithym's diam'a (Clarke, 1907), (p. 73). Tomhegan Formation. Brassua Lake quadrangle, Somerset County, Maine. 6. Impression of interior (posterior view X 1). USNM 126182. USGS 10c. 2869—SD. 7—11. Impression of interior: brachial valve, pedicle valve, side, posterior, and anterior views (X 2). USNM 1261643. USGS loc. 2869—SD. 12, 13. Impression and latex replica of interior of brachial valve (X 2). USNM 126167A. USGS loc. 2715—SD. 14—26. Mutationella parlinensis Boucot, Caster, Ives, and Talent, 1963. (p. 73). Lower sandstone of Tarratine Formation. USGS 10c. 2718—SD, Long Pond quadrangle, Somerset County, Maine. 14—48. Impression of interior: posterior (X 2), pedicle valve (X 1), side (X 1), anterior (X 1), and brachial valve (X 2) views. Holotype, USNM 126178A. 19. Impression of interior of brachial valve (X 2). USNM 126168A. 20. Latex replica of exterior of pedicle valve (X 1). USNM 126173B. 21. Latex replica of exterior of brachial valve (X 2). USNM 126186. 22, 26. Latex replica and impression of interior of pedicle valve (X 2). USNM 126166B. 23, 25. Latex replica (X 3) and impression (X 2) of interior of pedicle valve. USNM 126171. 24. ImpressiOn of interior of brachial valve (X 2). USNM 126185A. fi' U5. GOVERNMENT PRINTING OFFICE : I973 0-499-380 GEOLOGICAL SURVEY PROFESSIONAL PAPER 784 PLATE 23 25 GLOBITHYRIS AND MUTA TIONELLA - v". EARTH ENC”) b’RARy f Petrology and Stratigraphy of 7; / tho F ra Mauro Formation at the Apollo 14 Site GEOLOGICAL SURVEY PROFESSIONAL PAPER 785 Prepared on behalf of the National Aeronautics and Space Administration DOCUMENTS DEPARTMENT AUG 2 81972 LIBRARY UNIV'EWSIW 0‘ CA'JFOR'J'A Potrology and Stratigraphy of the F ra Mauro Formation at the Apollo 14 Sito By H. G. WILSHIRE and E. D. JACKSON GEOLOGICAL SURVEY PROFESSIONAL PAPER 785 Prepared on behalf of the National Aeronautics and Space Administration Distribution of Apollo 14 fragmental rocks, Classified into four groups, suggests that the Fra Mauro Formation consists of stratified ejecta UNITED STATES GOVERNMENT PRINTING OFFICE, \VASHINGTON : 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—600103 For sale by the Superintendent of Documents, US. Government Printing Office Washington, D.C. 20402 CONTENTS Page Page Abstract ..................................................................................... 1 Thin-section petrology — Continued Introduction ........................................................ 1 Matrix materials .............................................................. 15 Returned samples .............................................. 2 Matrix-Clast reactions ..................................................... 16 Megascopic structures ...... .. 4 Partial fusion .................................................................... 17 Hand-specimen petrology ........................................................ 4 Deformation of mineral and lithic clasts ..... 17 Homogeneous crystalline rocks ...................................... 5 Compound clasts ............................................... 17 Fragmental rocks ............................................................. 5 Classification of the rocks.“ 19 Thin-section petrology ............................................................. 6 Rock distribution ...................................................................... 21 Glass clasts .............. 7 Fra Mauro Formation ............................................................. 22 Mineral clasts... .............................. 7 Conclusions ................................................................................ 23 Lithic clasts .......................................... 9 Fra Mauro Formation ..................................................... 23 Igneous rocks ............................................................. 9 Pre-Imbrium terrane.... 24 Metamorphic rocks ................................................... 11 References cited ..................................................... 25 ILLUSTRATIONS Page FIGURE 1. Regional geologic map of the area surrounding the Apollo 14 landing site ........................................................... 2 2. Map of major geologic features in the Apollo 14 traverse area ........................ 3 3. Surface photograph and photomicrographs of Apollo 14 rocks... ............................ 4 4. Scheme used for subdivision of Apollo 14 rocks .............................................................................................................. 5 5. Frequency distribution of clasts >1 mm in fragmental rocks of groups F1, F2, F3, and F4 ............................... 6 6. Histograms showing relative abundance of glass, mineral, igneous, and metamorphic clasts and microclasts in the fundamental rocks of groups F1, F2, F3, and F4 ......................................................... 8 7. Triangular plot of plagioclase-pyroxene-olivine microclasts ........................................ 9 8—11. Photomicrographs of Apollo 14 rocks ...................................................................................... .10, 13, 14,18 12. Triangular plots of clasts and microclasts of glass, igneous, and metamorphic rocks ........................................... 20 13. Distribution of fragmental rock types at the Apollo 14 landing site ....................................................................... 22 14. Map showing outer mountain rings of major lunar basins, their relative ages, and the extent of their continuous ejecta blankets ............................................................................................................... 25 TABLES Page TABLE 1. Classification of rocks larger than 1 gram ........................................................................................................................ 5 2. Principal types of clasts and microclasts ................................................................................ 7 3. Clast sizes and grain sizes of some igneous clasts .............................................................. 9 4. Modal compositions of basalts ........................................ 9 5. Types of vitrophyres ............................... 11 6. Modes and average grain sizes of common cumulates .................................................................................................... 11 7. Distribution of mineral reactions ............................................................................................................ 16 8. Clasts and matrix types in compound clasts .................................................................................................................... 19 III PETROLOGY AND STRATIGRAPHY OF THE FRA MAURO FORMATION AT THE APOLLO 14 SITE By H. G. \VILSHIRE ABSTRACT The Apollo 14 rocks returned to earth are divisible into two broad groups: relatively homogeneous crystalline rocks and fragmental rocks (breccias). The crystalline rocks are, for the most part, metaclastic, but a few, including the second largest sample returned, are aphyric basalts. All of the crys- talline rocks are considered to be clasts dislodged from frag— mental rocks. These fragmental rocks constitute the Fra Mauro Formation, which is presumed to be the ejecta deposit from the Imbrium basin. Megascopic properties of the frag- mental rocks suggest a fourfold subdivision: F1, poorly con- solidated breccias with light clasts dominant over dark ones, with few clasts larger than 1 mm, and with comparatively abundant fragmental glass; F2, coherent breccias with light clasts dominant and with comparatively abundant fragmental glass; F3, moderately coherent breccias with dark clasts dom- inant and with no fragmental glass; and F4, coherent breccias with dark clasts dominant and with no fragmental glass. Systematic microscopic descriptions and clast counts of 19 different breccias revealed that (1) clasts of fine-grained thermally metamorphosed fragmental rock are by far the dominant type in the breccias; (2) there are eight main varieties of igneous clasts including basalt, anorthosite-gab- bro, plagioclase cumulates, and quartz-alkali feldspar inter- growths; (3) a significantly high proportion of mineral clasts occurs in the size range 0.1—1.0 mm; plagioclase is dominant among the mineral clasts, but the proportion of pyroxene is as much as 50 percent; and (4) lithic and mineral clasts commonly are cataclastically deformed. Most of the rocks contain compound clasts that are them- selves pieces of breccia. Two generations of breccia are com- mon, and as many as four occur in the samples studied. The rock types in compound clasts reveal earlier fragmentation- metamorphic episodes that predate the Fra Mauro Formation. Partial to complete reequilibration reactions have taken place between matrices and clasts in a number of the frag- mental rocks. These relations suggest postconsolidation thermal effects on the Fra Mauro Formation. Similar reactions are observed within clasts in all types of frag- mental rocks, whether glassy or crystalline. It appears that more than one generation of reequilibration reactions is represented in the Fra Mauro and that older breccias had thermal histories similar to that of the Fra Mauro Formation itself. The distribution of rocks at the Apollo 14 site suggests that the Fra Mauro Formation is stratified; units of dark- clast-dominant breccias (F3, F4) are deepest, and units of light-clast-dominant breccias (F2) are shallowest. Poorly consolidated glass-rich rocks (F1) are considered to be and E. D. JACKSON weakly lithified regolith, derived either from disaggregated F1 type breccias or from originally unconsolidated material at the top of the Fra Mauro Formation. This stratigraphic sequence is consistent with the comparative abundance of glass and the weakly developed reequilibration reactions in F2 rocks and with the dominance of light—colored metaclastic rocks among the lithic fragments of the soils. It is presumed that thermal blanketing of the lower dark-clast-dominant breccias allowed time for reactions to proceed and for anneal- ing and devitrification of glass. Thus, although we see that the Fra Mauro Formation is composed of three or four distinct units based on clast populations, its internal thermal metamorphism appears to be gradational. The compound clasts and great variety of lithic frag- ments indicate that the pre-Imbrium terrane, presumed to have existed prior to about 3.9—4.2 billion years ago, was already exceedingly complex. However, the similarity of older breccias to the Fra Mauro breccias themselves in- dicates that processes such as extrusion of basalt, intrusion and differentiation of basaltic magma, impact brecciation, and metamorphism prior to the Imbrium event were not un- like later events. The complex clasts of the Fra Mauro breccias therefore reveal a long history of impact-related events that probably record half a billion years of planetary accretion, of which the Imbrium impact was one of the last large events. INTRODUCTION Apollo 14 landed on an area of lunar terra composed of materials that premission studies indicated were ejecta from the Imbrium basin (Gilbert, 1893; Swarm and others, 1971; Swann and Field Geology Team, 1971). This deposit, forming a broad belt surrounding Mare Imbrium, has been designated the Fra Mauro Formation (Eggleton, 1963; Wilhelms, 1970). The landing site (fig. 1) is about 550 km (kilometers) south of Montes Carpathus, the southern boundary of the Imbrium basin, and 150 km north of the mapped edge of the ejecta deposit. The formation in this area is estimated to be about 500 m (meters) thick (Eggle- ton, 1963). The principal objective of Apollo 14 was to sample the Fra Mauro Formation in the area im- mediately west of Cone crater, a young blocky crater superposed on a ridge of the formation (fig. 2). Cone crater was presumed to have excavated Fra Mauro material from beneath the local regolith, making about 1 2 PETROLOGY AND STRATIGRAPHY, FRA MAURO FORMATION, APOLLO 14 SITE the upper 75 m of the formation accessible to sam- pling, description, and photography. Results of the preliminary geologic investigations of the site (Swann and Field Geology Team, 1971) and of the preliminary examination of samples (Lunar Sample Preliminary Examination Team, 1971) have been published. This paper (1) presents a general report on the petrology and petrography of the Apollo 14 samples, based on our examination of the textures and structures of the rocks in hand specimen and thin section, (2) classifies these rocks, and insofar as is possible at this time, (3) discusses the stratigraphy and origin of the Fra Mauro Formation. Acknowledgments.—This work was done under NASA contract T—4738A. We are much indebted to J. P. Lockwood and D. J. Milton, US. Geological Sur- vey, for critical review of the manuscript. RETURNED SAMPLES Apollo 14 returned about 43 kg (kilograms) of rock samples, of which 321 specimens weigh more than 50 g (grams) and 1241 specimens weigh more than 1 g (Lunar Sample Preliminary Examination Team, 1971). These samples were collected by Astronauts Mitchell and Shepard during two traverses covering a distance of about 3,175 m (Swann and Field Geol- ogy Team, 1971). The location of samples along the traverses is known for the majority of rocks; two rocks bigger than 50 g (14303, 14169) are still only tentatively located (Sutton and others, 1971). In contrast to rocks returned by Apollo 11 and 12, the Apollo 14 rocks are predominantly fragmental and consist of clasts (>1 mm) (millimeter) and micro- clasts (0.1—1.0 mm) in fine-grained matrices. Indeed, only two of the 32 rocks weighing more than 50 g and a correspondingly small proportion of the 124 rocks weighing more than 1 g are homogeneous crystalline rocks (Lunar Sample Preliminary Examination Team, 1971). The homogeneous crystalline rocks all have their counterparts as clasts in fragmental rocks (Lu- nar Sample Preliminary Examination Team, 1971). One sample (14053) was collected from the side of a boulder, and surface descriptions by the Apollo 14 crew indicate that it was a clast in a fragmental rock (Sutton and others, 1971). The other (14310) is con- siderably smaller than the largest clast observed in surface photographs and is lithologically identical to clasts in returned fragmental rocks. It appears likely, therefore, that all the homogeneous crystalline rocks are clasts that were dislodged from fragmental rocks, either on the lunar surface by natural processes or by handling after collection. lTotals combine samples that were broken by handling but that were given different Lunar Receiving Laboratory numbers (Swann and Field Geology Team, 1971). ao°w 20° I0°w zoo“ 20°N EXPLANATION E Alpes Formation llllllllllllllll Fra Mauro Formation Materials of mountains rim» ming the Imbrium Basin Materials of major post» Fra Mauro craters B Dark mantling material E Young mare basalt U Old mare basalt Undivided terra materials Mostly pre»l~‘ra Mauro HIE/(’- rlals. but includer snmc [ult’f‘ crater dcposim and prohah/t' volcanic rocks Pre-mare plains-forming material FIGURE 1.—Regional geologic map of the area surrounding the Apollo landing site. From Swann and Field Geology Team (1971). V Boulder ray deposits Continuous ejecta blanket Material of Cone Crater Is Smooth terrain material of the Fra Mauro Formation Ifr Ridge material of the Fra Mauro Formation 100 O RETURNED SAMPLES llllll I EXPLANATION Contact Long dashed where approximately lo cated; short dashed where location is inferred without local evidence —l—__T——_ Foot of scarp Bounds small mesa. Short dashed where location inferred. Triangles point down slope T_K—_E__ Edge of mu Long dashed where approximately lo- cated; short dashed where location in- ferred. Triangles point down slope 500 METERS l l I __)__—. Traverse route for EVA’s l and 2 Stations OB —Panorama station A61 -Station without panorama C/S —ALSEP Central Station LR3 -Laser Ranging Retroreflector Samples FSR —Football size rock 033:“ —Contingency sample ogglmp-Comprehensive sample 09 —Grab sample at Station D FIGURE 2. —— Map of major geologic features in the Apollo 14 traverse area. From Swarm, Trask, Hait, and Sutton (1971). 4 PETROLOGY AND STRATIGRAPHY, FRA MAURO FORMATION, APOLLO 14 SITE MEGASCOPIC STRUCTURES Swann and Field Geology Team (1971) and Hait (1972) provided descriptions and detailed maps of the internal structures in some large boulders photo- graphed by the astronauts. The main types of megascopic structures, besides the obvious fragmental textures, are fractures, color layering, lineation or foliation, and veins. The fractures occur in multiple sets of generally planar, parallel joints spaced from a few millimeters to a few centimeters apart. Color lay- ering is particularly well developed in boulders from the area near the “white rocks” (fig. 3A), where prominent white-gray layers about 1 m thick occur. In places, soft layers are etched into depressions, pre- sumably by micrometeorite scrubbing, while adjacent harder layers stand out as ribs. Some layering in essen- tially monotoned boulders is defined by differences in sizes of clasts from layer to layer; size layering also accompanies the color layering. Some layers have highly irregular contacts, whereas others are nearly plane parallel. There appear to be sizable discordances in attitude between sets of layers Within two boulders (Hait, 1972), and in places, layers are contorted. Rarely, a weak lineation or, more probably, foliation is produced by alinement of lenticular clasts, and one irregular white layer about 1 cm (centimeter) thick in Weird Rock (Swann and Field Geology Team, 1971) may be a vein. Features of this kind are, of course, less obvious in the smaller returned hand samples. Many rocks, in- cluding 14304, 14302—14305, 14312, and 14313 have shapes partly controlled by fracture sets, and a num- ber, including 14306 and 14316, are cut by glassy veins. Sample 14082, collected from one of the “white rocks,” is layered (fig. 38, C) with millimeter— to centimeter—thick irregular units differentiated by the abundance of darker clasts and by average size of debris in the matrix (fig. 30). Samples 14076 and 14453 are also layered fragmental rocks. Sample 14076 has an irregular but sharp contact separating rock with about 30 percent clasts from rock with 5 percent of the same types of clasts. Sample 14453 consists of light- and dark-colored laminae of fragmental material interlayered on a scale of 1—3 mm. HAND-SPECIMEN PETROLOGY One of~us (Jackson) was a member of the Apollo 14 Lunar Sample Preliminary Examination Team (LSPET) ; the other (Wilshire) was a representative of the Apollo 14 Lunar Geology Investigation Group to the LSPET. Together, we had the opportunity to examine 107 Apollo 14 hand samples larger than 1 g, mostly with the aid of binocular microscopes. Many of our descriptions were systematized by using a form that required notation of a wide variety of features FIGURE 3.—Surface photograph and photomicrographs of Apollo 14 rocks. A, NASA photograph ASl4—68—9448 show- ing complexly interlayered light and dark rocks in block near rim of Cone crater. Both rock types appear to be fragmental. Bar scale 2 m long. B, Photomicrograph of lay- ering in sample 14082,11. Plane polarized light View show- ing planar contact between a layer lacking dark clasts and one containing them. Bar scale 1 mm long. C, Same area as 38, crossed nicols. Layer lacking dark clasts has a coarser grained matrix. Bar scale 1 mm long. for each specimen. Structures, textures, and lithologic properties, such as degree of induration, grain size, and mineral proportions, were systematically described and measured. This information is presented here be- cause wide access to the hand specimens is no longer possible and because the broad petrologic framework required for classification and stratigraphic studies could not otherwise be established. The Apollo 14 rocks may be divided into two broad groups (fig. 4): relatively homogeneous crystalline rocks and fragmental rocks (or breccias). Both groups HAND-SPECIMEN PETROLOGY Fragmental rocks Homogeneous l I O I crystalline rocks g : Basaltic (B) Metac|astle (C) 8 | | I L ______________ Light Dark Clast color FIGURE 4. —- Scheme used for subdivision of Apollo 14 rocks. have been subdivided on the basis of megascopic fea- tures described below. In table 1, all specimens weigh- ing more than 1 g have been classified into one of six categories — two groups (B, C) representing sub- division of the crystalline rocks, and four (Fl—F4) representing subdivisions of the fragmental rocks. HOMOGENEOUS CRYSTALLINE ROCKS Group B (basaltic) rocks are even grained and aphyric, with textures ranging from intersertal to ophitic. Grain size varies from fine to medium, but most are conspicuously coarser than the metaclastic rocks. Two groups are recognized, one having about 40—50 percent plagioclase (14053, 14071, 14074) and the other having about 60—70 percent plagioclase (14073, 14078, 14079, 14276, 14310). Rock 14310 has small cognate inclusions that are mainly finer grained than the body of the rock but are modally similar to it (Melson and others, 1972). Clasts of basaltic rocks are widespread though not abundant in the fragmental rocks. Most of these rocks are richer in plagioclase, poorer in ilmenite, and have much lighter colored pyroxenes than the basaltic rocks re- turned by Apollo 11 and 12. Group C (metaclastic) rocks have granular textures and are fine grained to aphanitic. The rocks are typi- cally inequigranular, with 1—5 percent of irregular large mineral grains, usually plagioclase, but less com- monly olivine or pyroxene. Both light and dark rocks of this type are very common as clasts in the frag- mental rocks. All but two rocks in this group are light colored and consist principally of plagioclase and light-gray pyroxene. The two dark rocks (14006, 14440) are aphanitic and of unknown mineral com- position. These rocks are similar to the common brec— cia clasts. FRAGMENTAL ROCKS The fragmental rocks, like the crystalline rocks, are distinctly different from those returned by the Apollo 11 and 12 missions. Most are coherent, contain less glass, and have larger and more abundant clasts than the fragmental rocks of earlier missions, but a few are more friable and contain fewer and smaller clasts TABLE 1. — Classification of rocks larger than 1 gram1 (B) Homogeneous crystalline rocks, basaltic Sample No. 14053 14073 14078 14276 14071 14074 14079 14310 14072 (C) Homogeneous crystalline rocks, metaclastic Sample No. 14006 14077 14434 14444 14068 14274 14436 14451 14069 14429 14440 14070 14431 14443 (F1) Fragmental rocks, light clasts dominant, matrix friable Percent Percent Percent Percent Sample clasts Sample clasts Sample clasts Sample clasts No. >1 mm No. >1 mm No. >1 mm No. >1 mm 14041" 2 14056 2 14177 1 14427 1 14042“ 10 14057 1 14178 10 14432 2 14043“ 'I 14058 20 14185 'I 14437 1 14045‘ 1 14059 1 14250 10 14438 1 14047 3 14061 7 14251 1 14442 1 14049 1 14062 1 14282 1 14449 10 14055 5 14080 1 14286 5 14452 1 (F2) Fragmental rocks, light clasts dominant, matrix coherent to moderately coherent Percent Percent Percent Percent Sample clasts Sample clasts Sample clasts Sample clasts No. >1 mm No, >1 mm No. >1 mm No. >1 mm 14007 25 14269 10 14281 40 14315 20 14051 '3 1 14271 1 14285 10 14316 20 14195 1 14272 1 14288 25 14317 10 14201 10 14273 1 14294 10 14318 30 14255 2 14275 10 14297 10 14430 20 14265 10 14277 10 14301 18 14453 1 14267 10 14278 20 14307 20 14268 10 14280 40 14313 5 (Fa) Fragmental rocks, dark c lasts dominant, matrix friable Percent Percent Percent Percent Sample clasts Sample clasts Sample . clasts Sample clasts No. >1 mm No. >1 mm No. >1 mm No. >1 mm 14063 23 14082“ 10 14171 30 14426 5 14064 40 14083“ 15 14196 10 (F4) Fragmental rocks, dark clasts dominant, matrix coherent to moderately coherent Percent Percent Percent Percent Sample clasts Sample clasts Sample clasts Sample clasts No. >1 mm No. >1 mm No. >1 mm No. >1 mm 14066 25 14179 20 14266 25 14308‘ 20 14075 5 14180 1 14270 '! 14309 20 14076 5 14181 50 14279 1 14311“ 20 14169 25 14182 5 14296 15 14312 45 14170 5 14188 5 14302“ 28 14314 25 14172 6 14194 8 14303 28 14319 35 14173 20 14197 20 14304 25 14320 13 14174 10 14199 20 14305“ 28 14321 30 14175 1 14264 10 14306 15 14445 15 14176 1 Samples lacking sufficient information for certain classification Sample Tentative No. identification 14008 F1 14060 F1 14198 B "Identifies broken pieces of same sample: 14041, 14042, 14043, 14045 parts of same sample; 14302, 14305 parts of same rock; 14308, 14311 parts of same rock; 14082, 14083 parts of same rock. 1For location see Swann and Field Geology Team (1971) ; for weight see Apollo 14 lunar sample information catalog, NASA TMX—58062. than most Apollo 11 and 12 breccias. We have dis- tinguished four groups of fragmental rocks on the basis of proportions of light and dark lithic fragments and matrix coherence. All four groups of fragmental rocks are polymict and contain clasts ranging in shape from angular to subrounded. Group F, rocks are defined as fragmental rocks con- taining leucocratic lithic clasts in excess of mesocratic 6 PETROLOGY AND STRATIGRAPHY, FRA MAURO FORMATION, APOLLO 14 SITE and melanocratic lithic clasts and having friable ma- trices. There are 25 rocks in this category ranging in weight from 1.5 to 242 g (table 1). In more than half of these rocks, leucocratic clasts make up more than 90 percent of total number of clasts larger than 1 mm.2 The abundance of clasts in these rocks is low (fig. 5). The majority of the clasts are lithic crystal- line very fine grained feldspar-rich rocks; many ap- pear to have originally been fragmental. Angular glass fragments, broken spheres, and spheres, mostly me- dium brown, are common. Mineral fragments larger than 1 mm are rare. The matrices of these rocks are medium gray and, down to the limit of resolution, consist of the same materials as the clasts, but With much larger proportions of glass and mineral frag- ments. Many of these rocks may be weakly indurated regolith. Group F2 rocks are defined as fragmental rocks con- taining leucocratic clasts in excess of the total of mesocratic and melanocratic clasts and having mod- erately coherent to coherent matrices. There are 30 rocks in this category ranging in weight from 1.5 to 1,360 g (table 1). In about 60 percent of these, leuco- cratic clasts make up more than 90 percent of total number of lithic clasts larger than 1 mm. The abun- dance of clasts is much greater than in F] rocks (fig. 5) ; nearly 80 percent of F2 rocks contain 5 per- cent or more clasts greater than 1 mm. The majority of the clasts are lithic, and most of them are mega— scopically indistinguishable from those in group F]. Clasts of basaltic rock very similar to those of group B are present. Very fine grained melanocratic clasts that characterize groups F3 and F4 fragmental rocks, while never dominant, are present in many F3 rocks. A few small clasts of finely granular nearly pure olivine and pyroxene aggregates are present in some F2 rocks. Mineral fragments larger than 1 mm are rare, and glass clasts are distinctly less abundant than in group F1 rocks. Matrix colors range from light to dark gray. Mineral fragments are more abundant than lithic fragments in the size fraction smaller than 1 mm. Group F3 and F4 rocks are defined as fragmental rocks containing melanocratic and mesocratic frag- ments in excess of leucocratic clasts. The groups are further subdivided on the basis of matrix coherence: group F3 rocks have relatively friable matrices, and F4 rocks are moderately coherent to coherent. So de- fined, group F3 contains only six rocks, all relatively small. Two of these rocks (14082, 14083), the “White rocks,” appear to be unique, but 14063 and 14064 also QMost of the breccias have seriatc fragment sizes, We arbitrarily subdivide these as “clasts,” >1 mm, and “micruclasts,” 0.1—1.0 mm. “Matrix” refers to glass and fragmental material <1 mm. zollllllllll F1 F3 NUMBER FREQUENCY 0 —T 5 25 45 5 25 45 PERCENT CLASTS FIGURE 5. —— Frequency distribution of clasts >1 mm in frag- mental rocks of groups F1, F2, F3, and F4. have light-colored matrices. The “White rocks” owe their light color to the matrix material, but clasts larger than 1 mm are dominantly melanocratic. The F, category contains 35 rocks, the largest weighing nearly 9 kg (table 1). Clasts larger than 1 mm are abundant (fig. 5). The majority of the clasts are lithic crystalline aphanitic dark-gray to black rocks, many with metamorphic textures. Leucocratic clasts appar— ently like those of group F; are present but not domi- nant. Basaltic rocks similar to those of group B are present in some rocks, and clasts of granular olivine and pyroxene aggregates seem to be more abundant than in group F2. The matrices vary in color from light to medium gray. Almost all the samples can be classified according to the scheme described above. Among several speci- mens, however, that have noteworthy macroscopic features in addition to the characteristics used in the classification are glass-cemented breccias (14080, 14251, 14307), a single fragment largely composed of metal (14286), and a single small glass sphere (14425). THIN-SECTION PETROLOGY Thin sections of only 30 of the samples examined megascopically are available. We studied 200 thin sec- THIN-SECTION PETROLOGY 7 tions of these 30 rocks. Clasts that could be identified as such in individual thin sections of fragmental rocks were divided into glass, mineral, and igneous and metamorphic lithic types. These groups were further subdivided into a total of 22 recognizable clast types and several subtypes based on mineralogy, color, and texture. (See table 2 for list of types.) The generally small size of the thin sections often did not allow good statistical results to be obtained for clasts larger than 1 mm. Accordingly, counts were also made of frag— mental material in the size range 0.1—1.0 mm, which we refer to here as “microclasts.” Systematic micro- scopic descriptions and counts of clast types of 19 different fragmental rocks were made from 89 thin sections. Counts of 100 clasts and microclasts were generally made, although for some samples as many as 215 were systematically described and for others the lack of thin sections resulted in counts of as few as 15. A total of more than 3,000 clasts and their individual dimensions were thus identified. The results of these clast counts are given below. TABLE 2. —- Principal types of clasts and microclasts Identifying Clast-microclast symbols 7311176 A1 ................................................. Pale-yellow to brown glass. A: ................................................. Colorless glass. B1 .................................................. Plagioclase. B3 ..... . ............................................ Clinopyroxene. B3 .................................................. Olivine. B4 .................................................. Orthopyroxene. B7, .................................................. Opaque minerals. C1 .................................................. Intersertal basalt. C2 .................................................. Ophitic basalt. C3 .................................................. Intergranular basalt. C4 .................................................. Variolitic basalt. C7,..._ ................................... Vitrophyric basalt. Cc, .................................................. Graphic quartz-alkali feldspar. 07 __________________________________________________ Plagioclase, orthopyroxene cumulates. C3 .................................................. Hypautomorphic gabbro, norite. D1 .................................................. Recrystallized plagioclase. D2 .................................................. Light-colored metaclastic rock. ................. Metabasalt. ...... Dark-colored metaclastic rock. ................ Recrystallized olivine. ................. Recrystallized pyroxenes. D7 .................................................. Metagabbro. The problem of differentiating the matrices of the fragmental rock proved more difficult. In general, the finer component down to the limits of microscopic resolution consists of the same constituents as the clasts, progressively disaggregated as the grain size decreases. Much material, however, cannot be resolved optically. In some rocks, postconsolidation metamor- phism has further affected the compositions and tex— tures of the matrices (Wilshire and Jackson, 1972; Warner, 1972). GLASS CLASTS Two varieties of glass clasts (class A, table 2) were distinguished optically: (1) pale-yellow to dark-brown glass that usually contains mineral and lithic debris and that commonly contains newly formed crystallites (type A1) ; and colorless glass lacking both inclu- sions of rock debris and crystallites (A2). Glass clasts (>1 mm) are proportionally less abundant than microclasts (0.1—1.0 mm) in the total clast population (fig. 6). Most glass clasts and microclasts are irregu- lar, angular fragments that were clearly solid when incorporated in the fragmental rocks. Some, however, have shapes suggesting that they were soft and were deformed during their incorporation. Spheres and pieces of spheres of both yellow-brown and colorless glass occur sparsely. The colorless glass may largely be shocked plagioclase (maskelynite) ; the more com- mon colored glasses probably have the wide range of basaltic compositions reported for glasses in Apollo 14 soils (Jakes and Reid, 1971; McKay and others, 1972; Chao and others, 1972; Carr and Meyer, 1972). MINERAL CLASTS Mineral fragments (class B, table 2) make up a sizable proportion of microclasts in all the fragmental rocks but, as might be expected, are proportionally more abundant in the 0.1- to 1.0—mm-size fraction than in the >10 mm fraction (fig. 6). A small percentage, however, attain sizes as large as 2.5 mm. Plagioclase (B1) is clearly the dominant mineral clast (fig. 6), but proportions of pyroxene are as much as 50 percent (fig. 7). A few plagioclase grains appear to have been compositionally zoned along irregular boundaries and subsequently broken across the zones. Pyroxene clasts include pale-brown to colorless augite (B2), nearly colorless pyroxene with low birefringence and inclined extinction (presumed to be pigeonite, but not differen- tiated from augite in the counts), and orthopyroxene (B4). Orthopyroxene clasts, especially the large ones, commonly contain exsolution lamellae of clinopyroxene (Papike and Renee, 1971); the clinopyroxenes are commonly, but not always, slightly more abundant than the orthopyroxene (fig. 6). Olivine (B3) is pres- ent sparsely as mineral clasts as large as 2.3 mm. Optical properties and preliminary electron micro- probe data indicate that it is typically very magnesian. Minor constituents of the mineral microclast frac- tion (B5) include ilmenite, metallic iron, troilite, chro- mian spinel, ulvospinel, native copper, armalcolite, zircon, apatite, and potassium feldspar( ‘2) (Lunar Sample Preliminary Examination Team, 1971). In addition, we have observed quartz and a dark-green PETROLOGY AND STRATIGRAPHY, FRA MAURO FORMATION, APOLLO 14 SITE Fl ‘; 34047 3143011 3 3.1-1 3 >1 H714“ fl 1.11.113 11.11 I 314049? 514307: 3.1—1 E— 12112345112 345678512345678 ABC D M43073 3 >1 ‘ lflfimunnnr 111111111111111 5143133 5 .1-1 3 F2 F3.4 3140423 4301 3 § 14313 3 3140823111) 314303§ 043063 1-1 1 1-1 1 §>1 ‘ 3.1-1 1 3.1-1 1 3>1 ‘ Um?“ flunr 314082§(w) 3 .1-1 3 ‘11; n1 LnH ‘ n 540633 5 .1-1 3 1 311 ”.17 314318 i 3 .1 -1 1 3140633 1 >1 ‘ 1311; 11 ABC [— IZ17245311315678312315678 D $143183 1>1 111111111; 3140663 3 .1-1 : L111 AB C '1-1 1 Fin-1 l2£l23451l2315673512315678 D 040665 1>121 EI4171§ 5 >1 12(1214502 315670” 231567! AB C D 314308? 3 .1 -1 i §14306§ 3 .1-1 1 1311 1143085 i14314§ F .1-1 1 3143142 1 >1 5 _ AB C 12112345311345673212345333 D 1 1 H 11 1. '75123‘551 234567551 2315678 ABCD FIGURE 6.—Histograms showing relative abundance of glass (A), mineral (B), igneous (C), and metamorphic (D) clasts (>1 mm) and microclasts (0.1—1.0 mm) in the fragmental rocks of groups F1, F2, F3, and F4. THIN-SECTION PETROLOGY 9 32+4 Clinopyroxene + Orthopyroxene 31‘33'52+4 .l-lmm 0315 V V 3] Plagioclase Olivine B3 FIGURE 7. —— Triangular plot of plagioclase (B1) -pyroxene— (B2+B4) -olivine (B3) microclasts. amphibole(?) (thin section 14308,2) in some thin sections. LITHIC CLASTS IGNEOUS ROCKS Nine varieties of igneous rocks occur as clasts and microclasts (table 2) : five textural variants of largely crystalline basalt, vitrophyric basalt (five subtypes), graphic intergrowths of quartz and alkali feldspar, cumulates (seven subtypes), and hypautomorphic- granular rocks (two subtypes). Despite the consider- able variety of lithologies, clasts of igneous rock are not particularly abundant among the lithic clasts as a whole (fig. 6). The most abundant igneous rock type is aphyric intersertal basalt (CI). Clast sizes in thin section are as large as 6 mm (table 3), but rock 14310, inferred to be a clast, is of this type and measures about 18 cm TABLE 3. — Clast sizes and grain sizes of some igneous clasts Maximum clast size Average Group in thin section grain size (table 2) (mm) (mm) C1 6 0.03—0.08 C2 5 .1 — .2 C3 4 .05 C4 4 05 4 .1 — .2 Ce 2 .5 — .6 C7 2 .05— .13 C8 7.5 .2 —3.0 (centimeters) across. The majority of these rocks are somewhat finer grained than 14310, but 14310 is tex- turally inhomogeneous (Ridley and others, 1971; Mel- son and others, 1972) on a scale comparable to or larger than the average clast size of the basalts (fig. 8A). Rocks with intersertal texture are rich in plagio- clase (table 4), which typically forms slender euhedral crystals interspersed with subhedral clinopyroxene. Opaque minerals (ilmenite, troilite, metallic iron) are present in small amounts (table 4). The groundmass, ranging from about 1 to 40 percent, is composed partly of very fine grained apatite ( ?), ilmenite, troilite, alkali feldspar(?), glass, and unidentified minerals (Lunar Sample Preliminary Examination Team, 1971). It is possible that sample 14310 and similar clasts are products of near—surface rocks that were shock melted (Dence and others, 1972). TABLE 4.—Modal compositions of basalts, in volume percent [Tr., trace] Intersertal Ophitic Intergranular Plagioclase .60 55 60 58 70 55 40 50 50 80 50 30 19 39 58 45 45 10 50 ................ 1 5 1 5 5 2 . . 1 2Tr. 1 1 ........ 8.... Mesostasis ............. 10 10 39 10 10 ........................ Aphyric basalt clasts and microclasts with ophitic texture (Cg) are almost as abundant as intersertal basalts, and the two types may intergrade. In rock 14310, intersertal texture tends to grade toward ophit- ic or subophitic texture as the grain size increases. This group of clasts differs from the ophitic parts of 14310 mainly in having a small percentage of olivine. Clast sizes in thin section are as large as 5 mm, but the largest Cg clast returned is sample 14053. Average grain sizes are somewhat coarser than those of 01 rocks (table 3), as large as 0.6 mm in rock 14053 ( Melson and others, 1972). These rocks typically have lower plagioclase contents (40—50 percent) than inter- sertal basalts (table 3), probably in part due to non- representative sampling of small clasts. Plagioclase forms euhedral grains that penetrate larger anhedral to subhedral pyroxenes (fig. SB). Olivine occurs sparsely, usually as ragged grains in cores of pyrox- enes. Opaque minerals (metallic iron, ilmenite, troilite) are not abundant (table 3). Tridymite and cristobalite occur in interstices and vugs of some rocks, and some rocks contain 1 percent or less of interstitial material (including metallic iron, dendritic ilmenite, fayalite( ?), and glass (Lunar Sample Preliminary Examination Team, 1971). 10 PETROLOGY AND STRATIGRAPHY, FRA MAURO FORMATION, APOLLO 14 SITE THIN-SECTION PETROLOGY 11 Clasts of basaltic rock with intergranular (Cs) and variolitic (Cl) texture are moderately widespread (fig. 6) but are not abundant. Modes are given in table 4, size data in table 3. Clasts of basaltic rock with vitrophyric texture (Cs) have been found in thin sections of seven rocks (fig. 6). These clasts are as large as 4 mm (table 3) and either are irregular angular to subangular fragments or have nearly perfectly circular cross sections (fig. 8C). The circular ones are among the features called chondrulelike by King, Butler, and Carman (1971, 1972) and by Fredriksson, Nelen, Noonan, and Kraut (1972). Five subtypes, all containing very dark brown glass, were identified on the basis of their mineralogy (table 5). TABLE 5. -—— Types of vitrophyres Group Crystal phases C3. ..................................... Plagioclase only. 0.11.... _____ Olivine only. C7,..." ._..P1agioclase, olivine. Cad... _____ Plagioclase, clinopyroxene. -------------- Plagioclase, olivine, cllnopyroxene. 5e ---------------- Clasts of graphic to irregular intergrowths of alkali feldspar and quartz (C6) are widespread but not abundant. Irregular quartz grains are intergrown with larger (table 3) alkali feldspar grains (fig. 8D). In places irregular blades of quartz that are comparable in one dimension to the alkali feldspar are intergrown with alkali feldspar. FIGURE 8.—Photomicrographs of Apollo 14 rocks. A, Photo- micrograph of fine-grained cognate inclusion (upper left 1/3 of photograph) in 14310,10. Boundary with normal rock is gradational, and both parts are composed of plagioclase laths, pyroxene, a trace of opaque minerals, and fine-grained interstitial material. B, Photomicrograph of ophitic basalt, 140536 Plagioclase laths penetrate zoned pigeonitic pyroxenes. Opaque minerals and olivine are scarce. C, Photomicrograph of a round clast of vitrophyre composed of skeletal olivine prisms and a small amount of plagioclase in finely devitrified glass. 14318,8. D, Photo— micrograph of a clast that consists of a graphic intergrowth of quartz and alkali feldspar. 14318,8. E, Photomicrograph of a cumulus clast composed of euhedral cumulus plagioclase and orthopyroxene (higher relief than the plagioclase) in a groundmass of finer grained postcumulus plagioclase, pyroxene, opaque minerals, and glass. 14321,29. F, Photomicrograph of a cumulus clast composed of euhedral cumulus plagioclase with postcumulus clino- pyroxene. 14171,13. G, Photomicrograph of a dark aphani- tic (D4) metaclastic clast composed of porphyroclastic plagioclase and a finely recrystallized plagioclase clast (D1) near center, in a dark microcrystalline matrix. 1406328. H, Photomicrograph of a light-colored (D2) metaclastic rock with plagioclase porphyroclasts in a moderately coarse grained granoblastic matrix composed of plagioclase, orthopyroxene, and opaque minerals. 14042,7. Bar scale 0.5 mm long. Rocks with cumulus textures (CT) are among the most widespread igneous clasts (fig. 6), though they are not as abundant as either intersertal or ophitic basalts. They are typically subangular to subrounded and are as much as 2.0 mm in size (table 3). Seven varieties of cumulates are identified on the basis of types of cumulus and postcumulus phases (table 6). TABLE 6. —.l10des and average grain sizes of common cumu- lates Cumulus minerals Postcumulus Average cumulus (volume percent) phases grain size (mm) Plagioclase (45) Orthopyroxene (55) 0.05 Opaque (trace) Clinopyroxene and Plagioclase (58) 0.05—0.07 olivine (42) Opaque (trace) Plagioclase (37) Plagioclase (27) 0.05 Orthopyroxene (13) Pyroxene (20) Glass (3) Plagioclase (40) Clinopyroxene (55) 0.07 Opaque (5) Plagioclase (79) Clinopyroxene (15) 0.13 + postcumulus Glass (6) overgrowth(?) Opaque (trace) Clinopyroxene (42) Plagioclase (58) 0.06 Opaque (trace) Plagioclase (40) Plagioclase ......... (with tiny Orthopyroxene. olivine(?) Alkali feldspar. (60) 0'13 inclusions) Glass .................... The textures of these rocks vary considerably (figs. 8E, F), depending in part on the degree of post- cumulus modifications in the shape of cumulus grains. Plagioclase is the dominant cumulus phase in all but one variety and is the only cumulus phase in three (table 5). Other cumulus minerals are clinopyroxene, orthopyroxene, and olivine. Postcumulus material in- cludes plagioclase, glass, opaque minerals, clinopyrox- ene, orthopyroxene, and alkali feldspar. Clasts composed of igneous rock with hypautomor- phic granular textures (Cg) are moderately wide- spread (fig. 6). Most are subangular clasts as much as 7.5 mm in diameter (table 3). One is norite com- posed of about 45 percent plagioclase, 50 percent orthopyroxene, 4 percent clinopyroxene, and 1 per- cent opaque minerals. The other coarser clasts (table 3) are gabbros and anorthosites. Their coarse grain size suggests that small clasts will not yield represen- tative modes. METAMORPHIC ROCKS Seven types of metamorphic rocks (class D, table 2) occur as clasts and microclasts in the Apollo 14 frag- mental rocks: two varieties of metaclastic rocks (one dark and one light colored), metabasalts, metagabbros, and three main types of recrystallized mineral grains. 12 PETROLOGY AND STRATIGRAPHY, FRA MAURO FORMATION, APOLLO 14 SITE All these metamorphic rocks have some variant of hornfels texture, with relicts of angular material that suggest a premetamorphic cataclastic texture. Metaclastic rocks are the most abundant of all clasts in the Apollo 14 fragmental rocks. They are character- ized by the presence of relict angular crystal debris (porphyroclasts), ranging from a trace to 75 percent of the rock or more, in a granoblastic or poikiloblastic matrix. Plagioclase dominates as a relict mineral, but clinopyroxene, orthopyroxene, olivine, and quartz occur. Of the two varieties, the dark-colored clasts (D4) are more abundant than the light—colored ones. D4 metaclastic rocks are principally recrystallized cataclasite and ultracataclasite (Spry, 1969), which are dark both in hand specimen and in thin section (table 2; fig. 8G). Clast sizes in thin section are as large as 8.0 mm. Though there appears to be a grada- tion in this class from extremely fine grained very dark rocks to somewhat coarser lighter gray rocks, some breccias (for example, 14306, 14063, 14312, 14171) have distinct clasts at both ends of the spectrum, with few intermediate individuals. In places, these contrast- ing types are interbanded, and in others, one type may occur as clasts in the other. Even the coarsest grained rocks of this type are too fine grained to enable identi- fication of most minerals, but they appear to have a higher opaque-mineral content than the light—colored metaclastic rocks, and the pyroxenes appear to be principally clinopyroxene. Some of these clasts show evidence of having been partly fused (buchite), and small amounts of glass have, in a few samples, sur- vived recrystallization. In others, the original presence of a melt is indicated by occurrence of vesicles and of small microlites of olivine or plagioclase along with the coarser porphyroclasts. A second common variety of recrystallized cata- clasites (D2) are light colored in hand specimen and in thin section (table 2). These rocks are generally coarser grained than the dark metaclastic rocks, and they are characterized by granoblastic or poikiloblastic textures. Granoblastic textures (fig. 8H) tend to be— come granoblastic-polygonal With increasing grain size. In rocks with poikiloblastic texture (fig. 9A, B), the host mineral (orthopyroxene or clinopyroxene) enclosing plagioclase granules ranges in size from >005 to 0.10 mm. Some of the more coarsely re- crystallized clasts also show development of porphy- roclastic plagioclase (fig. 9C) and growth of small lath-shaped plagioclase grains. Where such rocks lack porphyroclasts, they resemble plagioclase cumulates, except for their fine grain size. In general, however, the former fragmental nature of these rocks is clearly evidenced by their angular grains and very poor sorting. Plagioclase is clearly the dominant mineral, although its proportion is extremely variable. In the majority, orthopyroxene is the most abundant mafic mineral. Clinopyroxene occurs with the orthopyroxene in some rocks and is the only pyroxene in a few; the consistently low birefringence of the clinopyroxene suggests that it is mainly pigeonite. Opaque minerals are a constant accessory, and small amounts of por- phyroclastic olivine are common. Metabasalts with granoblastic textures (D3) are moderately widespread but are not abundant (fig. 5). Most of these appear to have had intergranular or intersertal textures before metamorphism and now consist of a granular aggregate of augite, with or without olivine, equant opaque minerals, and ragged plagioclase that commonly tend to poikiloblastically enclose other minerals. The rocks closely resemble hornfelsed basalts in the Apollo 11 samples, described by James and Jackson (1970), though many are coarser grained. Typical modes and grain sizes are given in table 4. The metagabbros (D7) are not only rare, but also unusual rocks. A few of them may have been troctolites (figs. 9D, E) in which What appear to be cumulus olivine grains are distributed among plagioclase grains. Both the olivine and the plagioclase have been recrys- tallized to fine-grained granoblastic aggregates, but the original texture of the rock is preserved. An even more striking example is illustrated in figure 9A and B, in which original euhedral shapes of plagioclase, possibly ophitically enclosed by pyroxene, are preserved despite coarse recrystallization to a decussate mosaic. Virtually monomineralic clasts of plagioclase or olivine hornfels are common, and similar ones of clinopyroxene and orthopyroxene occur sparsely. The clasts typically are ovoid and subround but may be blocky and angular. Their size ranges from 0.01 to 3.0 mm but averages about 0.25 mm. Grain sizes vary FIGURE 9. —Photomicrographs of Apollo 14 rocks. A, Photo- micrograph of a light—colored (D3) metaclastic rock with poikiloblastic orthopyroxene enclosing small plagioclase grains. 14082,12. 8, Same area as 9A. Crossed nicols. Note zoning of porphyroclastic plagioclase. C, Photomicrograph of a light-colored (D3) metaclastic rock with porphyro- clastic plagioclase in a coarse-grained matrix composed of orthopyroxene poikiloblastically enclosing plagioclase laths, crossed nicols. 14063,25. D, Photomicrograph of a “meta- troctolite” composed of olivine and plagioclase. Plane polar- ized light. 14801,13. E, Same as 9D, crossed nicols, showing granoblastic recrystallization of both olivine and plagio- clase. F, Photomicrograph of a “metagabbro” clast with euhedral plagioclase grains enclosed by clinopyroxene. Plane polarized light. 142709 G, Same view as 9F, crossed nicols, showing decussate recrystallization of the plagioclase and granoblastic texture of the pyroxene. H, Photomicro- graph of clast with areas of a single relict plagioclase grain, crossed nicols. 14312,19. Bar scale 0.5 mm long. 13 THIN—SECTION PETROLOGY ‘r w ., n a. ‘ . Emma. «A: . .1 N?L€FM«X. JV <’ 14.. 7.. 14 PETROLOGY AND STRATIGRAPHY, FRA MAURO FORMATION, APOLLO 14 SITE THIN—SECTION PETROLOGY 15 somewhat (range 0.03 to 0.08 mm), with typical grain sizes for the plagioclase clasts at the lower end of the range and for olivine clasts at the upper end. Plagio- clase hornfels (D1) are by far the most abundant of the monomineralic hornfels and show the widest variety of textures. Most plagioclase clasts consist of fine interlocking anhedra that are either elongate blebs or irregular in shape. Spherulitic texture in the plagio- clase clasts is common and in some is spectacularly developed (fig. 10A). A few plagioclase clasts have granoblastic-polygonal textures that closely resemble those in Apollo 11 rocks described by Wood, Dickey, Marvin, and Powell (1970a, b), and others have strik— ing decussate textures resembling the feldspar in metagabbro clasts. Olivine hornfels (D5) typically have granoblastic-polygonal textures (fig. 103) and are composed of interlocking grains of undeformed olivine. Pyroxene hornfels (D6) are uncommon, and those composed of orthopyroxene are especially rare. Both have granoblastic-polygonal textures. Relicts of the mineral grains that recrystallized to hornfels are present in a number of these clasts; no polycrystalline relicts of mafic silicates were observed, the unrecrystallized material in all samples being rel- icts of a large single grain (fig. 100). However, a few incompletely recrystallized plagioclase clasts have rel- icts that suggest a polycrystalline source, and we can- not be sure that we have not counted some previously polygranular plagioclase aggregates in this group. It is nevertheless clear that the great majority of these monomineralic aggregates are recrystallized single FIGURE 10. — Photomicrographs of Apollo 14 rocks. A, Photo- micrograph of a clast of spherulitically recrystallized feld- spar. 14301,16. B, Photomicrograph of olivine hornfels microclast with granoblastic-polygonal texture. 140665. C, Photomicrograph of olivine microclast, showing partial recrystallization of single kink-banded olivine grain. 140667 D, Photomicrograph of partly reacted olivine microclast surrounded by a corona of pyroxene and an opaque mineral. Note lightened halo in matrix around corona. Small “olivine” adjacent to right side of large one is completely replaced. 14308,3. E, Photomicrograph of a plagioclase porphyroclast in a poikiloblastically recrystallized D2 meta- clastic rock. Plagioclase porphyroclast has a narrow sodic reaction rim. 14063,8. F, Photomicrograph of red spinel microclasts with colorless feldspathic reaction rims. Largest spinel grain (lower center) has darkened borders next to the corona, while smaller grains are dark throughout. 14171,13. G, Photomicrograph of quartz microclast, crossed nicols, showing quartz core surrounded by an envelope of glass that in turn is surrounded by a corona of pyroxene. 14066,7. H, Photomicrograph of clast of partly fused graphic quartz-alkali feldspar intergrowth, crossed nicols. Fusion occurs along grain boundaries (wide black bands) in the interior of the clast. Quartz at the edge of the clast has reacted with the matrix to form a pyroxene corona. 14305,102. Bar scale 0.5 mm long. crystals. It is possible that they were formed from dis- aggregated coarse-grained metagabbros like those illustrated in figure 9 (D—G). Ave’Lallement and Carter (1972) attributed the conditions of recrystal- lization to shock processes. However, we have observed olivine with a similar texture in an unshocked olivine pyroxenite inclusion in basalt from Kilbourne’s Hole, N. Mex. Also, Wilson (1969) described recrystalliza- tion of pyroxene in granular aggregates in terres- trial granulites without modification of original grain shapes. It seems quite clear that these lunar materials are partly to completely recrystallized mineral grains and do not represent fragments of lunar anorthosite, pyroxenite, or dunite. MATRIX MATERIALS Examination of surface photographs of boulders of breccia ejected from Cone crater (Hait, 1972) has shown the presence of some clasts that are larger than many of the returned samples. Since many of the clasts in the breccias are themselves fragments of breccia, the relative age of the “matrix” of any par- ticular breccia sample is difficult to establish. The problem of distinguishing the youngest, or Fra Mauro, matrix is discussed in a later section; in the following discussion, “matrix materials” refers to the youngest binding agents of any particular sample. It is apparent in hand sample and thin section that the clasts in the Apollo 14 fragmental rocks have a seriate distribution and that our division by grain size into clasts (>1.0 mm) and microclasts (0.1—1.0 mm) is arbitrary. Aside from a major increase in propor- tion of mineral microclasts in all rocks and of glass microclasts in those rocks that contain glass, the types and relative abundances of microclasts do not differ significantly from those of clasts. A comparison of the kinds of lithic clasts and lithic microclasts in the same rock (fig. 6) show that the proportions of rock types are much the same. Indeed, the grain sizes of the most abundant lithic clasts are such that they are commonly not disaggregated in the 0.1- to 1.0-mm-size fraction. One cannot readily account, therefore, for the origin of the large percentage of mineral fragments in the 0.1— 1.0—mm size range. (See fig. 6.) A small proportion of these fragments were almost certainly derived by comminution of porphyroclasts in metaclastic rocks; for example, some plagioclase fragments are broken across sodic reaction rims that characterize these por- phyroclasts. However, in most rocks, the proportion of plagioclase fragments is not high enough to suggest a porphyroclast source. Moreover, the comparative abundance of mineral microclasts is not matched by the generally low proportions of porphyroclasts, and the typical size of porphyroclasts is not larger than the mineral fragments. A more likely source for the 16 PETROLOGY AND STRATIGRAPHY, FRA MAURO FORMATION, APOLLO 14 SITE mineral fragments is the coarser grained gabbros with hypautomorphic granular texture. The mineral pro- portions in the clasts of gabbro are consistent with the mineral fragment distributions, although the rela— tive paucity of gabbroic rocks among the lithic clasts is diflicult to explain. Rocks containing much clastic glass in the >0.1- mm-size fraction (F1, F2) also appear to be bound by glass that, if fragmental, is too small to resolve micro- scopically in thin sections.3 Rocks lacking glass clasts (F3, F4) are bound by very finely crystalline appar- ently annealed material that is generally finer grained than most clasts of metaclastic rock. In some of these rocks, spherical and broken spherical fragments that presumably were formerly glass are now composed of fine crystalline material. Warner (1971, 1972) has subdivided the Apollo 14 breccias, using the relative abundance of glass clasts and of glass in the matrices as his principal criteria. He thus divided the rocks into six categories (from glassy “detrital” breccias to glass—free “annealed” breccias), which span our Fl—F4 types. MATRIX-CLAST REACTIONS In a number of fragmental rocks, partial to com- plete reequilibration reactions have taken place be- tween youngest matrices and some clasts and have clearly occurred subsequent to the formation of the breccia. In other rocks, similar reactions have been observed within the clasts but not within the youngest matrix, which suggests that these reactions are pre- consolidation features. Postconsolidation mineral reactions—Mineral clasts and microclasts of olivine, plagioclase, spinel, and quartz commonly have reacted with the matrix con- taining them. Reaction products of olivine and matrix are orthopyroxene and ilmenite (fig. 10D). The ap- parent proportions of reaction products vary widely. The matrix appears in places to have been lightened in a faint halo around reacted grains (fig. 10D) to form a rim half as wide as the original diameter of the olivine. Irregular clasts of plagioclase commonly are surrounded by a narrow reaction rim of more sodic plagioclase (fig. 10E) ; in one example, reversed zoning was observed. Pale-pink to orange- or red- brown clasts of spinel(?) typically have reacted with the matrix to produce a compositional zoning (dark- ening) on the spinel and a corona of colorless, often radially fibrous material (feldspar) (Roedder and Weiblen, 1972; fig. 10F). Quartz has reacted with the matrix to form a corona of pyroxene; the pyroxene rim is typically separated from the quartz by a thin 3This type of glass difi'ers conspicuously from “splash" glass that also cements some breccias, the largest of which is 14306, The "splash" glass contains very irregularly distributed fragmental material and is highly vesicular. band of colorless glass (fig. 100). In addition, both orthopyroxene and clinopyroxene appear to be com- positionally zoned by reaction. Opaque minerals have also reacted, but these have not been closely examined. The degree of reaction between minerals and matrix of the breccia containing them is very erratic, and grains showing well-developed reaction may occur side by side with those showing no reaction. However, there does appear to be a general sequence of reaction and also a correspondence between degree of recrys- tallization of the matrix (Warner, 1971, 1972) and degree of reaction. The earliest sign of reaction is the conversion of olivine to orthopyroxene + ilmenite. Interchange of components between the olivine and matrix is shown by light halos around such reaction rims. This reaction is observed in rocks in which no other minerals are visibly affected. The next reaction that occurs at a stage at which the pyroxene-ilmenite rims on olivine are still poorly developed is peripheral darkening of translucent spinel grains. Compositional zoning of plagioclase is commonly observed in rocks in which some olivine grains are completely reacted, and spinel grains may exhibit coronas of radially fibrous plagioclase. The olivine reaction is at best poorly developed in the glassy rocks; hence, the other, more advanced reactions probably occurred concomi- tantly with crystallization or recrystallization of the matrix of those rocks that now contain little or no glass. Careful examination of the phases involved may allow definition of facies of metamorphism. However, it should be emphasized that disequilibrium is the rule in these rocks and that meaningful temperature-pres- sure conditions of metamorphism will be most difficult to ascertain. Table 7 summarizes the observed reaction phenom- ena. The two F2 rocks contain fairly abundant glass clasts and microclasts. The remainder of the rocks exhibiting this reaction are classified as F4 ; only one (14306) contains a minor amount of glass. All the rocks in which we have observed the reacted plagio- clase and spinel are type F4 and glass free. Preconsolidation mineral reactions—The same evi- dence of disequilibrium is found between porphyro- clasts and their matrix in individual clasts as between mineral fragments and the youngest matrix in certain fragmental rocks: olivine has reacted to form ortho- TABLE 7. — Distribution of mineral reactions Sample Rock No. type Mineral reaction 14301 F2 Olivine. 14307 F2 0. 14304 F4 Olivine, spinel, plagioclase. 14306 F4 D0. 14308—14311 F4 D0. 14319 F4 DO. 14321 F4 Do. THIN-SECTION PETROLOGY 17 pyroxene and ilmenite, plagioclase has developed a sodic rim, and spinel has reacted to form a felsic(?) corona and complementary compositional zoning within the spinel. We have, in fact, noted these reactions in clasts in all of the rocks in which matrix-mineral clast reactions were noted above. In addition, we have observed reaction rims in clasts in several glassy frag- mental rocks in which no young matrix reactions were seen. These rocks include members classified as types F1, F2, F3, and F4. We can only conclude that the reequilibration that affected these clasts took place prior to their incorporation into the Apollo 14 frag- mental rocks. PARTIAL FUSION In several fragmental rocks (14270, 14303, 14305, 14306, 14315, 14318), quartz and quartz-alkali feld- spar intergrowths were partly melted. Isolated quartz grains and those at the edges of quartz-feldspar clasts have colorless glass rinds that have reacted with the matrix to form characteristic pyroxene coronas. In the interiors of quartz-alkali feldspar clasts, partial fusion began along the grain boundaries of quartz and feldspar (fig. 10H). As in the mineral reactions described above, the effects of partial fusion are quite erratic, with extensively fused clasts occurring side by side with similar clasts showing no fusion. Partly melted clasts occur both as fragments isolated in the matrix of the rock and as clasts within larger frag- mental clasts. In general, evidence concerning the time of melting, before or during formation of the breccia, is ambiguous. In one sample, however, melting clearly occurred after consolidation of the breccia: a melted part of a fragment of the metaclastic rock has in- truded adjacent matrix (fig. 11A). Similarly, reac- tion between the matrix and partly melted quartz at the edges of some clasts also indicates fusion after formation of the breccia. In some rocks, the glass resulting from partial fusion of quartz-feldspar intergrowths is the only glass remaining in the rock. (See also Dence and others, 1972.) In other rocks, material composed of fine-grained alkali-feldspar( ?)-bladed quartz inter- growths suggests that melted clasts have devitrified. Devitrified quartz-feldspathic material is remarkably similar, both texturally and in mode of occurrence as both angular clasts and irregularly “intrusive” bodies, to Apollo 12 rock 12013 (Drake and others, 1970; James, 1971). DEFORMATION 0F MINERAL AND LITHIC CLASTS All the varieties of mineral and lithic debris in the fragmental rocks described above show various de- grees of deformation. Most plagioclase microclasts are mildly deformed by fracturing, and many show weak lattice distortion by undulose extinction; some are more severely deformed, as evidenced by mosaic tex- ture and conversion to maskelynite. Many pyroxene microclasts are undeformed or weakly fractured, but some have well-developed shock-formed planar struc- tures. Olivine microclasts are more commonly de- formed, usually by development of broad kink bands. The lithic fragments responded in the same ways to deformation. Some of them have textures ranging from those of weakly deformed cataclasites to severely shocked rocks in which maskelynite is well developed. Cataclastic rocks, composed of partly crushed, incom- pletely recrystallized material, are especially conspic- uous among the small fraction of coarse-grained gabbroic clasts (fig. 118). Apparently plagioclase is more susceptible to recrystallization in these rocks, and the mafic silicates remain as crushed aggregates. Isolated mineral clasts, especially of olivine and pyroxene, closely resembling the same minerals in these cataclasites, occur sparsely in the fragmental rocks. Marginal crush zones and partial mixing with debris of the host rock matrix (fig. 110) indicate that some cataclasis occurred after isolation of the clast from its source rock. This texture is remark- ably similar to marginal crush zones on clasts in the impact breccias of the Vredefort ring, South Africa (Wilshire, 1971). Gradations from crushed rock that still retains its original textural identity to highly disaggregated debris in which fragments of neighboring minerals have been partly mixed are observed. Cataclastic de- formation of an olivine-bearing rock of intermediate composition may have been the source of a small piece of soil breccia interpreted by Taylor and Marvin (1971) as fragmented dunite that was invaded by noritic matrix material. Some gabbroic rocks (fig. 11D) are laced by irregu- lar crush zones showing varying degrees of recrys- tallization, and some olivine clasts show moderate to extensive crushing. Most anorthosite clasts that are moderately crushed were polycrystalline aggregates whose original textures appear to have been meta- morphic. Some partly crushed gabbros have igneous textures, others are too small to determine the original texture. Finer grained light-colored cataclastic rocks may represent cataclastically deformed cumulates or D2 metaclastic rocks (table 4), and dark ones may have been derived from both basaltic rocks and D4 metaclastic rocks (table 4). Severely shocked rocks in which maskelynite is well developed include intersertal basalts, D2 metaclastic rocks, gabbros, and plagioclase cumulates. COMPOUND CLASTS Virtually every rock examined contains some com- pound clasts (fig. llE—G), and in many, such clasts PETROLOGY AND STRATIGRAPHY, FRA MAURO FORMATION, APOLLO 14 SITE 18 CLASSIFICATION OF THE ROCKS 19 are abundant. Many of these are clearly fragments of breccias within which Clast and matrix are readily identified. Others consist of two or more lithologies without a clear host-Clast relation; usually, however, this can be ascertained by examination of several sections. In table 8 the common compound clasts are listed with the host lithology and clasts identified. The first column of table 8 gives the sample number and the main types of simple, noncomposite clasts observed. (See fig. 6 for the range and proportion of clast types in the entire sample.) For the most part, host-Clast lithologies are present as simple clasts in the sample. It is significant that only one of the compound clasts examined has an igneous host, the remainder being either light or dark metaclastic rocks. There is much greater variety among the included clasts than within their hosts; these clasts include not only an abundance of metamorphic rocks, but also cumulates, graphic quartz-alkali feldspar intergrowths, and intersertal, ophitic, and intergranular basalts. In contrast to sim- ple clast populations, intergranular basalts are more abundant than intersertal and ophitic types. More rarely, a third generation of breccia is evidenced by clasts enclosed in second-stage clasts (table 8, column 3), and in one example (14308), a probable fourth generation was recognized (table 8, column 4). Four FIGURE 11. — Photomicrographs of Apollo 14 rocks. A, Photo- micrograph of a completely melted quartz—alkali feldspar(?) clast in a dark metaclastic (D4) clast. The glass was squeezed out of the clast and intruded into the adjacent matrix. 14306,65. B, Photomicrograph of cataclastically crushed “troctolite.” Close view of crushed but not sheared olivine and plagioclase aggregate. 14083,8. C, Photomicro- graph of crushed zone along edge of anorthosite clast. Top of photograph is normal dark matrix of the host rock; below that is a partly mixed zone (now recrystallized), followed downward in the photograph by unmixed crushed anorthosite, and at bottom by relict anorthosite, crossed nicols. 14306,3. D, Photomicrograph of gabbro clast laced by thin crush zones now partly recrystallized, crossed nicols. 14321,21. E, Photomicrograph of compound clasts. Clast of plagioclase hornfels with decussate texture in a finer grained D2 metaclastic clast. 14306,60. Crossed polarized light. F, Photomicrograph of compound clast in coarse light-colored matrix (lower left corner of photograph). Clast is composed of several dark aphanitic fragments of metaclastic rock (D4) in a fine-grained slightly lighter matrix. The largest clast (right side) has light-colored metaclastic (D2) clasts within it. 14306,54. G, Photomicrograph of compound clast. Dark aphanitic (D4) clast (left half of photograph) en- closes two clasts of light metaclastic rock (D2), the top one » of which contains a large kink—banded olivine porphyro- Clast. 14306,4. H, Photomicrograph showing “pellet” texture characteristic of F2 rocks. Round particles in field of view are metaclastic rocks, but well-rounded variolites, vitro- phyres, and intersertal basalt fragments also are present in the rock. 14318,44. Bar scale 0.5 mm long. TABLE 8. — Clasts and matrix types in compound clasts Sample No.; main types Host clast Clasts in 1 Clasts in 2 Clasts in 3 of simple clasts l 2 3 4 14042—A1, D2 ............. Glass D4, A1 14047—A1, A2, D2 ....... Glass D2, A1 14049—A1, D2 .............. Glass D: 14063—D4 ________ D4 D1, D2 14066—D4 ..................... D2 D3» c7) D2y 1 D4 D2 14301—D2 .................... D2 CJ, (1337, D1, D4 C1, CT: D1 14303—D4 ..................... D4 Co, D2 14304—134 ..... _.. D4 D1, D2: D1 14305—D4 ..... D4 D1, D2 14306—D4..... D4 C1; D1, D4 D2: D4 14307—D2 ..................... D2 D3, D1 D4 cm I D3 Dl’ D4 14308—D4 ..................... D4, D4 D~_> D4 D5 4 D4 14313—A1, D4 .............. C7, D3 D4 D4 D3 D4 14314—D4 ____________________ D4 132,134, D3, 14315—132, D4 .............. D4 A’, C2; D4 generations of breccias were also observed in sample 14821 by Duncan and others (1972). In one sample (14306), the clasts within host D4 clasts were counted; although the proportions of these clasts are measurably different from those of the first generation of breccia, the essential properties of the F4 class of rock 14306 are retained (compare positions for BC and 14306 in fig. 12A and B). It is interesting and significant that older generations of clasts are of metaclastic rocks, clearly showing a history of multi- ple brecciation and metamorphism. CLASSIFICATION OF THE ROCKS Our classification of the Apollo 14 hand specimens is based primarily on coherence and on proportion of light and dark clasts and secondarily on the abundance of clasts and the proportion of glass (Jackson and Wilshire, 1972). The abundance of clasts (again arbi- trarily limiting this term to fragments greater than 1 mm in diameter) is difficult to evaluate in thin sections, both because of the small sample size repre- sented even by many sections and because of section- ing effects. Nevertheless, insofar as we could judge, the observations from hand specimens were supported by those from thin sections. We had difficulty in find- ing 1.0-mm fragments in rocks that we had classified as F, and F3, but no difficulty in making statistically meaningful counts of clasts in types F2 and F4. (See fig. 6.) Considering the relative abundance of the different clast types (fig. 6), it is clear that the light and dark clasts serving as a basis for the hand-speci- men classification are predominantly light and dark metaclastic rocks. In the following section, other less abundant lithic clast types are added to the light or 20 PETROLOGY AND STRATIGRAPHY, FRA MAURO FORMATION, APOLLO 14 SITE All A ' D1+2+3 ’04 >lmm )f\ A Glass Light clasts Dl+2 +3 D4 Dark metaclastic rocks ALLC'D]+2+3 ‘04 .l-l mm C Igneous Light clasts Dl+2+3 C FIGURE 12.—Triangular plots of clasts and microclasts of glass, igneous, and metamorphic rocks. A, Triangular plot of glass (A), light—colored metamorphic rocks (Dl-Dg), and dark-colored metamorphic rocks (D4), showing fields of rock groups F1, F2, and F4. B, Triangular plot of micro- clasts, as in part A above. C, Triangular plot of clasts of dark metaclastic rock proportions where appropriate, but the same results would be obtained if they were ignored. Coherence was impossible to measure in thin sec- tion, but the percentage of glass clasts was not. Figure 12A shows a triangular plot comparing the proportion of glass clasts to those of light-colored rocks (the total of D1, D2, and D3 types) and dark-colored rocks (D4) in thin sections available to us. A clear separation is Dark metaclastic rocks ALL A ‘Dl+2+3 '04 .l—l mm .049 .318 \ Fl/\_\'313775£'77e ,L 31 V 047v 31 B A Glass Light clasts 01.2 +3 C + D Igneous and meta- morphic rocks ALL A-ALL B-ALLC+D .l—lmm 4A Mineral clasts B A Glass D igneous rocks (C), dark-colored metamorphic rocks (D4), and light-colored metamorphic rocks (D1—D3), showing fields of rock groups F1, F3, F3, and F4. D, Triangular plot of clasts of igneous and metamorphic rocks (C and D), glass (A), and mineral grains (B), showing fields of rock groups F1, F2, F3, and F4. found between F1, F3, and F4 types in terms of light- and dark-clast ratios. It is also apparent that the F2 rocks are more apt to contain glass than the F4 types. Much better statistical reliability can be obtained from the microclast data. The same plot for fragments in the size range 0.1 to 1.0 mm is given in figure 123. The same clear separation of light and dark clasts of F2 and F4 rocks is apparent. Furthermore, the propor- tion of D4 clasts in F1 rocks is, on the average, even ROCK DISTRIBUTION 21 less than that of F2, although some overlap is seen. The proportion of glass microclasts is also significantly different between groups — only two rocks classified as F4 types contain glass microclasts, whereas all of the F2 types contain significant amounts, and F1 rocks contain more than 50 percent glass in this size range. Only two F3 rocks were available for counting, and sample 14082 proved to be layered; the counts suggest that its darker layered component is like that of F4 material, whereas its lighter component is unlike any of the other rocks. Sample 14063 has the same clast distributions as the F, group, but differs in having a white friable matrix. A plot of microclasts of igneous rocks (all of group C in table 3) against dark (D4) and light (D1, D2, and D3) metamorphic rocks is given in figure 120. The light and dark metamorphic rock separation is again apparent between F, and Fl—Fg groups. The proportion of igneous rocks in the F1 rocks is clearly very low, and there is an apparent tendency for F2 rocks to contain fewer igneous rock clasts than R, rocks, although the areas overlap somewhat. Finally, all lithic microclasts were plotted against glass and mineral fragments in figure 12D. While the clear separation of types F1, F2, and F3 is largely due to glass content, it is apparent that type F2 contains considerably higher proportions of lithic than mineral fragments in this size range. In contrast, the reverse is generally true for the other groups. F4 rocks tend to have higher pyroxene-plagioclase ratios than F1 and F2 rocks. With these features in mind, we can summarize the megascopic and microscopic characteristics of the four classification groups. Rocks in the F1 category are friable and contain few clasts larger than 1.0 mm. Where such clasts occur, they are very dominantly light colored. In the microclast size range, fragments consist of more than 40 percent glass with respect to mineral and lithic fragments. Mineral fragments are composed of 50—60 percent plagioclase, 35—50 percent pyroxene, and less than 10 percent olivine. Lithic fragments are strongly dominated by light-colored metamorphic rocks; ig- neous rocks make up less than 5 percent of the total lithic microclast population. Reactions between clasts or microclasts and matrix have not been observed, and multiple clasts are rare. Rocks in the F3 category are moderately coherent to coherent, contain abundant clasts, and have a strik— ing pelletlike texture (fig. 11H). The clasts are domi- nantly light colored, but not so exclusively as those of F1 rocks. In both clast and microclast size ranges, the fragments consist of more than 50 percent lithic frag- ments with respect to mineral and glass fragments. Glass fragments are everywhere present, however. Mineral fragments are principally plagioclase (60—90 percent), but olivine and pyroxene occur in a wide range of proportions (fig. 6). Lithic fragments are dominated by light-colored metamorphic rocks but contain 5—30 percent igneous rock fragments. Reac- tions between clasts or microclasts and matrix are present in at least two samples, but, in general, they are poorly developed or absent. Only two rocks in the F3 category were sectioned, and both have friable white crystalline matrices. One of these exhibited rather unusual banding. The dark bands exhibit F4 properties in general (figs. 12B—D) , but the rock is more friable. The light bands contain no D4-type clasts and hence are unlike any of the other groups. The second rock resembles the dark bands in the first. Rocks in the F4 category are the most abundant in the collection. They are moderately coherent to tough and contain abundant clasts that are generally more angular than those in F2 rocks. The clasts are domi- nantly dark colored, to the extent that there is no overlap in this respect with F1 and F2 rocks. In both the clast and microclast size ranges, glass is very scarce (figs. 12A, B) ; it has been observed only in two rocks in this category and then in minor amounts. Mineral fragments are relatively abundant with re- spect to lithic fragments and have approximately the same proportions as do those of F2 rocks (fig. 7). Lithic fragments are dominantly dark metaclastic rocks, which make up 50 percent or more of the lithic clasts (fig. 12C). Igneous rocks show a wide range (4—20 percent) of relative lithic clast abundance. Reactions between clasts or microclasts and matrix are common and occur in at least five of the rocks we examined in thin section. We emphasize that this classification is descriptive. Without intending to imply a genetic connotation, we note that the F1 group of fragmental rocks corre- sponds to the “porous, unshocked microbreccias,” and the F3 and F4 groups to the “thermally metamor— ph0sed microbreccias” of Chao, Boreman, and Des- borough (1971). ROCK DISTRIBUTION Figure 13 shows the field distribution of Fl—F4 rocks (locations are from Sutton and others, 1971). Homogeneous crystalline rocks, believed to be clasts broken free from F2 and F4 fragmental rocks, are not shown on the map. It is at once apparent that the sample abundance is heavily weighted toward sites close to the lunar module. This would be even more conspicuous if samples smaller than 50 g were deleted. However, there is a regular pattern in the rock-type distribution: 22 PETROLOGY AND STRATIGRAPHY, FRA MAURO FORMATION, APOLLO 14 SITE Comprehensive Sugle 2506 L 2510 303. 255A 309. 26k. 312. 265A 314. 315A 316 A 317 A 269‘ 3186 270. 319. 271A 320. 272A 273A EXPLANAI‘ION ,/-~-\' c1 A Fl A n 0 {3,4 266. 267A 268A Contingency A MNE— 041° 007A SR 252 A 33::Sam21e L260 427‘ 430‘ . j t; F \(\ 0750 076. 080A 432.4450L j 30". I“ \. TRIPLET K 307‘ 100 0 HUMAN ' 438A452A \ 00 METERS ALIA 1053 A \ FIGURE 13,—Distribution of fragmental rock types at the Apollo 14 landing site. 1. All F1 types were collected from the smooth ter- rain (fig. 2) outside the continuous ejecta blan- ket and blocky rays of Cone crater. 2. The great majority of F: rocks were collected from sites far removed from Cone crater (fig. 13), and some were almost certainly collected from the discontinuous outer parts of the Cone crater ejecta (Swann and Field Geology Team, 1971). F4 type rocks also make up a sizable proportion of rocks collected from the outer part of Cone crater ejecta (fig. 13). 3. With the single exception of rock 14051, which was thoroughly dust covered when examined and may well be incorrectly classified, the sam- ples collected from near the rim of Cone crater are F3 and F4 types. (See also Quaide, 1972.) That dark-clast-rich breccias are indeed the dominant variety of rock near Cone crater is further supported by the astronauts’ descrip- tions and by the surface photographs of boul- ders on the flanks of Cone crater (Swann and Field Geology Team, 1971; Hait, 1972). In terms of the general sequence of deposits in ejecta blankets (Shoemaker, 1960), the material col— lected from closest to the crater is presumed to repre- sent the lowest units reached during crater excavation. At Cone crater, that material is fragmental rock with dominantly dark clasts (F3 and F4). Material collected from farther out on the ejecta blanket (F2 and F4) is presumed to represent material from higher levels in the precrater section. Whether the mixture of F2 and F4 types in ejecta at the outer, discontinuous part of the Cone crater rays (fig. 13) is due to a thin layer of F2 or to complex intermingling of rocks by ejection through precrater fractures (Gault and others, 1968) cannot be ascertained. Our interpretation, therefore, is that the lithologic units described above are stratified in the ridgelike terrain excavated by Cone crater. The surface of the stratigraphic pile away from the young ejecta blanket of Cone crater consists of soils and weakly lithified soil breccias (F1) . Light-clast-dominant coherent but glassy fragmental rocks (F2) form the upper strata and most likely crop out in the upper walls of Cone crater. A still deeper unit of dark-clast-dominant coherent largely crystalline fragmental rocks (F4) was pene- trated by the Cone impact, and these rocks are con- centrated near the rim. The unique “white rocks” (F3) form layers in dark-clast-dominant fragmental rocks close to the rim of Cone crater and are pre- sumed to represent the lowest parts of the section excavated by Cone crater (Sutton and others, 1972; Hait, 1972). FRA MAURO FORMATION Whether the stratigraphic units described in the preceding section represent unmodified Fra Mauro Formation, reworked or partly reworked Fra Mauro Formation, or more than one formation remains an open question. The pre-mission map of the Cone crater area (Eggleton and Offield, 1970) shows craters 0f Eratosthenian age as much as half a kilometer in diameter and a crater of Imbrian age approximately 1 km in diameter located in the immediate vicinity of the landing area. Shoemaker (1972) further contends that the Fra Mauro surface was in a steady state for craters as much as 1 km in diameter; if so, this sug- gests considerable reworking of the upper parts of the original Fra Mauro deposits. In the immediate vicinity of the landing site, known craters half a kilometer in diameter may be expected to have ejected materials over the future Cone crater site to depths of perhaps 5—15 m (H. J. Moore, oral c0mmun., 1972). This is considerably less than the section ultimately sampled by the Cone event. If unrecognized larger impacts did indeed affect the Fra Mauro surface in the vicinity sampled, it is apparent that they did not modify the radial ridge pattern typical of the Fra Mauro surface. Hence, the bulk of the material ejected from Cone crater probably represents strata deposited before the oldest craters in the immediate area were formed. The uppermost units (Fl—F2) at the Apollo 14 land- ing site are lithologically distinct from underlying units and cannot be the result of reworking of lower units, despite suggestions to the contrary (Quaide, 1972; Dence and others, 1972). Hence, they must represent either the uppermost surviving units of the Fra Mauro Formation or younger, post-Fra Mauro rocks. The only tangible evidence we have bearing on this question is apparent gradational thermal behavior between F; and F4 rocks: metamorphic reactions be- CONCLUSIONS 23 tween mineral clasts and matrix are much better devel- oped in F4 rocks, but olivine has reacted locally with the glassy matrix of F3 breccias. This suggests that F3, F3, and F4 breccias represent a single cooling unit of the original Fra Mauro Formation. The viability of our conclusions is dependent on the correct recognition of the Fra Mauro matrix proper. The presence of breccia clasts in boulders ejected from Cone crater that are bigger than many of the returned samples does not allow an a priori identification of the youngest matrix in any particular sample as Fra Mauro. However, the packing density of identifiable large clasts is quite low (Hait, 1972), and the prob- ability of sampling the youngest breccia matrix in the boulders appears to be high. If the boulders around Cone crater are themselves clasts in the Fra Mauro matrix, the minimum clast size would have to be larger than several meters. “Turtle Rock” (Sutton and others, 1971), a 2—m-wide boulder in north boulder field, appears to have the same clast population as the two samples (14321, 14319) taken from the top of the boulder. Still larger boulders photographed near Cone crater show no signs of huge clast boundaries within them, and we find it difficult to believe that every block in the field is indeed a clast. Lacking tangible evidence to the contrary, we therefore presume that the youngest matrices of most breccia samples are the Fra Mauro proper and not matrices of older breccias. lf we accept the F2, F3, and F4 rocks as samples of the Fra Mauro Formation and the idea that these rocks represent a single cooling unit, then local tem— peratures on the order of 1,000°C are required. (See also Quaide, 1972; Dence and others, 1972.) Terres- trial examples suggest that temperatures of this mag— nitude are restricted to areas relatively closer to the locus of impact. In the Ries basin (Germany), the best—known large terrestrial impact crater that retains its ejecta deposits, the high-temperature ejecta (sue- vite) extends out only about half a crater diameter (Pohl and Angenheister, 1969 ; Dennis, 1971), whereas lower temperature breccia deposits (Bunte Breccia) underlie the suevite and extend nearly a crater diam- eter from the crater rim (Dennis, 1971). However, these deposits have been subjected to lengthy erosion, and the suevite, being in the most easily eroded posi- tion, may originally have covered a much larger area. The view that the Apollo 14 high-temperature rocks are derived from earlier basin-forming events rather than from the Imbrium impact (Dence and others, 1972) does not solve the problem because the source area of the Fra Mauro rocks was just as distant from earlier major basins, such as Serenitatis or the south Imbrium basin (Wilhelms and McCauley, 1971), as the Apollo 14 site is from the Imbrium basin. We must conclude, therefore, that the magnitude of major lunar basin-forming events is such that small terrestrial craters do not provide adequate models for tempera- ture distribution in the ejecta deposits. CONCLUSIONS The site descriptions at the Apollo 14 landing area and the nature of the rocks returned by the crew strongly support Gilbert’s (1893) conclusion that the area is covered by ejecta from the Imbrium basin. We can apply petrologic observations to a more detailed interpretation of the nature and stratigraphy of the Fra Mauro Formation by attempting to reconstruct an inverted stratigraphic section in the Cone crater ejecta. And, on the basis of the clasts in What we believe to be Fra Mauro rocks, we can say something about the terrane that existed in the Imbrium area prior to the enormous impact that produced that basin. FRA MAURO FORMATION The distribution of rocks collected in the Apollo 14 landing area suggests that the Fra Mauro Formation is stratified. The deepest unit sampled is composed of fragmental rocks rich in dark clasts with interlayered white fragmental rocks (F4 and F3), the next over- lying layer is composed of well-indurated fragmental rocks (F3) rich in light-colored clasts, and the upper- most unit is unconsolidated to weakly lithified (F1) regolith. F1 type fragmental rocks, characterized by an abun- dance of glass, weak lithification, and scarcity of clasts bigger than 1 mm, were collected from areas not cov- ered by Cone crater ejecta, and in most respects they closely resemble the unconsolidated soils of the area (Carr and Meyer, 1972; Quaide, 1972). They appear to represent poorly lithified regolith that formed either by disaggregation and weak re-lithification of F2 ma- terial or from an originally unconsolidated layer of fragmental material overlying the F2 material. Com- parison of the clast populations of F1 and F2 shows that F1 has more clastic glass, much more pyroxene in the microclast size range, and fewer clasts of igneous rock than F2. It is possible that all of these are effects of “gardening” of F2 material by small impacts, which would tend to produce glass and break down the coarser grained clasts more rapidly than the fine—grained clasts. However, it remains possible that the regolith and F1 rocks are derived from a surficial layer overlying F2 material that was originally of a different composition. In our opinion, F3 rocks represent the upper, con- solidated part of the Fra Mauro Formation and are probably exposed in the upper walls of Cone crater. Glass-filled fractures cross clasts and matrices alike in both F2 and F4 rocks and are presumably a conse- quence of the Cone impact, indicating that both rock 24 PETROLOGY AND STRATIGRAPHY, FRA MAURO FORMATION, APOLLO 14 SITE types were consolidated at the time of that event. The F2 rocks contain some elastic glass, some of which was soft during consolidation of the rocks, and are partly bound by glass. Moreover, a few of these rocks show the earliest signs of postconsolidation metamorphic reaction between mineral clasts and the matrix. The simplest explanation of these features is that part of the ejecta was hot when deposited and that insulation and compaction by upper units allowed induration and metamorphism of lower units. The preservation of glass and very limited metamorphic reaction suggest that all these rocks were near the original top of the Fra Mauro ejecta blanket. Comparison of the clast populations of F2 and F.1 rocks, as well as their general textures, shows that these are distinctive rocks and lack intermediate types ; this is especially evident in the relative abundance of light- and dark—colored metaclastic rocks. Hence, we infer that the contact between the F2 and F4 units in the Fra Mauro Formation is sharp. However, since neither glass nor metamorphic reactions are unique to F2 or F4, it seems likely that the thermal metamorphic effects are gradational across the contact. In this re- spect, the deposits are analogous to cooling units of welded tuff deposits. F3 and F4 rocks represent the lower consolidated part of the Fra Mauro Formation sampled. These rocks are largely crystalline, and the original glass clasts and glassy matrix material have largely been annealed and devitrified. Postconsolidation metamorphic reac- tions are likewise considerably better developed than in F3 rocks. This feature and the distribution of rocks around Cone Crater suggest a deeper source for F4 rocks in which the thermal blanketing effect of over- lying material allowed slower cooling and partial re— equilibration of an assortment of material thrown together by impact. Local temperatures were high enough (about 1,000°C) to reset partially or com- pletely certain radiogenic systems used to date the Apollo 14 rocks (Silver, 1972; Papanastassiou and Wasserburg, 1971; Turner and others, 1971; Comp— ston and others, 1971), but the very erratic distri- bution of metamorphic reactions and partial fusion effects indicates uneven heating of debris in the blanket. PRE-IMBRIUM TERRANE The stratigraphic sequence inferred for samples collected from the Cone crater ejecta is based on interpretations gleaned from small terrestrial craters. Studies of these craters have shown that, in a general way, the stratigraphy of the target is inverted and that younger beds are exposed farther from the source in the ejecta blankets. Application of these general rules to the Imbrium basin is perhaps an unwarranted extrapolation, and we cannot be fully confident that ejected materials from such large basins will behave in the same way. However, it is noteworthy that the samples collected 550 km from the rim of the Imbrium basin, and only 150 km from the mapped edge of the ejecta deposit, are of a character suggesting a rela- tively shallow origin. Glassy materials in the F3 rocks were presumably produced during the Imbrium event and give little information on the lithologic character of the source material. The mineral microclast population, however, offers more potential. We have pointed out before that these clasts are too large and too numerous to have been supplied by disaggregation of the most abundant (fine-grained metaclastic rocks) clasts in the breccias. Early results from Apollo 15 (Wilshire and others, 1972) show that coarser rocks of appropriate composi- tion were present in the Imbrium target area and that these may have supplied the mineral debris if com— minution was nearly complete. The lithic clasts in the Apollo 14 rocks are of a great variety, indicating a very complex target area of the Imbrium impact. Igneous clasts include fresh basalts of several varieties that are in general richer in feld- spar than basalts from the mare regions and rocks with cumulus and hypautomorphic textures that sug— gest igneous crystallization and differentiation at depth. Metamorphic clasts, the dominant clast type in Apollo 14 rocks, are of both fine-grained dark-colored types and coarser light—colored types. All of them appear to have formed by impact fragmentation and thermal metamorphism, clasts of each type containing relicts of the fragmentation and incomplete reequili- bration of materials during thermal metamorphism. Both the igneous and metamorphic rock types are themselves clasts in the common compound breccia clasts. In the record revealed by small thin sections, as many as four fragmentation events have been recognized, each apparently followed by thermal metamorphism. These relations show that impact fragmentation and subsequent thermal metamorphism happened many times and that materials similar to the Fra Mauro Formation were themselves disrupted by the Imbrium event. It seems likely that the Serenitatis blanket, which must certainly have covered most of the Imbrium basin area (fig. 14), provided one gen- eration of these materials. Hence, the pre-Imbrium terrane, presumed to have existed prior to about 3.9 to 4.2 billion years ago (Silver, 1971, 1972; Wasserburg and others, 1971) was already exceedingly complex. Multiple ejecta blan- kets from very large impacts had been laid down and metamorphosed. Basalts had been erupted and incor- porated in the ejecta blankets. Very likely, some flows REFERENCES CITED 25 8 FIGURE 14.—Map showing main outer mountain rings of major lunar basins, their relative ages, and the extent of their continuous ejecta blankets (based on a Mare Orientale model for basins older than Imbrium). Drawn by D. E. Wilhelms, U.S. Geological Survey. 1, Fecunditatis. 2, “South Imbrium.” 3, Tranquillitatis. 4, Nubium. 5, Procellarum. 6, Serenitatis. 7, Humorum. 8, Nectaris. 9, Crisium. 10, Imbrium. 11, Orientale. were extruded in the Imbrium area itself, intercalated with these ejecta blankets. At depth, either in the Imbrium area or in still older basins, igneous intru- sion, crystallization, and differentiation had occurred. Processes operative prior to 3.9 to 4.2 billion years ago, so far as we can determine, were not unlike those operative later, and the Moon probably appeared then about the same as it does now. Hence, the existence of a primordial crust of “anorthosite” (Wood and others, 19703, b) or “fused shell” (Smith and others, 1970) is not established. Rather, we see in the Fra Mauro clasts a long history of impact and ejection that prob- ably records half a billion years of planetary accretion, of which the Imbrium impact was one of the last large events on the Moon. REFERENCES CITED Ave’Lallemant, H. G., and Carter, N. L., 1972, Deformation of silicates in some Fra Mauro breccias, in Third Lunar Sci. Conf., Houston, Texas, January 1972: Lunar Sci. Inst. Contr. 88, p. 33—34. Carr, M. H., and Meyer, C. E., 1972, Petrologic and chemical characterization of soils from the Apollo 14 landing site, in Third Lunar Sci. Conf., Houston, Texas, January 1972: Lunar Sci. Inst. Contr. 88, p. 117—118. (‘hao, E. C. T., Boreman, J. A., and Desborough, G. A., 1971, The petrology of unshocked and shocked Apollo 11 and Apollo 12 microbreccias, in Second Lunar Sci. Conf., Houston, Texas, 1971, Proc.: Geochim. et Cosmochim. Acta, Supp. 2, v. 1, p. 797—816. (‘hao, E. C. T., Minkin, J. A., and Boreman, J. A., 1972, Apollo 14 glasses of impact origin, in Third Lunar Sci. ‘ Conf., Houston, Texas, January 1972: Lunar Sci. Inst. Contr. 88, p. 131—132. Compston, W., Vernon, M. J., Berry, H., and Rudowski, R., 1971, The age of the Era Mauro Formation—A radio— metric older limit: Earth and Planetary Sci. Letters, v. 12, p. 55—58. Dence, M. R., Plant, A. G., and Trail], R. J., 1972, Impact- generated shock and thermal metamorphism in Fra Mauro lunar samples, in Third Lunar Sci. Conf., Hous- ton, Texas, January 1972: Lunar Sci. Inst. Contr. 88, p. 174—176. Dennis, J. G., 1971, Ries structure, southern Germany, a review: Jour. Geophys. Research, v. 76, p. 5394—5406. Drake, M. S., McCallum, I. S., McKay, G. A., and Weill, P. F., 1970, Mineralogy and petrology of Apollo 12 sample 12013—A progress report: Earth and Planetary Sci. Letters, V. 9, p. 124—126. Duncan, A. R., Lindstrom, N. N., Lindstrom, D. J., McKay, S., Woodbury, J. A., Goles, G. G., and Fruchter, J. S., 1972, Comments on the genesis of breccia 14321, in Third Lunar Sci. Conf., Houston, Texas, January 1972: Lunar Sci. Inst. Contr. 88, p. 192-194. Eggleton, R. E., 1963, Thickness of the Apennine Series in the Lansberg region of the Moon, in Astrogeol. Studies Ann. Prog. Rept., August 1961—August 1962: U.S. Geol. Survey open-file report, p. 19—31. Eggleton, R. E., and Ofiield, T. W., 1970, Geologic map of part of the Era Mauro region of the Moon—Apollo 13: U.S. Geol. Survey Misc. Geol. Inv. Map I—708. Fredriksson, K., Nelen, J., Noonan, A., and Kraut, F., 1972, Apollo 14 — Glasses, breccias, chondrules, in Third Lunar Sci. Conf., Houston, Texas, January 1972: Lunar Sci. Inst. Contr. 88, p. 280—282. Gault, D. E., Quaide, W. L., and Oberbeck, V. R., 1968, Impact cratering mechanics and structures, in French, B. M., and Short, N. M., eds., Shock metamorphism of natural materials: Baltimore, Md., Mono Book Corp., p. 87—99. Gilbert, G. K., 1893, The Moon’s face— A study of the origin of its features: Philos. Soc. Washington Bull., v. 12, p. 241—292. Hait, M. H., 1972, The white rock group and other boulders of the Apollo site—A partial record of the Era Mauro history, 1'11 Third Lunar Sci. Conf., Houston, Texas, January 1972: Lunar Sci. Inst. Contr. 88, p. 353. Jackson, E. D., and Wilshire, H. G., 1972, Classification of the samples returned from the Apollo 14 landing site, in Third Lunar Sci. Conf., Houston, Texas, January 1972: Lunar Sci. Inst. Contr. 88, p. 418—420. Jakes, Petr, and Reid, A. M., 1971, Petrology of the Fra Mauro Formation [abs]: Geol. Soc. America Abs. with Programs, v. 3, p. 611. James, 0.13., 1971, Origin of lunar microbreccia 12013 [abs]: EOS (Am. Geophys. Union Trans.), v. 52, no. 4, p. 272. James, O. B., and Jackson, E. D., 1970, Petrology of the Apollo 11 ilmenite basalts: Jour. Geophys. Research, v. 75, p. 5793—5824. 26 PETROLOGY AND STRATIGRAPHY, FRA MAURO FORMATION, APOLLO 14 SITE King, E. A., Jr., Butler, J. C., and Carman, M. F., 1971, Apollo 14 sample results— Chondrule-like bodies, grain size analyses, and estimation of lunar surface ages [abs.]: Geol. Soc. America Abs. with Programs, v. 3, p. 623. 1972, Chondrules in Apollo 14 breccias and estima- tion of lunar surface exposure ages from grain size analysis, in Third Lunar Sci. Conf., Houston, Texas, January 1972: Lunar Sci. Inst. Contr. 88, p. 449—451. Lunar Sample Preliminary Examination Team, 1971, Pre- liminary examination of lunar samples from Apollo 14: Science, v. 173, p. 681—693. McKay, D. S., Clanton, U. S., Heiken, G. H., Morrison, D. A., Taylor, R. M., and Ladle, G., 1972, Characterization of Apollo 14 soils, in Third Lunar Sci. Conf., Houston, Texas, January 1972: Lunar Sci. Inst. Contr. 88, p. 529—531. Melson, W. G., Mason, B., Nelen, J., and Jacobson, S., 1972, Apollo 14 basaltic rocks, in Third Lunar Sci. Conf., Hous- ton, Texas, January 1972: Lunar Sci. Inst. Contr. 88, p. 535—536. Papanastassiou, D. A., and Wasserburg, G. J., 1971, Rb-Sr ages of igneous rocks from the Apollo 14 mission and the age of the Fra Mauro Formation: Earth and Planetary Sci. Letters, v. 12, p. 36—48. Papike, J. J., and Bence, A. E., 1971, Apollo 14 pyroxenes: Subsolidus relations and implied thermal histories [abs.] : Geol. Soc. America Abs. with Programs, V. 3, p. 666. Pohl, Jean, and Angenheister, Gustav, 1969, Anomalien des Erdmagnetfeldes und Magnetisierung der Gesteine im Nordlinger Ries, in Das Ries— Geologie, Geophysik, und Genese eines Kraters: Geol. Bavarica, no. 61, p. 327—336. Quaide, W. L., 1972, Mineralogy and origin of Fra Mauro fines and breccias, in Third Lunar Sci. Conf., Houston, Texas, January 1972: Lunar Sci. Inst. Contr. 88, p. 627—629. Ridley, W. 1., Williams, R. J., Takeda, H., Brown, R. W., and Brett, R., 1971, Petrology of Fra Mauro basalt 14310 [abs.]: Geol. Soc. America Abs. with Programs, v. 3, p. 682. Roedder, E., and Weiblen, P. W., 1972, Petrographic and petrologic features of Apollo 14 and 15 and Luna 16 samples, in Third Lunar Sci. Conf., Houston, Texas, January 1972: Lunar Sci. Inst. Contr. 88, p. 657—659. Shoemaker, E. M., 1960, Penetration mechanics of high veloc- ity meteorites, illustrated by Meteor Crater, Arizona: Internat. Geol. Cong., let, Copenhagen, 1960, Rept., pt. 18, p. 418—434. 1972, Cratering history and early evolution of the Moon, in Third Lunar Sci. Conf., Houston, Texas, Janu- ary 1972: Lunar Sci. Inst. Contr. 88, p. 669—698. Silver, L. T., 1971, U—Th-Pb isotope relations in some Apollo 14 (and 15?) materials [abs.]: Geol. Soc. America Abs. with Programs, v. 3, p. 706. __.fi1972, Lead volatilization and volatile transfer pro- cesses on Moon, in Third Lunar Sci. Conf., Houston, Texas, January 1972: Lunar Sci. Inst. Contr. 88, p. 701—703. Smith, J. V., Anderson, A. T., Newton, R. 0., Olsen, E. J., and Wyllie, P. J., 1970, Petrologic history of the moon inferred from petrography, mineralogy, and petrogenesis of Apollo 11 rocks, in Apollo 11 Lunar Sci. Conf., Hous- ton, Texas, 1970, Proc.: Geochim. et Cosmochim. Acta, Supp. 1, v. 1, p. 897—925. Spry, A., 1969, Metamorphic textures: New York, Pergamon Press, 350 p. Sutton, R. L., Batson, R. M., Larson, K. B., Schafer, J. P., Eggleton, R. E., and Swann, G. A., 1971, Documentation of the Apollo 14 samples: U.S. Geol. Survey open-file report, 37 p. Sutton, R. L., Hait, M. H., and Swann, G. A., 1972, Geology of the Apollo 14 landing site, in Third Lunar Sci. Conf., Houston, Texas, January 1972: Lunar Sci. Inst. Contr. 88, p. 732—734. Swann, G. A., and Field Geology Team, 1971, Preliminary geologic investigations of the Apollo 14 landing site, in Apollo 14 Prelim. Sci. Rept.: Natl. Aeronautics and Space Adm. Spec. Pub. SP—272, p. 39—85. Swann, G. A., Trask, N. J., Hait, M. H., and Sutton, R. L., 1971, Geologic setting of the Apollo 14 samples: Science, v. 173, no. 3998, p. 716—719. Taylor, G. J., and Marvin, U. B., 1971, A dunite—norite lunar microbreccia: Meteoritics, v. 6, p. 173—180. Turner, Grenville, Huneke, J. C., Podosek, F. A., and Wasser- burg, G. J., 1971, 40Ar—39Ar ages and cosmic ray ex- posure ages of Apollo 14 samples: Earth and Planetary Sci. Letters, v. 12, p. 19—35. Warner, J. L., 1971, Progressive metamorphism of Apollo 14 breccias [abs.]: Geol. Soc. America Abs. with Programs, V. 3, p. 744. 1972, Apollo 14 breccias—Metamorphic origin and classification, in Third Lunar Sci. Conf., Houston, Texas, January 1972: Lunar Sci. Inst. Contr. 88, p. 782—784. Wasserburg, G. J., Huneke, J. C., Papanastassiou, D. C., Podosek, F. A., Tera, F., and Turner, G., 1971, Lunar chronology and evolution [abs.]: Geol. Soc. America Abs. with Programs, V. 3, p. 745. Wilhelms, D. E., 1970, Summary of lunar stratigraphy— Telescopic observations: U.S. Geol. Survey Prof. Paper 599—F, p. F1—F47. Wilhelms, D. E., and McCauley, J. F., 1971, Geologic map of the near side of the Moon: U.S. Geol. Survey Misc. Geol. Inv. Map I—703. Wilshire, H. G., 1971, Pseudotachylite from the Vredefort ring, South Africa: Jour. Geology, v. 79, p. 195-206. Wilshire, H. G., and Jackson, E. D., 1972, Petrology of the ' Fra Mauro Formation at the Apollo 14 landing site, in Third Lunar Sci. Conf., Houston, Texas, January 1972: Lunar Sci. Inst. Contr. 88, p. 803—805. Wilshire, H. G., Schaber, G. G., Silver, L. T., Phinney, W. C., and Jackson, E. D., 1972, Geologic setting and petrology of Apollo 15 anorthosite (15415): Geol. Soc. America Bull, v. 83, p. 1083—1092. Wilson, A. F., 1969, Some structural, geochemical, and eco- nomic aspects of the metamorphosed East Fraser gabbro and associated pyroxene granulites of the Fraser Range, Western Australia: Indian Mineralogist, v. 10, p. 46—66. Wood, J. A., Dickey, J. 8., Jr., Marvin, U. B., and Powell, B. N., 1970a, Lunar anorthosites and a geophysical model of the Moon, in Apollo 11 Lunar Sci. Conf., Houston, Texas, 1970, Proc.: Geochim. et Cosmochim. Acta, Supp. 1, v. 1, p. 965—988. 1970b, Lunar anorthosites: Science, v. 167, p. 602— 604. fiGPO 781-165 Pleistocene Geology of the Northeast Adirondack Region, New York GEOLOGICAL SURVEY PROFESSIONAL PAPE («73‘1-13‘77 , [33$be UF CAL’/\ R 786 Pleistocene Geology of the Northeast Adirondack Region, New York By CHARLES S. DENNY GEOLOGICAL SURVEY PROFESSIONAL PAPER 786 A description and interpretation of ice fronts and water bodies in parts of the St. Lawrence and Champlain basins 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. 73-600215 For sale by the Superintendent of Documents, US. Government Printing Office Washington, DC. 20402 — Price $1.45 (paper cover) Stock Number 2401-02552 Abstract Introduction _____________________________________ Acknowledgments ____________________________ Bedrock geology and topography __________________ Deposits and landforms associated with the ice sheet _______________________________________ Glacial drift Till lithology ____________________________ Direction of ice movement ____________________ Moraines in the Saranac Valley _______________ Moraines in the Great Chazy Valley ___________ Spillways on the St. Lawrence-Champlain divide- Areas of bare rock _- __________________________ Flat Rock, near Altona ___________________ South of English River ___________________ North of English River ___________________ Origin __________________________________ History of deglaciation ___________________________ Loon Lake Episode (1) _______________________ Owls Head-Redford stand (2) _________________ Trout River-Moflitsville stand (3) _____________ Malone-Schuyler Falls stand (4) ______________ Chateaugay-Cadyville episode (4, 5, 6, 7, and 8)- Covey Hill episode (9 through 15) _____________ CONTENTS WNN Hg Deposits and shore features of late-glacial water bodies Beach deposits _______________________________ Delta deposits _______________________________ Texture _________________________________ Grain size ___________________________ Sorting _____________________________ Internal structure ________________________ Location and areal relations ______________ Volume Lake Vermont _______________________________ Coveville stage __________________________ Fort Ann stage __________________________ Champlain Sea ______________________________ Age _____________________________________ Rate of delta formation __________________ Environment ____________________________ Ausable River delta in Lake Champlain- Tilt of water planes __________________________ Summary and correlation _________________________ Northeast Adirondack region _________________ St. Lawrence Lowlands in New York State _____ Champlain Lowland in Vermont _______________ Appalachian region of southeastern Quebec ____ References cited _________________________________ Index ___________________________________________ ILLUSTRATIONS [Plates 1—‘7 in pocket] PLATE 39°???”F’3" Surficial geologic map of the northeast Adirondack region, New York. Map showing bedrock geology and percentage of clasts of granite gneiss and of anorthosite in till. Map showing bedrock geology and percentage of clasts of Paleozoic sedimentary rock in till. Surficial geologic map of the moraine near Cadyville. Surficial geologic map of the moraine near Ellenburg Depot along North Branch Great Chazy River. Surficial geologic map of the area at the headwaters of English River west of Cannon Corners. Map showing beaches on Cobblestone Hill and part of Flat Rock, near Altona. Page FIGURE 1. Photograph showing bouldery till on upland between the Saranac and Salmon Rivers __________ 4 2. Map showing percentage of clasts of Hawkeye Granite Gneiss in till in the Ellenburg-Lyon Moun- tain area _______________________________________________________________________________ 6 3. Photograph showing cross section of the moraine near Ellenburg Depot _______________________ 8 4. Photograph showing moraine near Ellenburg Depot along the 01d Military Turnpike ______________ 9 III IV CONTENTS Page FIGURE 5. Map of the moraine north of Clinton Mills _____________________________________________________ 10 6. Map of the moraine south of Miner Lake _____________________________________________________ 11 7-10. Photographs : 7. Flat Rock near Altona _______________________________________________________________ 12 8. Solution pit in Potsdam Sandstone _____________________________________________________ 12 9. Vertical joint face on Potsdam Sandstone _____________________________________________ 13 10. Boulder gravel near Altona ___________________________________________________________ 13 11. Map of ice-front positions between Covey Hill and Flat Rock ___________________________________ 14 12. Photograph of recessional moraine west of Cannon Corners _____________________________________ 15 13. Photograph of recessional moraine in Salmon River valley (Clinton County) _____________________ 17 14. Map showing moraines and maximum extent of late-glacial water bodies in parts of the St. Law- rence and Champlain Lowlands ___________________________________________________________ 21 15—21. Photographs : 15. Cross section of flaggy gravel beach ridge _____________________________________________ 27 16. Beach gravel of the Champlain Sea ____________________________________________________ 27 17. Cross section of beach on Cobblestone Hill ______________________________________________ 28 18. Boulder beach on Cobblestone Hill ____________________________________________________ 29 19. Gravel in Champlain Sea delta of the Saranac River ____________________________________ 32 20. Crossbedded sand and gravel in Fort Ann delta of Great Chazy River ____________________ 32 21. Foreset beds in delta of Champlain Sea _______________________________________________ 33 22. Maps showing texture of late-glacial sands in the Plattsburgh area _____________________________ 34 23; Graphs showing texture of late-glacial sands ——————————————————————————————————————————————————— 36 24. Scatter diagram showing relation of grain size to distance from apex on late-glacial deltas ________ 37 25. Graph showing relationship of upper Pleistocene deltas and their source areas ___________________ 38 26. Graph showing rate of delta formation versus relative relief of source area _____________________ 39 27. Graph showing rate of delta formation versus area 0f drainage basin __________________________ 40 28. Map of the Ausable River delta in Lake Champlain ____________________________________________ 41 TABLES Page TABLE 1. History of deglaciation of the northeast Adirondack region, New York __________________________ 18 2. Ice-marginal and shore features near Covey Hill, Quebec, as interpreted by Prest (1970) and by Denny (this report) _____________________________________________________________________ 23 3. Upper Pleistocene and Holocene deltas of the northeast Adirondack region, New York ___________ 30 PLEISTOCENE GEOLOGY OF THE NORTHEAST ADIRONDACK REGION, NEW YORK By CHARLES S. DENNY ABSTRACT The northeast Adirondack region includes Clinton County and parts of Franklin and Essex Counties in the northeast corner of New York State. Lowlands west of Lake Champlain and southeast of the St. Lawrence River are underlain chief- ly by Paleozoic carbonate rocks and are bordered by a dis- sected plateau held up by a massive quartzitic to arkosic sandstone (Potsdam Sandstone of Cambrian age). The ad- jacent uplands and mountains are underlain by Precambrian rocks, chiefly granitic and other gneisses, anorthosite, meta- gabbro, and metasedimentary rocks; in many valleys are thick masses of sand and gravel. In the lowlands the till is a calcareous silt loam containing a few clasts of sedimentary rocks; in the plateau areas underlain by Potsdam Sandstone, the till is a sandy loam or loam containing 10—20 percent sandstone clasts; and in the mountains the till is p‘ebbly and sandy. Directional data from striae, drumlins, and grooved drift and a study of till lithology by A. W. Postel indicate that the upper Wisconsin (Woodfordian) ice sheet moved south- west across the mountains and in a more southerly direction in adjacent parts of the St. Lawrence and Champlain Low- lands. However, the surface till in two small areas north of the mountains near Ellenburg contains many clasts of rocks found only in the mountains to the south, suggesting northward transport of these clasts by glacier ice or other processes prior to late Wisconsin time. In the mountains, the history of deglaciation involves a series of episodes of moraine building, outwash deposition, and drainage diversion; in the lowlands, ice-dammed lakes lengthened northward, to be replaced eventually by the Champlain Sea. Deglaciation began with the building of massive outwash plains by southwest-flowing melt-water streams in the southwestern part of the area, probably not earlier than about 12,700 years B.P. Deglaciation proceeded in a general northeast direction and involved thinning of the marginal zone of the ice sheet. In the mountains, parts of the terminus stagnated, and melt-water streams built kames and outwash plains on and adjacent to masses of dead ice. In the lowlands, the ice sheet maintained an active front and built small moraines and ice-marginal kames. In the low- lands north of Plattsburgh, an esker about 10 miles long was built by a subglacial stream that discharged south into glacial Lake Vermont. The principal streams, the Salmon, Trout, and Chateaugay Rivers on the edge of the St. Lawrence Lowlands in Frank- lin County and the Great Chazy, Saranac, and Ausable Rivers in the Champlain drainage basin were ponded at the retreat- ing ice front and diverted across interstream divides, cutting channels, now abandoned, in drift and bedrock, removing the drift from large areas of bedrock, and emptying into the glacial lakes in the adjacent lowlands. Along the western edge of the Champlain Lowlands, small moraines were built in or near the mouths of valleys drain- ing northeast. Northeast-flowing streams were diverted south- ward along the ice edge where they eroded channels in drift and in rock. The Champlain Valley ice lobe held in glacial Lake Vermont and was probably only a few tens of miles long. The edge of the lobe retreated northward, for the oldest ice-marginal features are to the south. The Saranac River was diverted south into the Ausable River, perhaps contributing to the formation of a large delta built by the Ausable River in Lake Vermont near the end of the Cove- ville stage, perhaps about 12,600 years B.P. A prominent moraine dams the Saranac River valley where it enters the lowlands. Abandoned channels in the moraine indicate that when the ice built the eastern part of the moraine, the Saranac River entered Lake Vermont only 2 or 3 miles south of the present course of the river. A prominent moraine near Ellenburg Depot in the Great Chazy drainage was built after that on the Saranac, perhaps at the beginning of the Fort Ann stage of Lake Vermont about 12,400 years B.P. As deglaciation proceeded, glacial Lake Iroquois, an ice- dammed lake in the St. Lawrence Lowlands southwest of Montreal, overflowed to the east across the divide into the Champlain Valley, cutting a deep rock-walled gorge across the divide just north of the International Boundary near Covey Hill, Quebec. Water escaping over the divide flowed southeast along the ice front into an ice—dammed lake in the valley of the Great Chazy River. This lake overflowed, perhaps catastrophically, to the southeast across the Great Chazy—Saranac River divide, where it washed clean Flat Rock, an area of essentially bare Potsdam Sandstone near Altona that measures about 2.5 by 5 miles. Farther northeasterly retreat of the ice front caused the water coming through the gorge near Covey Hill to flow south and empty into a small ice-dammed lake in the English River valley. This lake in turn overflowed to the southeast along the ice margin into another ice—dammed lake, cleaning off the bedrock ridge between the lakes. In this fashion, several bedrock ridges between valleys were washed clean of their drift cover. The final withdrawal of the ice sheet from the northeast prong of the Adirondack uplands caused Lake Iroquois in the St. Lawrence Lowlands to drain down to the Fort Ann 1 2 PLEISTOCENE GEOLOGY OF THE NORTHEAST ADIRONDACK REGION level of Lake Vermont in the Champlain Valley, and the two lakes merged. This withdrawal may have taken place about 12,200 years B.P. The strandlines of the late-glacial water bodies in the New York part of the Champlain drainage basin are marked by beaches and deltas. Beaches are prominent features where the drift is pebbly to bouldery, that is, in areas of Potsdam Sandstone. Beaches are best developed in the bouldery de- posits of the ice-marginal streams. Deltas were built where the principal streams from the mountains emptied into the late-glacial water bodies. Deltas built at low levels are composed in part of material eroded from those at higher levels. The presence of beaches of the Fort Ann stage in areas of washed bedrock and of Fort Ann deltas whose tops are below the highest stand of that stage suggest that the level of Lake Vermont rose perhaps 50—75 feet during Fort Ann time. The Champlain Sea came into existence when the retreat- ing ice front reached a position a short distance north of Quebec City, thereby allowing the sea to invade the St. Lawrence and Champlain basins. In the Champlain Valley the marine submergence lasted from about 12,000 to about 10,500 years B.P. The deposits of the Champlain Sea can- not be distinguished lithologically from those of glacial Lake Vermont; the distinction is based on altitude and the remains of marine or brackish-water organisms, largely mollusks. Streams from the mountains built deltas in the sea. The size of the deltas appears to be related to the size and aver- age slope of the drainage basins supplying the sediment. This relationship suggests that the environment of Champlain Sea time was not periglacial but more like the present. The older Lake Vermont deltas do not show such a relationship; they were built largely by melt-water streams. Lake Champlain came into existence about 10,000 years B.P., when gradual uplift caused the marine level to fall until the connection with the ocean was cut off. Except for the Ausable River, none of the larger streams emptying into the lake has built a delta into it comparable with those of earlier date. The crows-foot delta of the Ausable River is unique. This delta was built sometime after the formation of Lake Champlain, when the river’s course below the chasm was changed by piracy. Delta building was apparently much more rapid during Champlain Sea time than it has been during the life of the modern lake. INTRODUCTION The northeast prong of the Adirondack Mountains rises between the broad St. Lawrence and Champlain Lowlands. During deglaciation the ice sheet blocked these valleys, ponding the streams that flowed north and east from the mountains. Some streams were diverted across divides into adjacent valleys. In the mountains, the marginal zone of the ice sheet stag- nated, and kame terraces were built by ice-marginal streams. In the broad lowlands, the ice sheet main- tained an active front and built small moraines largely of water-laid materials. As the ice front re- treated out of the mountains, the ice-dammed lakes grew in size until glacial Lake Iroquois (Coleman, 1937) in the St. Lawrence Valley overflowed across the divide into glacial Lake Vermont (Chapman, 1937 ) in the Champlain Valley. The overflow washed clean large areas of bedrock. The ice front moved back and forth, opening and closing lake outlets. Water levels fell, rose, and fell again. Ultimately salt water invaded the St. Lawrence River lowlands, initi- ating the Champlain Sea episode. Deltas and beaches record the presence of the ancient water bodies that smoothed the contours of the glaciated landscape. The 'author spent about 14 months in the field, from 1961 to 1969, and mapped in detail the lowland and adjacent foothills west of Lake Champlain be- tween the Ausable River and the Canadian border; this includes the Dannemora and Mooers 15-minute quadrangles and the New York part of the Platts- burgh and Rouses Point 15-minute quadrangles (Denny, 1967, 1970). A reconnaissance study was also carried on in the remainder of Clinton County and adjacent parts of Franklin County (pl. 1). In company with the late A. Williams Postel of the US. Geological Survey, the lithology of stones in the till was analyzed in the field. Rapid mechanical analy- ses of sandy sediments were also made in the field. No samples of drift were studied in the laboratory. No attempt is made here to summarize the pre- vious geologic work in the region. The bedrock has been studied by Postel and his associates (Postel, 1952; Postel, Dodson, and Carswell, 1956; Postel, Wiesnet and Nelson, 1956; Nelson and others, 1956) and by Fisher (1968), Buddington (1937, 1953), and Miller (1919) . The area is covered by the Adirondack Sheet of the Geologic Map of New York (Fisher and others, 1962). The chief contributors to the knowl- edge of the Quaternary geology of the region are Woodworth (1905a, b), Fairchild (1919), Chapman ( 1937 ), and MacClintock and Stewart (1965). Mac- Clintock and Stewart present summaries of previous work. The Vermont Geological Survey has recently issued “The Surficial Geologic Map of Vermont,” scale 1:250,000, prepared by Stewart and MacClin- tock (1970; see also Stewart and MacClintock, 1969) . ACKNOWLEDGMENTS The study has had the benefit of assistance and advice from many individuals and institutions. The late Paul MacClintock of Princeton University intro- duced me to the Adirondack region and gave freely of his wide knowledge of the Quaternary geology of northeastern United States. Stimulating and pleas- urable discussions with the late John C. Goodlett of The Johns Hopkins University were cut short by his INTRODUCTION 3 death in 1968. S. A. Kirsch, R. D. Lawrence, and A. H. Strahler served as able field assistants. To‘ them and to many of my colleagues in the U.S. Geological Survey I am most grateful, especially to J. E. Hazel, W. A. Hobba, Jr., Carl Koteff, W. N. Newell, J. P. Owens, Meyer Rubin, and J. P. Schafer. A radio- carbon age determination was done in the laboratory of the U.S. Geological Survey. Thanks are also due to the New York State Geo- logical Survey, especially J. G. Broughton, D. W. Fisher, and Y. W. Isachsen; to the Vermont State Geological Survey, especially to C. G. Doll and D. P. Stewart; and to many others who contributed in various ways, including G. G. Connally, J. L. Craft, Klaus Flach, D. S. Fullerton, L. B. Gillett, J. H. Hartshorn, C. B. Hunt, W. H. Lyford, J. H. Moss, E. H. Muller, W. S. Newman, H. M. Raup, William Shelton, L. A. Sirkin, W. P. Wagner, D. R. Wiesnet, and M. G. Wolman. It is a great pleasure to acknowledge the assist- ance given by Canadian geologists and geographers, especially N. R. Gadd, E. P. Henderson, B. C. Mc- Donald, and V. K. Prest, of the Geological Survey of Canada; J. A. Elson and R. C. Zimmerman, of McGill University; and Pierre LaSalle, Quebec Dept. of Natural Resources. BEDROCK GEOLOGY AND TOPOGRAPHY Uplands and mountains of Precambrian crystal- line and metamorphic rocks are bordered by foothills of Paleozoic sedimentary rock that descend north— ward to the St. Lawrence Lowlands (Bostock, 1970) and eastward to the Champlain Valley. The north- east prong of the upland forms the St. Lawrence- Champlain divide and extends across the Interna- tional Boundary into southern Quebec Province. The Precambrian rocks form broad valleys and dome-shaped hills and low mountains commonly not more than 1,000 feet high, except for a few isolated peaks such as Lyon Mountain and Mount Whiteface. Granite gneiss and anorthosite each constitute about 30 percent of the area of uplands and mountains. These rocks are generally medium to very coarse grained and massive, with widely spaced joints. The rest of the area is underlain by other gneisses, meta- gabbro, and metasedimentary rocks (pl. 2; see also Fisher and others, 1962; Broughton and others, 1966). Sheeting is a common weathering feature of the gneissic rocks, and exfoliation domes are promi- nent topographic forms, as, for example, Silver Lake Mountains about 10 miles northwest of Ausable Forks (pl. 2). The Precambrian rocks tend to weather into large clasts, commonly of boulder size, and these in turn break up into fine gravel and sand. A foothills belt, Which‘i‘s a' féw‘to as much as 20 miles wide, borders the uplands and mountains on the northeast and north. It includes both isolated hills and dissected plateaus underlain by the Potsdam Sandstone (pl. 3), a quartz sandstone or arkose that is locally conglomeratic and that includes beds of shale. The Potsdam Sandstone forms beds a few inches to a few feet thick. Crossbedding and ripple marks are common. Dips are low, commonly less than 5°. Till derived from Potsdam Sandstone contains many clasts of pebble and cobble size, whereas that from anorthosite and granite gneiss contains a high proportion of sand-sized particles. The lowlands adjacent to the St. Lawrence River and Lake Champlain are formed on sedimentary rocks, largely of Ordovician age (pl. 3), commonly calcareous, and generally having a larger proportion of thin-bedded units than is found in much of the Potsdam Sandstone. In the area near Plattsburgh, dolostone, limestone, and dolomitic quartz sandstone are common (Fisher, 1968). Along English River in Quebec Province, and in some areas near Lake Cham- plain, the Potsdam Sandstone extends into the low- lands. DEPOSITS AND LANDFORMS ASSOCIATED WITH THE ICE SHEET GLACIAL DRIFT In the mountains the glacial deposits are, in large part, pebbly and sandy till that covers valley floors and lower mountain slopes. Many valleys also con- tain thick masses of sand and lesser amounts of gravel. Outcrops of bedrock are abundant along some of the larger streams and on upper slopes and moun- tain summits. Elsewhere, bedrock outcrops are scarce. (See Postel, 1952, pl. 1; Postel, Dodson, and Carswell, 1956; Denny, 1967, 1970.) On lower mountain slopes the drift may be more than 100 feet thick. W. A. Hobba, Jr. (written commun., 1967) re- ported that a water well on the upland about a mile south of Dannemora penetrated 130 feet of drift without reaching bedrock. No samples of drift were studied in the laboratory. The composition and texture of the till reflect the kind of bedrock. In areas underlain by crystalline rocks of Precambrian age (pl. 3) , the till is commonly a sandy loam or loamy sand containing pebbles and boulders of crystalline rocks and of Potsdam Sand- stone. In most exposures clasts constitute only a few 4 PLEISTOCENE GEOLOGY 01" THE NORTHEAST ADIRONDACK REGION percent by volume of the deposit. The scarcity of clasts may be the result of two factors: First, the Precambrian crystalline rocks weather to boulders and then to sand without the formation of abundant pebble-sized clasts; second, the valleys drain toward the ice edge and were dammed by it, so that the glacier overrode and incorporated sandy alluvial de- posits, either proglacial outwash or older alluvium. For example, in the valley of Alder Brook, a north- east-flowing tributary of the Saranac River, the till is a massive loamy sand containing only a few peb- bles and boulders (less than 1 percent by volume). In the foothills underlain by Potsdam Sandstone, the till is stonier and has a finer grained matrix (fig. 1) than the till in the mountains. Pebbles and boulders of Potsdam Sandstone may compose per- haps a fifth of the volume of the drift; the matrix is a sandy loam or loam. In the lowlands near Lake Champlain and in the St. Lawrence Valley, where the bedrock is chiefly dolostone, limestone, sandstone, and shale, largely of FIGURE 1.—Bouldery till at east end of the upland be- tween the Saranac and Salmon Rivers. Boulders are chief— ly of Potsdam Sandstone. Exposure in borrow pit about 2.5 miles north of Peasleeville. Ordovician age (pl. 3), the till is a pebbly loam to silt loam, has an alkaline reaction, and contains clasts of sedimentary rock that make up 5—10 percent of the total volume. The water-laid drift, largely sand, also reflects the nature of the adjacent till and bedrock. The deposits laid down in association with ice form kames, out- wash plains, and ice-channel fillings that are de- scribed in the sections on the various episodes of deglaciation. TILL LITHOLOGY The late A. Williams Postel identified in the field the stones or clasts in samples of till collected by him and the author (Denny and Postel, 1964) and placed each clast in one of several lithologic groups. The lower size limit of the stones collected was about 0.5 inches. Isopleths showing the percentage of clasts of Paleozoic sedimentary rock, granite gneiss, and anorthosite in the till (pls. 2, 3) trend roughly from northwest to southeast at right angles to the general direction of ice movement shown by directional fea- tures. Clasts of Paleozoic sedimentary rock, largely, and in many samples exclusively, Potsdam Sand- stone, are persistent. In most samples of till from the area underlain by granite gneiss, clasts of Potsdam Sandstone are more abundant than those of granite gneiss; clasts of Potsdam Sandstone appear to be more resistant to wear during glacial transport than those of the granite gneiss. It should be emphasized that the abundance (the number) of clasts in the till is not necessarily a true measure of the lithology of the till matrix. The clasts may constitute only a few percent of the total volume of till at a sample locality. Thus, the till as a whole might be 90 percent local material on a volume basis, whereas 70 or 80 percent by number of the clasts might be from a distant source. Some of the pebble counts appear anomalous. In a sample of till from the southwest corner of the map area (pl. 3), 50 percent of the clasts are sandstone. This sample comes from a broad lowland north of the high peaks where the drift cover is extensive. Perhaps beneath the drift there are small outliers of Potsdam Sandstone (Kemp, 1921, p. 65). On a high plateau of Potsdam Sandstone north of the mountains near Ellenburg (pls. 2, 3), the drift contains many clasts of granite gneiss and other crystalline rocks. The presence of these clasts 2—4 miles north of the nearest outcrop of crystalline rocks suggests either an inlier of Precambrian rock north of Ellenburg, now buried by drift, or the north- DEPOSITS AND LANDFORMS ASSOCIATED WITH THE ICE SHEET 5 ward movement of the clasts by glacier ice, mass movements, or running water. The till exposed in a roadcut about 2 miles west of Ellenburg contains about 25 percent clasts of Pre- cambrian crystalline rocks (pl. 2) ; the rest are clasts of Paleozoic sedimentary rocks. Twenty-one percent of the clasts of Precambrian rocks are of the Lyon Mountain Granite Gneiss, and 2 percent are Hawk- eye Granite Gneiss (Postel, 1952). A 16-inch boulder of the Hawkeye Granite Gneiss (not in the count) lay on the roadbank, and in the adjacent pasture are many boulders of Precambrian crystalline rock more than 4 feet in diameter and one about 6 feet in diameter. The till in an exposure about 5 miles south- southeast of Ellenburg and north of the Precam- brian-Paleozoic rock contact contains about 29 per- cent clasts of Precambrian crystalline rocks includ- ing 12 percent Lyon Mountain Gneiss and 12 percent Hawkeye Granite Gneiss. The clasts of Hawkeye Granite Gneiss at the locality south of Ellenburg could have come from outcrops of this granite gneiss on the crest of Ellenburg Mountain (fig. 2), about 3—4 miles to the south and about 600 feet above the elevation-1 of the sample site. The nearest source of Lyon Mountain Granite Gneiss is only about 1.5 miles south of the sample site. At the roadcut about 2 miles west of Ellenburg, the clasts of Hawkeye Granite Gneiss are about 7 miles north of their nearest source area on Ellenburg Mountain, and those of Lyon Mountain Granite Gneiss are about 4 miles from their nearest source to the southwest. At both localities, the drift that contains the clasts of Precambrian crystalline rocks is thin and rests on Potsdam Sandstone. The clasts do not have con- spicuous weathering rinds or shells. Some are stained throughout; others are essentially fresh. It is‘prob- able that the drift at the two localities was deposited by the last ice sheet, and that this ice picked up these clasts from a nearby source and moved them south- ward a short distance. Prior to this last movement, the clasts may have been carried northward by gla- cial ice from the mountains to the south (Cushing, 1899, p. 8). This transport could have taken place during the onset of the last glaciation, or it could have taken place during an earlier Wisconsin or pre— Wisconsin glacial age. The inferred northward move- ment of these clasts could also have been by mass movement or by catastrophic flood. DIRECTION OF ICE MOVEMENT Striae, drumlins, grooved drift, and till lithology indicate ice movement southwest across the north- east prong of the Adirondack Mountains (pls. 2, 3). In the lowlands south of the St. Lawrence River west of Clinton County, ice movement was south and southeast into the mountains, whereas in the low- lands west of Lake Champlain, ice movement was south up the Champlain Valley. The consistent trend of all the directional features suggests that they record ice movement during deglaciation when the ice edge was nearby. The few recessional moraines that can be traced for several miles—near Ellenburg Depot, north of Clinton Mills, and south of Miner Lake (pl. 1)—trend roughly at right angles to the direction of ice movement inferred from striae or grooved drift. The preservation of grooves in drift implies that the ice was not loaded with debris suf- ficient to conceal these streamlined forms when the ice had disappeared. The lithology of the stones in the till is consistent with the direction of ice movement recorded by the directional features, except for the samples from the two small areas near Ellenburg. In general, the per- centage of clasts of Paleozoic sedimentary rock (pl. 3) decreases southwestward and also as the altitude of the sample site increases (compare Lyon, Terry, and Whiteface Mountains). The trend of the iso- pleths suggests movement of ice southwest up the Saranac River valley. The pebble counts do not sug- gest southward ice movement across the Lyon Moun- tain—Johnson Mountain highland north of the river, nor is there evidence in the counts for a northward ice movement from the high peaks region (Craft, 1969) south of the area shown on plate 2. Isopleths for anorthosite are consistent with a southerly ice movement. Small drumlins are abundant in the northeast corner of Clinton County near Champlain (Denny, 1970), where the bedrock is chiefly dolostone, lime- stone, sandstone, and shale, and where the till form- ing the streamlined features is a pebbly silt loam or loam. The azimuth of the long axis of 50 of the drumlins near Champlain ranges from N. 9° W. to N. 14° E. The estimated mean value is N. 6° E. Glacial grooved or fluted drift consisting of long narrow ridges and swales is found in small areas near Miner Lake and on the crest of the ridge be- tween Saranac and Peasleeville, where the bedrock is Potsdam Sandstone. Striae adjacent to grooved drift trend in approximately the same southwesterly di- rection. The individual ridge does not have a bedrock knob at its northeast (up-ice) end as do the ridges described by McDonald from the Appalachian region of southeastern Quebec Province (McDonald, 1966, 1967). 6 PLEISTOCENE GEOLOGY OF THE NORTHEAST ADIRONDACK REGION 74 °OO’ Cen r’ _ /— \—\/l I 44°rx\7l"‘/ "\‘4 /\/ / /\‘_\/\ /.§// ,\/\/I V, -+\, 45*'I,‘/\:\:\ ‘/‘ I \/\_l:|,\]\/\/\/l:\l,\~./ / \—I _\ /\\/ I -’\ / __ \/ / \xlT—\/\/l?_\/>I ‘ I \ amp/V“ /\\/,/ , fl,» \,Lyon Mountai / , ., . \ /\\\‘/ / 73°45’ , Ellen urg O 5 MILES |_;1__|__1_l E X P LA N ATI O N , I E . . , . . . l“ Glacial and alluvial 3: deposits 0 % ' . Potsdam Sandstone 0 ~ / x I l (s V 4 Z V. e” 9 < 9.5"— 2??% E Chiefly Granite Metagabbro Meta- ‘11 Hawkeye g'neiss and sedimentary 3 Granite Includes Lyon amphibolite rocks 0 Gneiss Mountain Granite Chiefly Gren— UJ Gneiss Mlle Series E x2 Sample site Value is percentage of clasts of Hawkeye Granite Gneiss larger than about 0.5 in. in diameter NARY BRIAN MORAINES IN THE SARANAC VALLEY At Cadyville on the Saranac River, a belt of re- cessional moraine consisting of linear ridges and knolls of till and kames of sand and gravel crosses the valley from north to south (pl. 1, from ice-front position 4b east to West Beekmantown). The belt ranges in width from about 1 mile to 2.5 miles. East of Cadyville, the river descends 300 feet in a rock- walled gorge. The wide belt of moraine crosses the valley on the tread of this giant bedrock step and was perhaps localized in part by it. The moraine was built by a small southwest-moving ice tongue from a large ice mass in the Champlain Valley. The moraine is a part of Taylor’s DeKalb moraine (Taylor, 1924, p. 665). The linear till ridges (pl. 4) range in height from a few feet to about 60 feet above the adjacent swales and in length from several hundred feet to a quarter of a mile. The ridges are curving or sinuous in plan and are spaced commonly 200—500 feet apart from crest to crest. The crests are rounded and boulder covered. The swales between the ridges are ZOO—1,000 feet wide and are floored with sand and a few bould- ers. These linear features are neither as regular nor as continuous as the 20-foot contours of the map indicate. The ridges are composed of pebbly, sandy till that includes small masses of stratified sand and gravel. The stones in the till ridges are largely Paleozoic sedimentary rock; pebbles of Precambrian rock gen- erally make up less than 5 percent (pl. 3) of the total clasts. Bedrock outcrops are scarce in the area of the moraine. The linear ridges in the moraine south of the river form arcs bowed out in an easterly direction, where- as the moraine as a whole is bowed in the opposite direction. The explanation for these opposing trends may involve stagnation and minor oscillations of the ice near the front. The ice tongue first advanced to its southwest limit about a mile west of Elsinore and stagnated. Then active ice east of Elsinore moved southwest diagonally up the valley wall south of the dead ice mass to form a narrow tongue of ice about a mile wide in the area between Duquette Road and the areas of bare rock south of Hardscrabble Road. The arcuate ridges bowed out to the east are inter- FIGURE 2.—Percenta.ge of clasts of Hawkeye Granite Gneiss in till in the Ellenburg—Lyon Mountain area, Churubusco and Lyon Mountain 15-minute quadrangles. Geology gen- eralized from Fisher and others (1962). Percentage of clasts estimated by rapid method described by Denny and Pastel (1964). DEPOSITS AND LANDFORMS ASSOCIATED WITH THE ICE SHEET 7 preted to be lateral moraines on the northwest side of this small tongue of ice. North of the Saranac River the moraine includes prominent linear till ridges, abandoned stream chan- nels eroded in drift, and bodies of sand and gravel believed to be ice-marginal kames. Most of the sand and gravel in the moraine prob- ably came from the ice sheet, although some of the material may have come from the erosion of chan— nels in the drift. Deposition was largely by melt- water streams flowing southwest and south along the ice margin. In addition, water ponded north of the Saranac Valley-Smith Wood Brook divide may have spilled south across the divide and down the north slope of the Saranac Valley where it eroded the drift-walled channels, now largely abandoned. As the edge of the ice sheet retreated northeast, the place Where the water spilled south also moved east. Thus, the abandoned channels on the north side of the Saranac Valley are progressively younger northeast- ward. The linear ridges in the moraine suggest radially inward shrinkage of the ice tongue in the valley. Water, ponded by ice, overflowed along the southern margin of the ice tongue around the eastern end of the hill between the Saranac and Salmon River (Clin- ton County) valleys (pl. 1) and eroded channels both in drift and in bedrock; one of the latter is cut about 60 feet into nearly horizontal beds of the Potsdam Sandstone. The water also washed away most of the drift cover from bedrock areas as large as 0.5 mile in diameter. As the edge of the ice tongue retreated to the northeast, channels at lower and lower alti— tudes were utilized by overflow from the Saranac Valley. The materials eroded by the water flowing along the south side of the ice tongue were deposited just southeast of the areas of bare bedrock as boul- dery glaciofluvial deposits on the north slope of the Salmon River valley. The eastern edge of the belt of recessional moraine (pl. 4) is marked by a low ridge that extends south from Cadyville for about 0.5 mile and north of the river to a sand and gravel kame along State Route 374. Farther north, another ridge follows the same course for a short distance before turning northeast. An exposure in the ridge on the north side of State Route 3 showed till resting on sand, suggesting a minor oscillation of the ice front. The kame north of Cadyville is probably an ice-marginal deposit from south-flowing water in a channel that crossed the site of the present Saranac River and continued south, entering Lake Vermont, where it built a small delta (Qs, pl. 4), about 1.5 miles south of Woods Mills. The broad open valley of the Saranac River be- tween Cadyville and Saranac contains no traces of an ice-dammed lake, yet the moraine near Cadyville demonstrates that an ice tongue dammed the valley. Either the moraine-dammed lake was too short lived to leave conspicuous evidence of its presence, or the broad valley upstream from the moraine was filled with stagnant ice. It is not known whether the mo- raine near Cadyville marks a stationary ice-front or a readvance of the ice up the Saranac Valley from a position in the Champlain Valley. MORAINES IN THE GREAT CHAZY VALLEY An 8-mile-long belt of recessional moraine crosses the North Branch Great Chazy River near Ellenburg Depot (ice-front position 8, pl. 1), and shorter mo- raine segments occur to the northwest near the Canadian border north of Clinton Mills and to the southeast in the Great Chazy River valley south of Miner Lake. If these three segments are essentially contemporaneous, as seems probable, they show that the ice front crossed the valley of the Great Chazy River and the North Branch with only a slight bulge upvalley to the southwest. The morainal belt attains its greatest dimensions at Ellenburg Depot where it is bisected by the North Branch Great Chazy River (pl. 5; Taylor 1924, fig. 12; MacClintock and Stewart, 1965, p. 58 and pl. 13) South of the river, a north-trending ridge ranging from about 80 to 100 feet in height and from about 1,000 to 1,600 feet in width is composed chiefly of sand and gravel. West of the ridge, the valley walls are smooth or gently rolling, but to the east they have a pronounced local relief of knolls- and ridges. In exposures in a borrow pit near the river (point A, pl. 5), deformed beds of gravel and masses of till are separated by essentially vertical faults from southwest-dipping beds of coarse sand and pebble gravel that grade into horizontally bedded sand near the west side of the ridge (fig. 3). These beds prob- ably are part of a delta built by melt-water streams from the ice sheet on the east side of a small ice- marginal lake that was west of the moraine and was dammed by it. In a second borrow pit (at point B, pl. 5), about 4 feet of till overlies 10—15 feet of thin-bedded medium sand. At the base of the exposure is a second till, largely a structureless mixture of sand and silt con- taining a few striated stones. Between the lower till and the bedded sand is a 10-foot horizon of mixed till and sand consisting of tilted and faulted beds of sand and masses of till. In places, the till masses are PLEISTOCENE GEOLOGY OF WA ADIRONDACK REGION .m .3 .V £52” .3an Manama .30: £9 33.3: 3 Eamomxfl i=2. 3533.793 :25 a no fan 0.3 m—umnaaus .539me HO ~50 3:95 553 an ".53 33.3 hnflnafinon 9:: 35.5 35 ~25?» 01509 and "Eda 3.38 «a mug MummmmcémoB Scum 33am 2&5.-wa .3 ceaananwm .8303 no 98 3%: 5.3: a.» Ea mc 33%: can $32» me 35.— wagomon demon Managua .39: 2:98:— 23 mo #39 a awn—959 now—own muonvlé ESE DEPOSITS AND LANDFORMS ASSOCIATED WITH THE ICE SHEET 9 vertical plates 2—3 feet thick and at least 10 feet long that lie against beds of sand cut by many closely spaced high-angle faults. It appears that the ice edge oscillated back and forth across the floor of the ice- dammed lake, depositing a relatively stone-free till, deforming its own delta deposits, and laying till on top of them. At its north and south ends, the moraine near Ellenburg Depot rises to an altitude of about 1,100 feet or about 100 feet above the crest of the moraine along the river, suggesting that when the ice- dammed lake overflowed, it was at least 200 feet deep. If so, the sands near the top of the moraine at the river may have been deposited at a depth of more than 100 feet below the surface of the lake. There is no other evidence for an ice-dammed lake in the headwaters of the North Branch beyond that presented by the moraine itself. South of Dannemora Crossing (pl. 5) the moraine ascends the south slope of the valley of the North Branch. In sand pits along Plank Road, deformed delta deposits of sand and gravel pass upward into lake deposits of thin-bedded fine sand containing a few boulders. West of the pits, the slope of the ridge is covered with boulders. Where the moraine crosses Old Military Turnpike (fig. 4), it is a small ridge only 6 feet high and about 80 feet wide. North of Ellenburg Depot, the moraine is a broad low ridge of till 5—20 feet high and 500—1,000 feet wide. Near its northern end, it becomes a narrow steep-sided boulder-covered ridge as much as 50 feet high and 400 feet wide (pl. 6). The short segment of recessional moraine north of Clinton Mills extends east-west across the uplands FIGURE 4.—Along the Old Military Turnpike the moraine near Ellenburg Depot is a boulder-covered ridge about 6 feet high and 80 feet wide. Photograph is looking north across the Turnpike (pl. 5) . for about 1.5 miles at altitudes ranging from 1,050 to 1,100 feet (pl. 1). It is a low sinuous ridge (fig. 5), 5—20 feet high and about 100—300 feet wide. Side slopes are steep, the surface is bouldery, and the material appears to be till. West of the well-defined segment of moraine for a distance of about 3 miles are other small east-trending linear ridges (not shown in fig. 5) that may also be recessional mo- raines, perhaps modified by wave action. MacClin- tock and Stewart (1965, pl. 1B) mapped some of these as beach and nearshore deposits. The other small segment of recessional moraine south of Miner Lake is a series of long narrow ridges on the south slope of the valley of the Great Chazy River (pl. 1). It extends from a point near the Old Military Turnpike (fig. 6), west for about 4.5 miles to Alder Bend Road (Woodworth, 1905a, p. 11). The individual ridges range in length from about 300 to 4,000 feet, in width from about 150 to 600 feet, and in height from about 2 to 30 feet. The ridges are boulder covered; they are probably com- posed of sandy till. Just east and west of the Great Chazy River are sand and gravel kames. SPILLWAYS ON THE ST. LAWRENCE-CHAMPLAIN DIVIDE The plateau on the Potsdam Sandstone northeast of the mountains (pls. 1, 3) forms the divide between the St. Lawrence and Champlain Lowlands and ex- tends as far as Covey Hill, Quebec Province. North and east of Covey Hill, the surface drops steeply to the floor of the St. Lawrence Lowlands, the descent being nearly 800 feet in about 2 miles. Abandoned stream channels or cols, which served as spillways for ice-dammed lakes, cross the divide (pl. 1) . The abandoned stream channels south of Churu- busco in northwestern Clinton County, at altitudes ranging from about 1,200 to 1,300 feet, are generally not more than 100 feet wide and 10—20 feet deep; one is about 300 feet wide. (See Churubusco 71/2- minute quadrangle.) Their east ends “hang,” and their floors slope west, indicating that they carried water from east to west, perhaps into ice-marginal streams that cut some of the abandoned stream chan- nels west of Churubusco and south of Chateaugay (pl. 1; see also MacClintock and Stewart, 1965, fig. 21). About 1.5 miles north of Clinton Mills (3 miles east of Churubusco) is a large channel about 500 feet wide and 10 feet deep at an altitude of about 1,085 feet (fig. 5). This channel hangs at both ends and could have carried water either east or west. No bare-rock areas are nearby. 10 PLEISTOCENE GEOLOGY OF THE NORTHEAST ADIRONDACK REGION I OH 5 HUPNTIFEDOE-CQ I 3’62“?” 45000! C ' J-"CLINTW v ’ fz‘ ‘ ' 027 r ~- ' - ' 0 V2 1 MILE Base from U.S. Geological Survey L I l 1 I l I I I l I Geology by C. S. Denny, 1967 Churubusco and Ellenburg Depot 0 .5 1 KILoMETER 1964 |_L__;I_1_I_I_J__I_I_J CONTOUR INTERVAL 10 FEET DATUM IS MEAN SEA LEVEL EXPLANATION ;___/ x. . . .. r'fi Recessional moraine Narrow boulder-covered ridge of till (.9) C01 FIGURE 5.—Recessional moraine north of Clinton Mills and abandoned stream channel or col, altitude about 1,085 feet, on the St. Lawrence Valley—Champlain Valley divide. The abandoned stream channel west of Covey Hill, described more than 100 years ago (Emmons, 1842; Quebec, just north of the Canadian border (pl. 1; see Woodworth, 1905a, Fairchild, 1912; and Goldthwait, also Chateaugay sheet, Quebec-New York, west half 1913. The following description is based largely on [31 H/4 WEST] scale 1:50,000, published by the MacClintock and Terasmae, 1960). The channel is at Canadian Dept. Mines and Tech. Surveys) was first an altitude of about 1,010 feet and is about 90 feet DEPOSITS AND LANDFORMS ASSOCIATED WITH THE ICE SHEET below the high point on the upland about 2 miles to the northeast. It is about half a mile wide and about a mile long; the floor of this channel is bedrock, which in part is mantled by a 6-foot layer of peat. At the east end of the channel is a box canyon, 75 to 100 feet deep and about a mile long, cut in the Pots- dam Sandstone. An abandoned waterfall at the head of the canyon is more than 125 feet high; about 75 feet of this precipice is now beneath the surface of a small pond. Half a mile from the dry fall is a second stagnant pool, The Gulf (pl. 6). The channel near Covey Hill carried the outflow of glacial Lake Iroquois (Coleman, 1937) into the Champlain basin, supplying much of the water that washed the drift from large areas of bedrock. AREAS OF BARE ROCK At the northeast end of the Adirondack uplands in northern Clinton County are large areas of bare rock (Potsdam Sandstone). The largest, Flat Rock, near Altona (Woodworth, 1905a, p. 16—24), is about 5 miles long, 2.5 miles wide (pl. 1), and presents a striking appearance on aerial photographs. The “flat rocks” are either bare rock or bedrock that supports an open forest with a luxurious under- growth of low shrubs which grow in a thick mat of organic matter resting directly on bedrock (fig. 7). There is little inorganic mineral matter in the soil. White pine, pitch pine, red oak, and other species 7.3“42’30” 11 common to the northern hardwood forest compose the open forest. Jack pine is also abundant, although it is rarely found growing in the adjacent drift- covered areas. Ripple marks and curving joints that are the out- cropping edges of foreset beds are a characteristic minor surface feature. The bedrock surface is slightly weathered, and solution pits are common (fig. 8). At only one locality was the surface pol- ished, presumably by glacier ice. The large areas of abundant outcrop have little local relief beyond that of risers a few inches to 10 feet high between the essentially horizontal bedding planes (fig. 9). The flat surfaces commonly extend for hundreds of feet between risers. The sandstone may be broken by joints as much as l-foot wide and several feet deep. In some places steep-sided bedrock masses rise 10—20 feet above the surrounding flat. About 1 mile northwest of Cobblestone Hill (pl. 7), a steep-sided ridge more than 50 feet high is sur- rounded by a flat bare-rock surface on three sides. The ridge has a closely packed mantle of angular sandstone blocks ranging from 1 to 15 feet in diam- eter; 2—6 feet is the most common range in size. The mantle of angular blocks extends down to a sharp contact with the bare rock. Bedrock crops out on the crest of the ridge near its north end. The large areas of bare rock are on divides be- tween the larger streams (pl. 1) and are restricted 73°40 Base from US. Geological Survey 1 1:24,000 Jericho 1966 l '1 MILE Geology by C. S. Denny, 1967 I 1 KILOMETER I_J_l_J_l_l_l_.L_l_L_J CONTOUR |NTERVAL 20 FEET DATU‘M IS MEAN SEA LEVEL EXPLANATION o. I. .; ~'.--'-" _.. o . .0..'Qk.o a. . “fl . I'. a... Kame Sand and gravel QUATER- Narrow boulder-covered ridge of till (.9) NARY —> Abandoned stream channel or col Glacial striation Point of obsmat'llon at tip of arrow FIGURE 6.—Moraine south of Miner Lake. 12 PLEISTOCENE GEOLOGY OF THE NORTHEAST ADIRONDACK REGION FIGURE 7.—Flat Rock, near Altona. Potsdam Sandstone with patches of low shrubs and scattered trees. View looking West just east of Miner Lake. largely to a belt about 3 miles Wide, extending from the Canadian border near Cannon Corners southeast nearly as far as West Chazy, a distance of about 20 miles. The belt ranges in altitude from about 1,000 feet to about 600 feet. FLAT ROCK, NEAR ALTONA Flat Rock, southeast of Altona (pl. 1), is about 5 miles long and 2.5 miles Wide. The bedrock surface slopes to the north and east from an altitude of more than 1,000 feet down to an altitude of about 600 feet. There are several narrow rock-walled gorges in the southeastern part of the bare-rock area, such as Bear Hollow (pl. 7). The bare-rock area is drained by Cold Brook, which has its source in the Dead Sea, a small rock-walled pond shown in the northwest corner of plate 7. The Dead Sea is in a former plunge FIGURE 8.—Solution pit in Potsdam Sandstone in bare-rock area south of English River. pool, part of an abandoned stream channel (Wood- worth, 1905a, pls. 7, 8, 9, and 10). The border of Flat Rock is generally sharp but highly irregular. The surface of bare rock passes under the surround- ing drift or other surficial cover without a prominent topographic break. Flat Rock is ringed on the north and east by seg- ments of the beaches of glacial Lake Vermont (Chap- man, 1937). The highest stand of Lake Vermont in this area, the Fort Ann stage, overlapped the exist ing outer limit of the area of bare rock by as much as 0.5 mile and stood almost 100 feet above the low- est elevation of bare rock (pl. 7 ). Bouldery material at the edge of Flat Rock was presumably in large part washed off the bare-rock area and redeposited by ice-marginal streams along the border. The total volume of such material is small. The gravel that forms the well-developed bouldery beach ridges on the east slope of Cobble- stone Hill could have been reworked from material washed from the bare-rock area to the northwest. South of Cobblestone Hill along the east edge of the bare-rock area (pl. 7) are two small deposits of bouldery material. Along the stream that crosses the southern deposit, 10—20 feet of boulder gravel is exposed. Bedrock crops out nearby, and the total volume of gravel is small. A borrow pit near Flat Rock (fig. 10) excavated in boulder gravel shows many clasts ranging from 1 to 5 feet in longest diameter. Southeast of Cobblestone Hill are bouldery depos- its that may be recessional moraines composed of material washed from the bare-rock area, deposited in Lake Vermont at the edge of the ice, and subse- quently molded into moraine during a slight advance of the ice edge. Woodworth made the suggestion that the abundance of gravel beaches and the scarcity of bedrock outcrops in the valley of the Little Chazy River west and southwest of West Chazy might be due to the presence of “unusually thick deposits [that] are probably to be attributed primarily to the wash from the flat rock districts” (Woodworth, 1905, p. 20). However, even assuming that all the materials described above were washed from Flat Rock, the total volume of material is so small that one must conclude that Flat Rock probably never had a thick drift cover. SOUTH OF ENGLISH RIVER The area of bare rock on the divide between Eng- lish River and North Branch Great Chazy River (“Blackman’s Rock,” Woodworth, 1905a, pl. 5) is DEPOSITS AND LANDFORMS ASSOCIATED WITH THE ICE about 3 miles long and about 1 mile wide at its widest point (pl. 6; fig. 11). The bare-rock surface slopes northeast and is bordered by recessional moraines trending at right angles to the divide. The moraines record the presence of a southeast-trending ice edge that dammed the English River, causing it to over- flow into the valley of the North Branch. Such over- flow progressively removed the drift from the slopes of the divide as the ice front retreated to the north- east. Figure 11 shows the stages of the northeasterly retreat of the ice front across the bare-rock area south of English River (ice-front positions 10-14) . Material washed from the higher or southwestern part of this area of bare rock was deposited as gravel and sand adjacent to the recessional moraines shown near the southern edge of plate 6. Material washed from the lower or northeast end was deposited in Lake Vermont where it was reworked by lake water into beaches and spits. A mass of gravel and sand about 0.8 mile south of Cannon Corners (pl. 6) extends from the bare-rock area east of the highest beaches of the glacial lake. The surface of the deposit is mantled with boulders as much as 6 feet in diameter and ranges in altitude from about 790 feet near the flat rocks down to about 710 feet along the highest stand of the glacial lake. The boulders could have been concentrated on the surface by the washing of underlying till, but the FIGURE 9.—Vertical joint face between two bedding-plane sur- faces on the Potsdam Sandstone south of English River (1°C. A, pl. 6). Block of sandstone at base of joint face at left. Note shovel for scale. 13 SHEET FIGURE 10.—Boulder gravel, probably debris washed from an area of bare rock. Exposure in borrow pit about 1 mile south of Altona. presence of shallow ice-block holes (not shown in pl. 6; see Denny, 1970) suggests that much of the surface is underlain by water-laid material deposited in association with wasting glacial ice. At one place near the southeast edge of the bare- rock area south of English River (loc. A, pl. 6), the surface of the riser and adjacent tread is highly polished but not striated. Presumably, this is glacial polish that elsewhere has been removed by weather- ing. The polished surface extends for perhaps 50 feet along the edge of the bare rock. To either side of the polished surface, the riser and the tread are essen- tially smooth or slightly pitted. There are no sandstone blocks at the base of the polished riser, but one or two blocks lie on the adja- cent flat at the base of the riser to either side. None of the surfaces of these blocks is polished; all appear slightly weathered. NORTH OF ENGLISH RIVER The area of bare rock (“Stafford’s Rock,” Wood- worth, 1905a, pl. 5) north of English River is about 3 miles long and 1 mile wide (pl. 6) and is much like that south of the river. At the north end, however, are several bedrock knolls that range from about 200 to 1,000 feet in diameter and that rise 5—20 feet above the surrounding flat bedrock surface. These small steep-sided residual knobs probably were carved by streams that flowed east from the gorge near Covey Hill, Quebec (The Gulf), roughly along ice-front position 13 (fig. 11). The streams eroded 14 PLEISTOCENE GEOLOGY OF THE NORTHEAST ADIRONDACK REGION 73'50‘ 45' 40' 73'35’ l I I Covey me oCovey Hill R o. 1 ¢§13,14, 5 QUEBEC CiNAD—A — — —— " - _— 45-00' 3 '73“ , _ NEW YORK _— —_ UNITED STATES #3 )4 55' - Miner Lake River 44-50' _ Base from U.$. Geological Survey 0 1 2 3 4 SMILES 1262,500 Churubusco and Mooers I l I I 1964 and 1956 0 1 i 3 4 5 KILOMETERS EXPLANATION -4539 a) Recessions] moraine Large area of bare rock Abandoned stream channel showing direction of flow WM W Assumed connection between , recessional 'moraines nghest stand of Fort Ann stage Number mum. ice-front may," of Lake Vermont in sequence; 8 is oldest FIGURE 11.—Map of ice-front positions in area from Covey Hill, Quebec, to Flat Rock, showing recessional moraines, areas of bare rock, abandoned stream channels, and highest stand of Fort Ann stage of Lake Vermont. HISTORY OF a bedrock ridge (cuesta) to form the knolls and cleaned off the drift to expose the adjacent bedrock. A prominent recessional moraine about 2 miles long and as much as 0.4 mile wide lies east of the bare-rock area (ice-front positions 13 and 14, fig. 11), and several small narrow morainal ridges lie on the western part of the bare-rock area (fig. 12) . At the south end, just north of White Road, two moraines unite to form a prong whose surface is ex- ceedingly bouldery; blocks 6—10 feet in diameter are common. The two moraines enclose a triangular- shaped swamp floored with organic material that appears to rest directly on bedrock (pl. 6) . ORIGIN The presence of large areas of bare rock (Pots- dam Sandstone) or rock thinly veneered with soil and vegetation indicates that in some areas, condi- tions at or near the ice edge were such that drift was not deposited, or once deposited, has since been re- moved. Because most of the large areas of bare rock are found in a limited region, whereas the Potsdam Sandstone is a widespread formation, it is clear that the bare surfaces do not owe their origin to some characteristic of the bedrock. I believe that the bare-rock areas were swept clean by the outflow of glacial lakes held between the ice front and the northeast slope of the uplands. Water flowing in spillways along the ice edge removed the drift that overlay the bedrock. This is the explana- tion presented by Woodworth, following a suggestion by G. K. Gilbert (Woodworth, 1905a, p. 16—24) , and my observations support their thesis. Only a few small masses of bouldery material have been found adjacent to the bare-rock areas, and it is logical to assume that such areas probably never had a thick drift cover. The cleaning of the drift cover from large areas of bedrock was accomplished by the sudden, perhaps catastrophic, draining of ice-dammed lakes. In the St. Lawrence Lowlands northwest of Covey Hill, glacial Lake Iroquois (Coleman, 1937) stood at an altitude of nearly 1,100 feet, its outlet near Rome, N.Y., southwest of the Adirondacks (fig. 14) . A slight rise in lake level due either to uplift of the Rome outlet or to a slight northeasterly retreat of the ice front could have caused Lake Iroquois to over- flow into the Champlain basin, supplying the ice- marginal streams with large volumes of sediment- free water. At first, the outflow from Lake Iroquois may have been by way of cols or abandoned channels just south DEGLACIATION 15 FIGURE 12.—Boulder—covered recessional moraine, at right, and turf-covered pasture and pond on bedrock at left. View north from White Road at a point about 1.2 miles west of Cannon Corners (pl. 6). of the Canadian border (fig. 11). Later, the outflow came by way of the channel near Covey Hill, Quebec. By this means, the volume and perhaps the velocity of the water that flowed along the ice front were in- creased to such an extent that over large areas all the debris on the bedrock was removed. As the ice front retreated (ice-front positions 9 to 15, fig. 11) , the ice-marginal streams—the spillways of Wood- worth (1905a)—migrated north and east and cleaned off large areas of bare rock. HISTORY OF DEGLACIATION In the mountains, deglaciation involved stagnation of the marginal zone of the ice sheet, whereas in the St. Lawrence and Champlain Lowlands the ice sheet maintained an active front. Presumably this con- trast in mode of deglaciation is related to the topo- graphic contrast between the two areas. In the low- lands, as the marginal zone of the ice-sheet thinned, it remained active and built small moraines. In the mountains, the thinned marginal zone became sepa- rated from the actively moving ice. Melt water built kame terraces and outwash plains of debris, prob- ably obtained from both stagnant and active ice bodies. Drainage was toward the ice front, and ice- marginal lakes were formed. The Ausable, Salmon (Clinton County), and Saranac Rivers (pl. 1) flow northeast into the Champlain Valley, that is, in the direction of ice retreat. As deglaciation proceeded in a generally northeast direction, the headwaters of the rivers were uncovered first. The streams were ponded at the ice front. These ice-dammed lakes, fed 16 PLEISTOCENE GEOLOGY OF THE NORTHEAST ADIRONDACK REGION by melt water and by runoff, overflowed across di- vides into adjacent valleys. Ultimately, the overflow reached glacial Lake Vermont, which itself length- ened northward as the ice edge retreated down the Champlain Valley. When the Adirondack uplands were largely un- covered but the ice sheet still rested on the St. Lawrence-Champlain divide near the International Boundary, ice-dammed lakes were present on both sides of the divide. At first, the lake on the northwest side overflowed to the southeast. Then the flow was reversed and small ice-dammed lakes southeast of the divide flowed into the St. Lawrence Lowlands. Later, the glacial lake west of the divide, Lake Iro- quois (Coleman, 1937), again overflowed to the east, cutting the gorge near Covey Hill, Quebec, cleaning ofi‘ large areas of bedrock, and emptying ultimately into Lake Vermont. Finally the two lakes merged. Only in the Champlain Valley are ice-marginal features continuous enough for one to draw, with reasonable certainty, the form of the ice front at any one time. Elsewhere, these features are widely sepa- rated, and their correlation one with another is based only on inference. As previously mentioned, the recessional moraines in the Great Chazy and Saranac valleys suggest that the ice sheet in the foothills and adjacent parts of the Champlain Valley had a straight to gently curved front, influenced in only a small way by the local topography. Narrow tongues of active ice apparently did not project far up the major east-draining valleys, at least not at the time when the edge of the ice lobe in the Champlain Valley lay along the east edge of the Adirondack uplands. If we assume that throughout deglaciation the ice edge was straight or gently curved, it is possible to reconstruct a logical picture of deglaciation of the northeast Adirondack region from the scattered patches of recessional moraine and other ice-mar- ginal features that remain. Plate 1 shows several inferred positions of the ice front beginning with number 1 in the southwest corner of the region and progressing in a general northeast direction toward Covey Hill, Quebec. The dating of these ice-front positions is discussed in the final section of this paper. The history of the various features shown on the map is summarized in table 1 and briefly described in the following pages. LOON LAKE EPISODE (1) When the edge of the southwest-moving ice sheet was at ice-front position 1 (pl. 1), melt-water streams flowed southwest, laying down thick masses of glaciofluvial deposits. This is the oldest event rec- ognized in the region. The ice-contact slope on the northeast side of the recessional moraine is as much as 200 feet high. Drainage was to the west by way of the St. Regis River. OWLS HEAD-REDFORD STAND (2) In the St. Lawrence River drainage, Franklin County, the most prominent feature of the Owls Head-Redford stand is the south-sloping outwash fan at Owls Head (2a, Watertown moraine of Taylor, 1924, p. 660). The fan is apparently com- posed of sand and gravel as much as 200 feet thick and rests on a bedrock sill, north of which the Sal- mon River drops 300 feet in less than a mile. The ice front was on the north side of the fan and discharged melt water to the south. The water ultimately es- caped west by way of the col between Humbug and Titusville Mountains. The Salmon River valley south of Owls Head probably still contained masses of stagnant ice. Southeast of the St. Lawrence-Champlain divide, the ice sheet blocked the Saranac Valley near Red- ford (2b) , and the river was diverted south into the headwaters of the Salmon River (Clinton County). The Salmon River valley also was blocked by ice, and the drainage went south across another low divide into a south-flowing tributary of the Ausable River. The position of the ice dam in the Salmon River valley is marked by a small recessional moraine con- sisting of steep-sided and boulder-covered knolls (fig. 13) that merge to the west with a small pitted outwash plain of sand and gravel, whose upper sur- face slopes to the south. The recessional moraines on the east slope of the uplands northwest of Clin- tonville are cut by the abandoned channels of streams that flowed south along the west side of a tongue of ice in the valley of the Little Ausable River (pl. 1), perhaps to join the Ausable River at the delta of the Coveville stage near Clintonville. The diversion of the Saranac River near Redford, south to the Ausable River near Ausable Forks, is marked by several abandoned stream channels. The older channels, formed when the ice front was slightly southwest of position 2b, are on the Saranac River—Ausable River divide southeast of Redford. They range in altitude from about 1,200 to 1,230 feet and carried Saranac River water into the headwaters of Blake Brook (pl. 1) and thence south to the Aus- able River. N ortheasterly retreat of the ice edge to position 2b opened other channels at lower altitudes, 1,130—1,200 feet, that conveyed water from the Sara- HISTORY OF DE-GLACIAT‘ION nac Valley into headwaters of the Salmon River and thence south to the Ausable River valley. North of the village of Black Brook, a flat-floored valley 800 feet wide and 3—4 miles long ends near the village in a series of terraces that are composed of boulder and cobble gravel and that extend south for about 1.5 miles to the Ausable River. The terraces decline in altitude from about 1,060 feet to 900 feet, and the material becomes finer grained. Such extensive boul- dery deposits are not common in this region; pre- sumably they were deposited as an alluvial fan or ice—marginal kame on the north side of the Ausable River valley. The Ausable River has since incised its bed 100—200 feet. The delta west of Clintonville built by the Ausable River in the Coveville stage of glacial Lake Vermont has its top at an altitude of about 670 feet. It is possible but by no means certain that water coming down the stream channel north of Black Brook contributed directly to the building of this delta. TROUT RIVER-MOFFITSVILLE STAND (3) West of the St. Lawrence-Champlain divide, a re- cessional moraine and south-draining ice-marginal channels near the mouth of the Trout River valley mark the Trout River-Moflitsville stand (3a). To the east, the ice front is marked by recessional moraines in the valley surrounding Upper Chateaugay Lake. In the Champlain basin, the ice-front (3b) is not well defined. Recessional moraines and other ice- marginal features are scarce, except for the small moraine, composed of till, sand and gravel, between Harkness and Keeseville. The overflow from the Saranac Valley abandoned its southerly course to the Ausable River and turned east down the Salmon River valley as an ice-mar- ginal stream, depositing gravel and sand and escap- ing around the east end of Terry Mountain. As the ice in the Salmon River valley gradually shrank, the late-glacial Saranac River cut down nearly 100 feet into the recessional moraine 4 miles southwest of Peasleeville and spread gravel on top of some of the sand deposited earlier. Near the moraine, the gravel- covered flood plain of the Salmon River slopes east at about 50 feet per mile and has conspicuous bars and swales. About 2 miles east of the moraine, the valley floor descends 100 feet in a distance of about a quarter of a mile. Farther east, the river meanders on a sandy and silty flood plain at a gradient of less than 20 feet per mile. The relations suggest that near the moraine the floor of the valley is essentially that of the flood plain of the late-glacial Saranac River 17 FIGURE 13.—Boulder—covered knolls form small recessional moraine in Salmon River valley (Clinton County) about 4 miles SOUthWBSt 0f Peaslee- ville. View to north and northeast. Ice stood on east Side of moraine and discharged melt water to the west. 18 PLEISTOCENE GEOLOGY OF THE NORTHEAST ADIRONDACK REGION TABLE 1.——-H’istory of deglaciation of the northeast Adirondack region, New York Episode or stand Location of stand or area deglaciated Age of episode (numbers refer to ice- during episode (points near ice front or stand front position shown listed in order from northwest to Description and interpretation (estimates in on plate 1) southeast) years B.P.) Loon Lake episode (1)--Duane Center—Walker Mill— Knolls and ridges of gravel and till form recessional moraine About 12,700. Loon Lake—Merrilville. with ice-contact slope on northeast side as much as 200 feet high. Glaciofluvial deposits that form pitted outwash plains were laid down by southwest-flowing melt-water streams from the ice front. Drainage was west by way of St. Regis River valley. Kames and ice channel fillings west of Duane Center, not shown on pl. 1, mark head of older esker system extending southwest for about 40 miles (Chad- wick, 1928; Buddington and Leonard, 1962, fig. 2). Owls Head—Redford Owls Head—Ragged Lake— In the St. Lawrence basin, the ice front retreated 4 to 8 miles About 12,600. stand (2). Middle Kiln—Redford—Clintonville northeast as far as position 2a, leaving masses of stagnant ice in some of the valleys. Knolls and ridges of sand, gravel, and till form recessional moraine near Owls Head. Extensive glaciofluvial deposits south and southeast of the moraine were laid down by south-flowing melt-water streams that escaped west around and across masses of dead ice into the valley now occupied by Lake Titus. Glaciofluvial deposits south of Ingraham Lake, recessional moraine near Ragged Mountain. and abandoned ice-marginal stream channels near Middle Kiln probably mark the same ice-front position (2a). In the Champlain basin, the Saranac River valley was blocked by the ice sheet near Redford (2b), and the waters were di- verted by way of several abandoned stream channels into the Ausable River Valley. perhaps contributing to the formation of the Coveville delta of glacial Lake Vermont near Clinton- ville. The ice front (2b) is marked by small recessional mor- aines near Redford, in the Salmon River valley west of Peasleeville, and on edge of the uplands northwest of Clinton- ville. Trout River—Moflits- Whippleville—Lyon Mountain— In the St. Lawrence basin, the ice front retreated 3 to 6 miles ville stand (3). Moi-fitsville—Keeseville. north, leaving masses of stagnant ice in the Salmon River valley and probably elsewhere. The ice front (3a) is marked by knolls and ridges of drift (recessional moraines) near mouth of Trout River valley and in the valley south of Upper Chateaugay Lake. In the Champlain basin, the ice front retreated 2 to 4 miles to position 3b, leaving dead-ice masses in the valleys of True Brook, Salmon River, and probably elsewhere. Melt-water streams built kames adjacent to the stagnant ice. The ice front (8b) is marked by small recessional moraines east of Peasleeville, abandoned stream channels, and glaciofluvial de- posits east of Terry Mountain, and small recessional moraines between Harkness and Keeseville. The Saranac River flowed east down the Salmon River valley, dissecting the moraine (2b) southwest of Peasleeville. Malone—Schuyler MaloneBrainardsville— In the St. Lawrence basin, the ice front retreated 8 to 7 miles Falls stand (4). Dannemora—Schuyler Fails. to the mountain front (4a), leaving masses of stagnant ice in some of the valleys. Ice-marginal channels and small masses of glaciofluvial deposits are present near Brainards- ville and south of Ellenburg Center (just east of the St. Law- rence-Champlain divide). In the Champlain basin, the ice front retreated 2 to 6 miles leaving stagnant ice masses in the Saranac and the Salmon River valleys. During the retreat, glaciofluvial deposits (kames) were laid down by melt-water streams, and south of the Salmon River, other ice-marginal streams cut channels in drift emptying into glacial Lake Vermont at the Coveville stage. Icefront position 4b marks the western limit of the prominent recessional moraine near Cadyville. To the south, this ice front formed the dam at the north end of Lake Vermont. Chateaugvay—Cadyville Area between Malone—Brainardsville— In the St. Lawrence basin, the ice front retreated 8 to 10 episode (4, 5, 6, 7, Dannemora—Schuyler Falls stand miles, almost to the International Boundary (5). During the and 8). and Frontier—Clinton Mills— retreat an ice-dammed lake in the St. Lawrence basin drained Ellenburg Depot—West east across cols into the Champlain basin, removing the drift Beekmantown stand. to expose bedrock in an area extending from near Clinton Mills south to the vicinity of Ellenburg. The ice front readvanced southwest to position 6, about 3 miles south of Churubusco. Outlet streams from small ice- dammed lakes in the Champlain basin crossed into the St. Lawrence basin where, augumented by melt water, they out many west-draining ice-marginal channels in the area between Belmont Center and Chateaugay as the ice front retreated from position 6 to position 7. The Trout. Little Trout, and Chateaugay Rivers built deltas into glacial Lake Iroquois (Coleman 1937). A short retreat of about 2 miles brought the ice front to posi- About 12'400' tion 8 where several prominent moraines were built dam- ming a lake in the valley of the North Branch and English Rivers which overflowed to the west through a col north of Clinton Mills into the St. Lawrence basin. In .the Saranac River valley near Cadyville, a belt of reces- 12,400 to 12_5oo. snonal moraines (4b to 8) was built by a southwest-moving ice tongue that blocked the mouth of the valley, causing runoff and melt water to flow south along the ice edge. The moraine may record a minor readvance. As the ice front retreated northeast into the Champlain basin, it built several linear morainal ridges, and south-flowing ice-marginal streams HISTORY OF DEGLACIAT‘ION 19 TABLE 1.—Histo'ry of deglwciation of the northeast Adirondack region, New York—Continued Location of stand or area deglaciated during episode (points near ice front listed in order from northwest to southeast) Episode or stand (numbers refer to ice- front position shown on plate 1) Age of episode or stand (estimates in Description and interpretation years B.P.) Chateaugay—Cadyville episode (4, 5, 6, 7. and 8)—Continued Covey Hill episode (8, 11, 14, and 15). Lowering of level of Lake Vermont. Area between Frontier—Clinton Mills—Ellenburg Depot—West Beekmantown stand and Franklin Centre—Covey Hill—Cannon Corners stand. Near the St. Lawrence—Champlain divide, the ice front re- cut successive channels on both the north and the south slopes of the Saranac River valley, washed clean, large areas of bedrock, and emptied into Lake Vermont at the Coveville stage. treated about 2 miles to the north and east, and Lake Iroquois in the St. Lawrence basin overflowed to the east across the divide at a point 2 to 3 miles west of Covey Hill, Quebec. At about this time, the level of Lake Vermont dropped 150 to 175 feet to that of the Fort Ann stage. The overflow from Lake Iroquois emptied into a series of ice- marginal lakes in the northeast-draining valleys of the Eng- lish, North Branch, and Great Ch‘azy Rivers .The south- easterly overflow of these lakes, perhaps in part catastrophic, cleaned ofi bedrock divides in an area extending roughly from West Chazy northwest to the International Boundary. The outflow from Lake Iroquois cut the deep rock-walled gorge on the divide west of Covey Hill and discharged into Lake Vermont at the Fort Ann stage. During this interval, the Fort Ann lake rose 50 to 80 feet to its highest stand. There were probably several minor advances and retreats of the ice front as Lake Iroquois in the St. Lawrence basin was lowered to the level of Lake Vermont in the Champlain Basin. During the Covey Hill episode, a subglacial stream built the Ingraham esker (Woodworth, 1905s) in the low- lands north of Plattsburgh. Ice front retreated to vicinity of Montreal. St. Lawrence Low- lands still dammed by ice sheet near Quebec City. Diflerential uplift closed southern outlet of Lake Vermont, and it drained northeast along the southeast side of St. Lawrence Lowlands between the ice sheet and the valley wall (glacial Lake New 12,400 to 12,200. York, Wagner, 1969). Champlain Sea ...... Ice front retreated to north side of St. Lawrence River, and 12,000 to 10,500. marine waters invaded the St. Lawrence and Champlain val- leys. Formation of Lake Champlain. Differential uplift closed the connection to the ocean, and the marine waters were replaced by a fresh-water lake. 10,500 to 10,000. rather than the product of the small modern stream. The late-glacial river dissected the moraine west of Peasleeville to form the broad gravel-covered flood plain and to build a gravel delta into a small ice- dammed lake. The eastward growth of the delta stopped when the ice front in the Saranac River valley to the north had retreated east to the point where the Saranac River no longer overflowed south into the headwaters of the Salmon River but assumed an easterly course toward Cadyville. The stream in the Salmon River valley was greatly reduced in size and its load greatly diminished. MALONE-SCHUYLER FALLS STAND (4) In the St. Lawrence basin this ice-front position (4a) is largely inferred. In the Champlain basin, the Malone-Schuyler Falls stand (4b) is drawn along the western edge of the moraine near Cadyville in the Saranac Valley and the western edge of the areas of bare rock on the uplands to the south. This stand marks the ice-front position at the beginning of the formation of the moraine and of the abandoned stream channels, kames, and bare-rock areas associated with it. In the Saranac Valley during the ice-front retreat from Mofl‘itsville (3b) to the west side of the mo- raine near Cadyville (4b) , glaciofluvial deposits were laid down as ice-margina1(?) kames. Sand bodies near Moffitsville rise to altitudes slightly above 1,000 feet and presumably were deposited by the Saranac River as it flowed east toward Cadyville along the north side of the ice (stagnant?) in the Saranac Valley. The floor of the lowest abandoned stream channel on the Saranac-Salmon divide is more than 100 feet above the top of the sand bodies near Moffits- ville. East of Terry Mountain, small recessional mo- raines and south-draining ice-marginal channels that end in glaciofluvial deposits were formed during ice retreat from position 3b to 4b. CHATEAUGAY-CADYVILLE EPISODE (4, 5, 6, 7, AND 8) This episode includes the uncovering of the up- lands from the northeast front of the mountains (4a and b) north almost to the Canadian border, and 20 PLEISTOCENE GEOLOGY OF THE NORTHEAST ADIRONDACK REGION the moraines in the valleys of the Great Chazy River and its North Branch (8) and of the Saranac River near Cadyville (4b to 8) . In the St. Lawrence basin, the ice edge retreated northward from the mountain front to a point on the St. Lawrence-Champlain divide about 3 miles south of Churubusco at an altitude of about 1,350 feet (ice-front position 6, pl. 1) . A small ice-dammed lake east of the divide overflowed west, cutting small west-sloping stream channels, now abandoned, at altitudes of 1,335, 1,325, and 1,105 feet (pl. 1). Five to 10 miles west of these small channels are many large west-draining abandoned stream chan- nels at altitudes ranging from about 1,300 feet to about 1,050 feet. These channels are presumably ice- marginal channels cut by melt water and (or) by overflow from the small lake east of the divide as the ice edge retreated northward (positions 6 and 7). Evidence suggests that, prior to the cutting of the abandoned stream channels between Churubusco and Malone, the ice edge may have retreated nearly to the Canadian border (ice-front position 5) and then readvanced to position 6, about 3 miles south of Churubusco. In northern Clinton County, west of the St. Lawrence-Champlain divide, as MacClintock and Stewart pointed out (1965, p. 58, pl. 13), the bed- rock is largely concealed beneath thick drift, whereas east of the divide between Clinton Mills and Ellen- burg, the drift is thin and patchy. Over large areas the bedrock is covered only by a thin layer of rubble or of peat. This contrast in extent and thickness of surficial mantle suggests the washing of the east slope of the divide by running water that spilled over the divide from an ice-dammed lake to the west. This easterly flowing water could have entered the Champlain basin by way of several cols on the divide at altitudes ranging from just above 1,300 feet about 4 miles south of Churubusco down to about 1,050 feet near the International Boundary. The easterly flowing water washed the east slope of the divide roughly from Clinton Mills south to Ellenburg and presumably entered a local ice-marginal lake in the headwaters of the North Branch Great Chazy River. This washing antedates that which cleaned off the large areas of bare rock between Covey Hill, Quebec, and West Chazy, because these latter bare—rock areas are at lower elevations than the area between Clin- ton Mills and Ellenburg. The ice-dammed lake west of the divide that over- flowed through the cols must antedate the cutting of the west-draining abandoned stream channels be- tween Churubusco and Malone. No deltas or beaches of this high-standing lake (alt 1,050—1,300 ft), have been positively identified. This lake may have been of limited extent, or it may have been an arm of Lake Iroquois. In the valley of the North Branch Great Chazy River, the Chateaugay-Cadyville episode ended with the building of the moraine near Ellenburg Depot (ice-front position 8, pl. 1 and fig. 11), where the front may have paused for a time. The ice built the moraine along the east edge of an ice-dammed lake that drained to the west, probably by way of the large col north of Clinton Mills at an altitude of 1,085 feet. If the ice that built the moraine south of Miner Lake dammed the valley of the Great Chazy River and formed a lake, this lake drained north into the valley of the North Branch, perhaps washing the drift cover from part of the area of bare rock on the divide between the two valleys. There is no evidence of the escape of lake waters from the Great Chazy Valley around the east end of the mo- raine south of Miner Lake (ice-front position 8). However, the area of bare rock on the Great Chazy— North Branch divide extends down to an altitude of about 1,050 feet, too low relative tothe col at 1,085 feet to have been washed completely by northwest- flowing lake waters. Before the washing of the area on the divide was completed, the direction of flow changed from northwest to southeast. At about the time the ice blocked the valley of the North Branch and built the moraine near Ellenburg Depot (ice-front position 8), a stream flowing south in or under the ice sheet in the low- lands north of Plattsburgh began to build the Ingra- ham esker, a ridge of gravel about 10 miles long (Woodworth, 1905a, pl. 4). The esker stream dis- charged into Lake Vermont and, as the ice edge re- treated north, the bouldery ridge formed in the ice was partly buried by sand deposited in the glacial lake. Later, submergence of the esker in the Cham- plain Sea resulted in extensive modification of the ridge by erosion and deposition (Denny, 1972). V. K. Prest (1970), in his thoughtful review of the present state of knowledge of the Quaternary geology of Canada, has presented a slightly different interpretation of the late-glacial history of the area northeast of the mountains, as shown in table 2. Glacial Lake Iroquois (Coleman, 1937) in the Lake Ontario basin drained east near Rome, N .Y. (fig. 14) , southwest of the Adirondacks, into the Mohawk River valley. Coleman (1937 ) and most later workers believe that the northward retreat of the ice front in the Ontario basin and the St. Lawrence Valley in time opened a new outlet for Lake Iroquois, the channel near Covey Hill, altitude 1,010 feet. The Rome outlet was then abandoned. Field observations made _by Prest in 1967, in company with E. P. Hen- HISTORY OF DEGLACIATION 48° 78" 76° 74° 72' 70“ l l I I l EXPLANATION We S‘- M" / 8 12.400 1921/ r Moraine or drift border _ Ticks point toward glacier. Small num- ber imitcates ice-from position in north: _ east Adirondack region; large number indicates years before present 6' Riviera du Loup @IQ R) . ,. . 0. IN Lake Vermont and Lake Iroqums “9‘ a). l ' \9 \x/ x- 0° 0 a <0 as .x i“ rv we /~ 10 VI} Champlain Sea 0/ “Q . O / N O 25 50 75 100 MILES 46° — Lb;_l__l__l 0 1 2 3 4 KILOMETERS L l—gL—L" / 2 I MQNTREAL \ ( x ' ‘ <52. ‘2‘ o 4 gift, OTTAWA ,, Granby SherbrookeJ‘lj CHAMPLAIN SEA \’ St. Lawrence Seaway] _ 1 '70 9.? 00 Malone ? O ‘ I 00 x7}, 44° -— . O . \@ :F'YX‘ W _ Ont WWI NEW HAMPSHIRE I Rome NEW YORK l l FIGURE 14.—Generalized map showing location and possible age of moraines and late-glacial water bodies in parts of the St. Lawrence and Champlain Lowlands. As used'here, Lake Iroquois includes the Frontenac and Sydney phases of the post-Iroquois lakes as defined by Prest (1970, p. 727 and fig. XII—16f and g). Compiled largely from MacClintock and Stewart (1965), Prest, Grant, and Rampton (1968), and Stewart and MacClintock (1969). derson and the author, led Prest (1970, p. 727) to question the earlier interpretation. Though the Iroquois shoreline is reported at only 1,100 feet southwest of Covey Hill there is evidence of shoreline features to about 1,250 feet, 1%. miles south of Churubusco. This is considered to be an Iroquois shore rather than that of local lake. Prest also suggested that a small embayment of Lake Iroquois extended eastward from Churubusco [into the Champlain basin] as a reentrant between the receding northern and eastern ice-fronts around Adirondack Mountains. This ice, probably Fort Covington [MacClintock and Stewart, 1965, p. 60], built an end moraine 22 PLEISTOCENE GEOLOGY OF THE NORTHEAST ADIRONDACK REGION and ribbed moraine complex south and southwest of Covey Hill. These are the moraines near Ellenburg Depot and north of Clinton Mills (ice-front position 8, pl. 1). C. S. Denny (USGS, 1968) believes that an outlet opened southeast of Ellenburg, N.Y., which allowed main Lake Iroquois to breach the end moraine and discharge into glacial Lake Vermont farther south in Lake Champlain val- ley. This discharge was thought to mark the beginning of the washing of Flat Rock, near Altona. Prest (1970, p. 727) also described how the drain- ing of the arm of Lake Iroquois southeast of Ellen- burg lowered lake level about 75 feet to what he named the “Ellenburg phase of the post-Iroquois lakes.” He went on to say as glacier recession was resumed, Covey Hill outlet (sill ele- vation 1,010 feet) was uncovered and the post-Iroquois lake level was lowered a further 75 feet. This lake phase was named glacial Lake Frontenac and is herein considered the Frontenac phase of the post-Iroquois lakes * * *. I now question some parts of Prest’s interpretation of the history of the divide area. The abandoned stream channels trending diagonally down the west slope of the St. Lawrence-Champlain divide between Malone and Churubusco (pl. 1; see also MacClintock and Stewart, 1965, fig. 21) are best explained as the work of ice—marginal streams. Successive channels were cut as the ice edge retreated northward (posi- tions 6 and 7). Because the channels do not appear to have been modified by ice or by wave action, I believe that they record the retreat of the ice margin when Lake Iroquois was still discharging into the Mohawk River valley by way of the Rome outlet. The moraine near Cadyville was built by a short tongue of ice moving southwest up the Saranac Val- ley from a large ice mass in the Champlain Low- land. There may have been stagnant ice in the Saranac Valley west of the moraine. At the begin- ning of the episode of moraine building, the ice front stood near thewest edge of the moraine (4b) and crossed over the east end of Johnson Mountain into the headwaters of the Great Chazy River, which presumably were dammed by the ice edge, although no record of this lake has been found. South of the Saranac River, the ice front (4b) pressed against the east end of the ridge to the south for a short distance before turning east into the Champlain basin to form the ice dam at the north end of Lake Vermont. Under the reconstruc« tion of glacial history presented here, the building of the moraine near Cadyville took place during the time that the ice front in the St. Lawrence basin retreated from the mountain front (4a) nearly to the Canadian border (8). The ice tongue in the Saranac Valley built small moraines as its edge retreated to the northeast. The Saranac River. blocked by the ice, was diverted south around the east end of the bedrock hill that separates the Saranac and Salmon River valleys; the waters carved channels in drift and in bedrock, washed drift from areas of bedrock, and deposited their load close to the north end of Lake Vermont. Drainage ponded north of Johnson Mountain flowed south into the Saranac Valley, cutting channels in drift and depositing gravel and sand against the ice in the valley. As the ice edge retreated, many channels were cut on the uplands west of West Beekmantown by water ponded north of the Saranac Valley—Smith Wood Brook divide. The water escaped southward into the head of Lake Vermont, which was then only a few miles south of the Saranac River. COVEY HILL EPISODE (9 THROUGH 15) The interval between the withdrawal of the ice from the moraine near Ellenburg Depot (8) and the uncovering of Covey Hill and the merging of the glacial lakes in the St. Lawrence and Champlain Valleys is called the Covey Hill episode. During this episode, the channel near Covey Hill was eroded, large areas were swept clean of their drift cover, and the ice-dammed lake in the upper St. Lawrence and Ontario basins was lowered more than 200 feet to the level of the glacial lake in the Champlain Valley. The ice edge had to maintain a position across the divide close to Covey Hill in order to hold the ice-dammed lake in the St. Lawrence Valley south- west of Montreal in a position where it could over- flow into the similar lake in the Champlain Valley. If the ice front had retreated a mile or two north- east, the lakes would have become confluent, and the water in the upper St. Lawrence and Ontario basins would have discharged catastrophically into the Champlain Valley. The channel near Covey Hill, altitude 1,010 feet, that controlled lake level in the upper St. Lawrence and Ontario basins, was about 270 feet above the level of Lake Vermont in the Champlain Valley. On the other hand, if the ice front had readvan-ced southwest a few miles, it would have blocked all the possible outlets by which St. Lawrence Valley waters could have overflowed into the Champlain Valley. During the Covey Hill HISTORY OF DEGLACIATION 23 TABLE 2.—Ice-marginal and shore features near Covey Hill, Quebec, as interpreted by Prest (1970) and by Denny (this report) Feature and altitude Denny (this report) Prest (1970) Gravel and sand near Churubusco, N.Y.; alt 1,250 ft. Large area of abundant outcrops of Potsdam Sandstone on divide be- tween Graves Brook and North Branch about 2 miles southeast of Ellenburg (pl. 1) ; alt 1,200—1,300 ft. Channel near Covey Hill, Quebec; alt 1,010 ft. Large area of abundant outcrops of Potsdam Sandstone north of Eng- lish River; alt 750—880 ft. on pl. 1). Glaciofluvial deposits (not shown Ice-dammed lake in valley of North Branch overflowed across divide into valley of Graves Brook and washed clean the bare-rock area between the two streams. Outlet of Lake Iroquois. Beach deposits of Lake Iroquois. Ellenburg phase of the post-Iroquois lakes (“Outlet * * * southeast of Ellenburg”; Prest, 1970, p. 727). Frontenac phase of the post-Iroquois lakes. Lake Iroquois overflowed by way of an ice-marginal stream coming around the north side of Covey Hill. The stream is assumed to have been the outlet that controlled the level of Prest’s Sydney phase of the post-Iroquois lakes, alt about 885 ft (Prest, 1970, p. 727). The lake water washed clean the bare-rock area. Beaches marking the highest stand of Fort Ann stage of Lake Ver- mont; altitude at International Boundary about 740 ft. basins. Lake Vermont merged with lake in upper St. Lawrence and Ontario Belleville-Fort Ann phase of the post- Iroquois lakes. episode, as here interpreted, the ice edge moved back and forth across the area of the present In— ternational Boundary several times, and the wash- ing of the bedrock and the lowering of the level of ponded water in the St. Lawrence basin took place during several intervals, not all at one time. Retreat of the ice front from the moraines north of Clinton Mills and near Ellenburg Depot (ice- front position 8, fig. 11) uncovered the col (1,010 ft) on the divide west of Covey Hill; the opening of this col permitted Lake Iroquois to overflow into the Champlain basin and presumably to abandon the outlet near Rome. Southwest of the col, the divide is bare rock up to an altitude of about 1,050 feet, sug- gesting that Lake Iroquois may have stood at that level when it first crossed the divide. Presumably the col at 1,010 feet soon became the control for the level of outflow of Lake Iroquois. At first the ice-marginal streams turned to the south down the slope (ice-front position 9, fig. 11) just east of the moraine near Ellenburg Depot and emptied into a long narrow ice-marginal lake or series of lakes between ice-front positions 8 and 9. These ice-dammed lakes drained east across the southern edge of Flat Rock. Continued retreat of the ice front caused the water escaping from Lake Iroquois to flow south along the ice edge (position 10) at altitudes ranging from about 900 to 950 feet; the flow cut bedrock( ?) channels a mile or two to the south of the Interna- tional Boundary and emptied into a long narrow ice- marginal lake or chain of lakes that overflowed to the east across Flat Rock, washing it clean and cut- ting narrow gorges in the bedrock. Some of the ma- terial removed from Flat Rock was deposited to the east on Cobblestone Hill (pl. 7), where it was later V heaped up into conspicuous beach ridges. On Flat Rock, the floor of the Dead Sea at the head of Cold Brook (pl. 7) is said to be 90 feet below the rock rim on the west side of the rock basin. Woodworth thought that such a deep pool could not have been eroded by a small ice-marginal stream de- scending over “so slight a fall as the rock cliff at its head, but it is quite conceivable that a heavy torrent might have produced the results” (Woodworth, 1905a, p. 20). On the north and east sides of Flat Rock, beach deposits of the Fort Ann stage of Lake Vermont extend up to a point 80 feet above the lowest part of the area of bare rock. The overlap of the Fort Ann beaches on the bare rock suggests that during the washing of Flat Rock the level of glacial Lake Ver- mont may have stood as much as 80 feet below its highest stand. If this was so, then the level of Lake Vermont rose during the Fort Ann stage, probably because of uplift of the outlet at the south end of the Champlain basin near Fort Ann, N.Y., about 100 miles to the south (fig. 14). Woodworth (1905b, p. 162) interpreted the overlap of beach on bare rock as indicative of a rise in lake level after the drift cover had been washed from the bare—rock area by ice-marginal streams. When the ice edge had retreated as far as position 11 (fig. 11), water from the Covey Hill channel turned south into a lake in the English River valley. This lake in turn overflowed to the southeast along the ice edge across the divide between the English River and the North Branch. Small northwest- 24 PLEISTOCENE GEOLOGY OF THE NORTHEAST ADIRONDACK REGION trending recessional moraines just south of the Eng- lish River—North Branch divide mark positions of the ice front that controlled the overflow. Perhaps much of the morainal material and the gravel asso- ciated with it was washed off the bare-rock area to the northwest by ice-marginal streams. Removal of drift by overflow is essentially the explanation ad- vanced by Woodworth (1905a, p. 12). As the ice edge retreated to the northeast (toward ice-front position 12), water from the Covey Hill channel escaped south across a col at an altitude of about 860 feet (pl. 6) and flowed into a prominent bedrock channel that runs south to drain into the ice-dammed lake in English River valley. This lake continued to drain across the divide to the south, cleaning off the bedrock. As the ice front continued to retreat (position 12) , water from the Covey Hill channel followed a more southeasterly course across the western part of the bare-rock area north of English River, along the west side of a prominent recessional moraine. Near English River, the ice front is assumed to have turned east and crossed the English River—North Branch divide where water ponded in the English River valley continued to overflow to the southeast, removing the drift. When the ice edge had retreated to the northeast down the English River—North Branch drainage di- vide a short distance beyond ice-front position 12, the overflow from the lake in the English River val- ley deposited bouldery material at the northeast end of the bare-rock area, about 0.75 mile south of Can- non Corners (pl. 6). Shallow kettles in these boul- dery deposits suggest deposition in association with glacier ice where the overflow entered Lake Ver- mont. The large area of bare rock north of English River could have been washed by lake water coming either by way of the gorge west of Covey Hill or around the north side of the hill. The existing segments of mo- raine are not sufl‘icient to define the ice edge pre- cisely, but in the reconstruction of figure 11, I have assumed that at ice-front position 12, water came from the gorge and at ice-front positions 13, 14, and 15 the water was coming around the north side of Covey Hill. The history of deglaciation presented here calls for the lowering of the level of glacial Lake Iroquois in several stages rather than in one catastrophic out- flow. If the lake in the St. Lawrence Valley at the level of the Covey Hill outlet, altitude 1,010 feet, had been lowered at once to the Fort Ann level, at about 750 feet, one would expect to find on the north and east slopes of Covey Hill large areas of bare rock swept clean by the flood; instead, these slopes are boulder covered. I assume that the lake to the west was lowered gradually, perhaps because the ice edge moved up and down the north slope of the hill. Some of the water may have escaped beneath the ice. Prest (1970, p. 727) believes that the existing strandlines in the Ontario basin call for a brief still- stand of a post-Iroquois lake on the north side of Covey Hill below the level of the lip of the gorge. When the ice withdrew from the northern and eastern flanks of Covey Hill there was a major dr0p in lake levels of some 125 feet. The short stand at this level, probably occasioned by an ice—marginal fluctuation on the northern side of Covey Hill, is termed the Sydney phase lake by E. Mirynech (1967). In the Trenton embayment in eastern Ontario the lake was lowered a further 30 to 75 feet according to Mirynech, prior to a significant halt responsible [for development of the Belleville beach. Isobases drawn on the Belleville beach would place the strandline on Covey Hill at about 750 feet. This is the same as that of the Fort Ann phase of glacial Lake Vermont which expanded northward as the eastern side of Covey Hill was uncovered by the ice. I suggest that the removal of drift from part of the bare-rock area north of English River (positions 13, 14, and 15) was accomplished during the “Syd- ney phase” by lake water escaping from the St. Lawrence Lowlands around the north side of Covey Hill. This water eroded low bedrock knolls near the Canadian border (pl. 6), washed bare the bedrock area now exposed, and probably entered Lake Ver- mont near Cannon Corners. The position of the recessional moraines in the English River valley suggests a minor readvance of the ice front. The reconstruction of figure 11 shows the ice front first retreating to position 13 near Can- non Corners, then advancing up the English River valley about a mile (position 14), and then retreat- ing again to a line north of Cannon Corners (posi- tion 15). The evidence for this minor readvance is as fol- lows: The ridge of recessional moraine on the east side of the bare-rock area north of English River (pl. 6) joins at its south end a small northwest- trending segment of recessional moraine; together the two enclose a small swamp. These ridges would have partly blocked the southerly flow of water es- caping around the north side of Covey Hill. Perhaps the segments of moraine near the swamp are giant gravel bars on the bed of a south-flowing river. If so, the sequence of events may have been as follows: When the edge of the ice sheet retreated to the east side of the bare-rock area, water from the north side of Covey Hill flowed south along the ice edge, cleaned off an area of bedrock somewhat larger than the area DEPOSITS AND SHORE FEATURES OF LATE-GLACIAL WATER BODIES 25 of bare rock existing today, and deposited its load in the valley of English River. During that interval, the level of a lake in the St. Lawrence Valley was low- ered about 125 feet and was stabilized at about 885 feet for a short time (the “Sydney phase lake” of Prest) . In the valley of English River, following this brief episode of erosion, the ice edge readvanced about a mile west across the southern part of the area previously cleaned oif, plowing up bouldery ma- terial from the valley floor to form the moraine along White Road. The water from the north side of Covey Hill escaped around the west end of the lobe (point B, pl. 6) into the English River valley and east across the divide to the south into Lake Vermont just south of Cannon Corners. During the shrinkage of this minor lobe, the water escaped across the bouldery ridges, perhaps first near point C and later near point D (pl. 6) on the east side of the bare-rock area. Immediately thereafter, the water followed the channel at the north end of the bare-rock area (ice-front position 15) to enter Lake Vermont near the present International Bound- ary (point E, pl. 6). The ice sheet soon retreated 01f Covey Hill for the last time, and Lake Vermont merged with the lake in the upper St. Lawrence and Ontario basins (fig. 14). DEPOSITS AND SHORE FEATURES OF LATE-GLACIAL WATER BODIES The late-glacial history of the Champlain Valley involves an ice-dammed fresh-water lake followed by an incursion of the sea, the latter ending about 10,500 years B.P. Woodworth, who made detailed studies throughout the Hudson-Champlain lowland (Woodworth, 1901, 1905a, b), recognized both ma- rine and fresh-water deposits. Later, Fairchild A (1919) concluded that the beaches and deltas de- scribed by Woodworth were formed in an arm of the sea extending from New York City to Montreal. Chapman’s (1937) study of the Champlain Valley largely reaflirmed Woodworth’s belief that the higher shoreline features are lacustrine and the lower ones marine. Lake Vermont, named by Woodworth (1905b, p. 190—206) , was the fresh-water lake that occupied the Champlain Valley during late-glacial time. It was dammed on the north by the edge of the ice sheet and overflowed to the south across a bedrock divide into the headwaters of the Hudson River. Since Wood- worth’s day, the only descriptions of the lake as a whole are by Chapman (1937) and by Stewart and MacClintock (1969). Chapman recognized and named two stages of lake history. The older stage, the Coveville, he believed was graded to Woodworth’s Coveville outlet, an abandoned channel and dry wa- terfall on the west bank of the Hudson River near Schuylerville about 30 miles north of Albany. (See Schuylerville 15-minute quadrangle.) Chapman named the younger stage the Fort Ann for its aban- doned outlet, which contains giant potholes, near Fort Ann, N.Y., east of Glens Falls (Chapman, 1937, figs. 6, 7, and 8). (See also Fort Ann 15-minute quad- rangle.) A still higher and older stage, the Quaker Springs, has been recognized in the southern part of the Champlain basin (Woodworth, 1905b, p. 103; Stewart and MacClintock, 1969, p. 163; Connally and Sirkin, 1971). The marine invasion that followed Lake Vermont has long been known as the Cham- plain Sea (Karrow, 1961; Elson, 1969). In the lowlands of the Plattsburgh area west of Lake Champlain, the deposits of glacial Lake Ver- mont resemble those of the Champlain Sea; they cannot be separated on a lithologic basis. The depos- its of these late-glacial water bodies are in two belts that parallel the edge of the uplands. The western or higher belt consists of elongate bodies of sand and gravel (Lake Vermont, pl. 1) that are separated by areas of ground moraine from the fossiliferous sand, gravel, silt, and clay of the eastern or lower belt (Champlain Sea, pl. 1). Only along the principal streams are the deposits of the two belts in contact. Chapman studied the beaches and deltas of the two belts in parts of the Dannemora quadrangle and made a precise survey of the beaches of the upper group in a small area west of Peru (Chapman, 1937, fig. 9). The upper belt as far north as the Saranac River is considered by Chapman to include both the Coveville and Fort Ann stages of Lake Vermont, but north of the Saranac River only the Fort Ann stage is present. I have adopted Chapman’s terms, Fort Ann and Coveville, for the two groups of deposits and shoreline features of the upper belt. The fossilif- erous deposits and shoreline features of the lower or eastern belt I assign, as did Chapman, to the Cham- plain Sea. The deposits of glacial Lake Vermont cannot be traced continuously from the Plattsburgh area south to the region of the outlet, nor are there fossils or carbon-14-dated materials with which to correlate the isolated deposits from place to place. It is a well- established fact that the glaciated area of northern United States and Canada was uplifted when the ice sheet disappeared, the amount of uplift increasing inward from the periphery of the glaciated area (Daly, 1934) Therefore, if the altitudes of the shore features of glacial Lake Vermont are plotted on a 26 PL‘EISTOC’ENE GEOLOGY OF THE NORTHEAST ADIRONDACK REGION profile running north-south more or less in the as- sumed direction of maximum tilt, the upper limit of these features should define the position of the high- est stand of the lake. The limit rises northward and is probably, but not certainly, a time line; all fea- tures on the line formed at about the same time. The correlation of the shore features (beach, spit, ter- race, wave—cut cliff, or delta) throughout the Cham- plain basin has been and is based largely on their location and altitude in relation to possible outlets, as shown on such north-south topographic profiles (Woodworth, 1905b, pl. 28; Chapman, 1937, figs. 15 and 16; Denny, 1967, 1970). The deposits of the late-glacial water bodies are dominantly sand; grave] is less abundant and com- monly occurs only near the larger streams. Near Lake Champlain in the drainage of the Great Chazy River, silt and clay are abundant. In favorable locali- ties, such as on the east slope of the uplands near West Beekmantown, the upper limit of wave action on recessional moraine is clearly marked (Stephens and Synge, 1966, figs. 9—12). In many places the upper foot or two of the till is a faintly stratified gravelly material, and the local relief of the surface of the ground moraine or of the glaciofluvial deposits is less than on similar materials above the limit of wave action. BEACH DEPOSITS Beach deposits closely resemble the adjacent drift from which they were derived. Where the drift con- tains many fragments of Potsdam Sandstone, the beach ridges are composed of a flaggy gravel (fig. 15). (See also Denny and Goodlett, 1968, figs. 2—5.) Where the drift contains abundant large clasts, the beach-forming materials are boulder or pebble gravel, commonly massive or faintly stratified (fig. 16). Only a few sandy beach ridges are recognized, in part because such features have been smoothed by frost action, tree-throw, or plowing. The beach ridges commonly occur in groups, some of which may be 3 or 4 miles long. Individual ridges range in length from a few hundred feet to nearly a mile, but most are less than half a mile long. The width of a ridge from trough to trough ranges from about 40 to 100 feet, the height from crest to trough, from about 1 to 5 feet. The availability of gravelly materials and expo- sure to wave action appear to be important factors in the location and abundance of beaches. Beaches are well developed on headlands. The most conspicu- ous examples are those near and southwest of Sciota (Denny, 1970), where a headland underlain by the Potsdam Sandstone projected northeast into the ad- j acent water body. Just south of Sciota, west of State Highway 22, are beach gravels of the Cham- plain Sea at the base of low sandstone bluffs that Woodworth (1905a, pls. 22—24) called sea cliffs. The most spectacular beaches, on Cobblestone Hill at the southeast end of Flat Rock (pl. 7), were de- scribed in detail by Woodworth (1905a, p. 32-35, pls. 12, 13, and 14). Pebble to boulder gravel and sand mantle the north and east sides of Cobblestone Hill; the deposit is more than a mile long and as much as 0.3 mile wide and ranges in altitude from about 580 feet at the base of the hill, where the material is chiefly sand, up to about 670 feet where the material is boulder gravel. The gravel is heaped up into beach ridges, as much as 3 feet high and 100 feet wide, that extend along the east side of the bill for at least a quarter of a mile to its south end, where they curve around to the west and north enclosing small de- pressions between adjacent ridges. At the north ends of the beach ridges, rounded clasts of white sandstone form a pebble and cobble gravel (fig. 17) , but near their south ends the ridges are mantled by angular blocks of sandstone (fig. 18) , suggesting that the beaches did not grow southward but rather were formed in place. Thus, the shape of the beach ridges, even though resembling a curved spit, is related instead to the form of Cobblestone Hill. The crest of the hill declines southward to pass beneath the position of the highest stand of the Fort Ann stage. The curvature of the beaches near their southern ends appears to be the result of the shape of the hill crest where it passes below lake level rather than the result of the southward movement of material by longshore currents. The beach deposits are reworked alluvial materials washed off the adja- cent bare—rock area just before beach formation. The sand and gravel at the east base of the hill are prob- ably material removed by wave action from the gravel near the top of the hill. DELTA DEPOSITS Sand and gravel form deltas at the mouths of the larger streams where they debouched into the late- glacial water bodies (pl. 1). Similar materials that blanket gently sloping plains between deltas are called, for convenience, nearshore deposits. These de- posits are, in part, the bottomSet beds of deltas and, in part, material spread by longshore currents. The deltas are composed of medium to fine sand and lesser amounts of coarse sand and pebble to cob- DEPOSITS AND SHORE FEATURES OF LATE-GLACIAL WATER BODIES 27 FIGURE 15.—Cross section of flaggy gravel forming beach ridge of Champlain Sea. View looking north along Mooers- Chazy town line and 0.6 mile east of State Route 22. ble gravel. Gravel is most abundant near the apex of a delta (figs. 19, 20), where the channel of the delta- building stream was restricted by the adjacent val- ley walls. Farther downstream the material is largely a well—sorted medium sand (fig. 21) . The rivers were fed largely by local runoff from their drainage ba- sins, except perhaps during Lake Vermont time when glacial ice may have contributed melt water and load. If Craft (1969) is correct in supposing that ice re- mained for a time in the high peaks region after the lowlands had been deglaciated, then perhaps the rivers continued to carry some melt water as late as Champlain Sea time. Some of the deltas retain much of their initial form; others built at high stands were partly eroded, and the material was redeposited to form deltas at lower levels. The town of Keeseville is on top of a delta built by the Ausable River in Lake Vermont during the Fort Ann stage. Except where dissected by the river, the initial top and foreslope of the delta are well preserved. Along the Saranac River, the sands form a series of discontinuous terraces from the top of Lake Vermont to Lake Champlain. Waves and currents spread a thin blanket of sand along the shore between the deltas. In broad inter- stream areas near Lake Champlain these sands may be several tens of feet thick. FIGURE 16.—Beach gravel of the Champlain Sea. Material contains a few fragments of marine shells. Exposed in bor- row pit on ridge crest, altitude about 300 feet, about 4 miles west of Champlain. 28 PLEISTOCENE GEOLOGY OF THE NORTHEAST ADIRONDACK REGION FIGURE 17.—Water-worn cobble and boulder gravel in beach of Lake Vermont. Exposure is in highest beach of Fort Ann stage along east-west road crossing south end of Cobblestone Hill (pl. 7). TEXTURE For the Plattsburgh area, mechanical analyses of sandy delta and nearshore deposits and sandy glacio- fluvial deposits were made in the field, using small sieves as the sorting unit and a 10-cc graduate as the measuring unit (Azmon, 1961). Most of the ap- proximately 120 samples, whose locations are shown in figure 223, came from a depth of about 3 feet. GRAIN SIZE The sand is dominantly medium and fine (fig. 22A). Coarse sand is generally restricted to deltaic deposits near the large rivers and to the glaciofiuvial deposits. The marine sands are largely deltaic, and size tends to decrease away from the rivers, except along the Salmon River below Schuyler Falls. Grain size shows no consistent relation to slope as meas- ured on a topographic map. Dune sand and sand of the Fort Ann stage are a little finer and more uni- formly sized than the marine sand or that of the Coveville stage (fig. 23). In the area between the Salmon and Saranac Rivers (fig. 22A), layers of uniformly sized sand parallel the ancient shorelines of Lake Vermont, suggesting that here the sand was spread by wave action near shore. Each of the four rivers in the study area built deltas into Lake Ver- mont and the Champlain Sea. Size tends to decrease away from the apex on some of these deltas (fig. 24) , but, in general, the relation is not consistent. SORTING The sands are well sorted. The Trask sorting co- efficient ranges from about 1.2 to 1.8 (fig. 223) ; in general, the coarser the sand, the poorer the sorting. INTERNAL STRUCTURE Neither the delta nor the nearshore deposits were well exposed during the course of this study. Excava- tions in the delta near Keeseville showed both topset and foreset beds. Pits open in the nearshore deposits during the construction of Interstate Highway 87 showed a more or less horizontal stratification and some rather massive beds a few feet thick. A few lenses of crossbedded sand and gravel were observed (fig. 20). In some of the delta deposits, contorted beds be- tween horizontal strata suggest deformation con- temporaneous with deposition, the result of the load- ing of sand on silty beds, of slumping, or of dewater- mg. West of Peru, sands of the Fort Ann stage form small knolls and ridges where the deposits lap up against the upland east of Terry Mountain (west and south of Clark Corners; Denny, 1967). The knolls are 10—30 feet high and 100—300 feet in diameter. They are on a drift-covered surface that slopes east- ward with a gradient of about 3° to 4°. Beds of fine sand are highly deformed. In cross section the folds are fan shaped or irregular and measure 3—5 feet across. The deformation appears to die out at depths of 5 or 6 feet below the surface. Perhaps the knolls of deformed sand are material that slumped down the till-covered slope when Lake Vermont drained away. LOCATION AND AREAL RELATIONS Deltas were mapped along four large rivers (pl. 1), the Salmon (Franklin County) and its principal tributary the Trout, the Great Chazy, the Saranac, and the Ausable. The drainage basins of these rivers range in size from about 225 to 600 square miles (table 3). Deltas were also mapped along five smaller streams, including the Little Trout and Chateaugay Rivers in Franklin County and Park Brook and the Salmon and Little Ausable Rivers in Clinton County. The drainage basins of the five smaller streams range DEPOSITS AND SHORE FEATURES OF LATE-GLACIAL WATER BODIES 29 ,y w v 3’. § 5! é FIGURE 18,—Boulder beach on Cobblestone Hill at southeast end of Flat Rock near Altona (pl. 7). View is looking north from near south end of the highest beach. Photograph by G. K. Gilbert. in size from about 25 to 150 square miles. The Sara- nac and Ausable Rivers head in the high peaks of the Adirondacks; the maximum relief of the Ausable drainage basin is about 5,000 feet. Of the four large river basins, only one, that of the Great Chazy River, is underlain predominantly by Paleozoic sedimentary rocks, chiefly the Potsdam Sandstone (pl. 3) . The others are largely in crystalline rocks of Precam- brian age. About two-thirds of the Ausable River drainage basin is underlain by anorthosite (pl. 2). VOLUME The volumes of the deltas have been calculated (table 3) in order to compare them with the topo- graphic and lithologic character of the drainage ba- sins upstream from the deltas. Many of them are dissected by streams that have cut down to bedrock, so that along the stream the thickness of the delta deposits can be measured. Well logs gathered from various sources by W. A. Hobba, J r., of the Geologi- cal Survey (written commun., 1967), are also avail- able. The estimated volumes are rough but appear to be the right order of magnitude. For some of the deltas, two estimated volumes are given, one for the original volume and one for the existing remnants. The younger deltas of the Ausable River are built, in part, of material eroded from older ones. The Coveville delta along the Ausable River west of Clin- tonville is estimated to have lost 203 million cubic yards of sediment by erosion since it was formed (table 3). Of this, about 122 million cubic yards are estimated to have been eroded by the Ausable River, the remainder by the Little Ausable River. Because the next younger and lower Ausable River delta, the Fort Ann delta near Keeseville, has an estimated original volume of about 261 million cubic yards, nearly half its sediment could have come from ero- 30 PLEISTOCENE GEOLOGY OF THE NORTHEAST ADIRONDACK REGION TABLE 3.—Upper Pleistocene and Holocene deltas of the northeast Adirondack region, New York Characteristics of drainage basin (above highest stand of Champlain Sea) Deltas (estimate volumes in millions of cubic yardS) Lithology1 Coveville stage2 Main stream (percent of total area of basin) River Maximum Drainage relief Length Slope area. Other Paleozoic Existing Loss by (ft) (miles) (ft (sq Anor- Pre- and Original remnants erosion per mile) miles) thmite cambrian younger Salmon and Trout (Franklin County) ..... 2,855 26 110 228 -_ 67 33 __ -- -- Trout ...................................... 2,030 16 127 47 __ 55 45 42 -- -- Little Trout ............................... 1,930 17 114 23 -_ 44 56 86 .- -- Chateaugay ................. 3 110 26 120 159 -_ 83 17 24 -- -- North Branch Great Chazy - 2,164 19 114 105 __ 13 87 -- -_ -- Park Brook ................ 780 8 98 27 __ __ 100 __ -- -— Gl'eat Chazy ............................ 3 340 26 128 91 _- 40 60 -_ .- -— Great Chazy and North Branch _____________ 3 340 26 128 196 __ 25 75 -- -_ -— Sarenac ....... _--_ - 4 472 68 66 589 34 63 3 48 28 20 Salmon (Clinton County) ___________________ 1,910 15 127 37 __ 48 52 141 125 16 Little Ausable _______________________________ 1,700 9 189 36 __ 45 55 __ -- -— Ausable .................................... 4,984 48 104 501 60 34 6 352 149 3203 1 Fisher and others (1962). '“Lake Iroquois" stage in St. Lawrence Valley (MacC'lintock and Stewart, 1965, pl. 1B). 3 81 to Little Ausable River and 122 to Ausable River. sion of the older delta. The Fort Ann delta of the Ausable River near Keeseville lost by erosion about 54 million cubic yards (table 3), but this figure is only about one-ninth of the estimated original vol- ume of the younger Champlain Sea delta down- stream from Ausable Chasm (446 million cubic yards). I assume, therefore, that most of the delta at the mouth of the chasm is built of debris that came from points farther upstream than the Cove- ville delta near Clintonville. The volumes of the deltas and the characteristics of their drainage basins are compared to see if there are any consistent relationships between them. The volume of a delta does not show a close relation to the length of the delta-building stream (fig. 25), ex- cept for the deltas of the Champlain Sea, which show a slight tendency to increase in volume as length of streams increases. This is readily understandable be- cause the older Lake Vermont deltas were built in part by melt-water streams, and thus the volumes of these deltas need have no relation to present drain- age basins. Of the six streams that built deltas into the Champlain Sea, the delta of the combined Salmon and Trout Rivers in Franklin County is large com- pared with those of the other five, whereas that of the Great Chazy River and North Branch is small. The large size of the Salmon-Trout River delta could be related to the large mass of glaciofluvial deposits upstream, a readily available source of delta-building material. Perhaps the small volume of the Great Chazy—North Branch delta is related to the large area of Potsdam Sandstone in the river’s drainage basin (75 percent). The lowlands of crystalline rocks in the Adirondack region are buried beneath a sandy glacial drift, whereas in areas of Potsdam Sandstone the drift cover is more stony and perhaps thinner and less extensive. As most of the bedload carried by the rivers probably was supplied by drift, it follows that the late-glacial rivers in areas of Precambrian rock may have carried a larger volume of sandy sediment than those in areas of Potsdam Sandstone. LAKE VERMONT The deposits and shoreline features of Lake Ver- mont in the lowlands from the Ausable River to the Canadian border record only the latter part of the history of this glacial lake. During most of the Cove- ville stage, the Plattsburgh area was beneath the ice sheet. COVEVILLE STAGE The history of the Coveville stage in the northeast Adirondack region opened with the deposition of the large delta in the Ausable River valley west of Clin- tonville during the Owls Head-Redford stand (ice- front position 2, pl. 1). The original volume of this delta is estimated at 352 million cubic yards (table 3). This figure is several times larger than that for the Saranac River delta (48 million cubic yards), although the Saranac is a somewhat longer stream than the Ausable. The Ausable River delta of Cove- ville age is also considerably‘ larger than that of the younger Fort Ann stage near Keeseville (261 million cubic yards). The explanation for the large volume DEPOSITS AND SHORE FEATURES OF LATE-GLACIAL WATER BODIES 31 TABLE 3.—Upper Pleistocene and Holocene deltas of the northeast Adirondack region, New York—Continued Deltas (estimated volumes in millions of cubic yards) Rate of delta forma- tion in Champlain Sea (assuming that it lasted about 1,500 years) Fort Ann stage Unit rate Lake (acre it per Rate Champlain Sea Champlain square mile (millions of of drainage cu yd per Existing Loss by Existing Loss by basin per yr) Original remnants erosion Original remnants erosion Original yr) galmon and Trout (Franklin County) ...... 529.0 441.0 88.0 382 ___ ___ ___ 0.692 0.255 rout ..... -__- ___- ___ -_- --- ___ --_ -_-- ___- Little Trout ........................... 6.5 ___- ___ --- --- ___ ___ ___- ___- Chateaugay ........................... 29.0 ---_ --- ___ ___ _-- __- ___- ___- North Branch Great Chazy -- ___- 10.4 5.7 4.7 ___ --- _-- --_ _--- ___- Park Brook _____________________ ___- 0.6 ___- --- --- ___ --- ___ _--- ___- Great Chazy ....................... ___- 3.3 _-_- _-- --- --- ___ ___ ___- -__- Great Chazy and North Branch ___- _-_- ___- ---_ --- 80 ___ _-_ _-- .169 .053 Saranac ......................... __-- 95.0 70.0 25.0 835 291 44 ___ .235 .223 Salmon (Clinton County) ................... 77.0 53.0 24.0 100 ___ ___ ___ 1.120 .067 Little Ausable _______________________________ -_-- ___- --_ 118 ___ -__ ___ 1.351 _--_ Ausable 261.0 207.0 54.0 446 239 207 ‘212 .367 .297 ‘Includes both the delta at month of river and the submerged delta near Wickham Marsh. of the delta near Clintonville during Coveville time is probably that the Ausable was augmented by over- flow from the Saranac drainage basin. The ice dam at the north end of Lake Vermont blocked the Champlain Valley near Keeseville, and the delta was built into an arm of the lake extending upvalley to Clintonville. The ice front gradually retreated north during late Coveville time to the vicinity of the Saranac River. The time of retreat corresponds to that involved in the Trout River- Moffitsville stand, the Malone-Schuyler Falls stand, and much of the building of the moraine near Cady- ville. Only one or two small beaches were formed during Coveville time in the area between the Ausable and Saranac Rivers; perhaps at this latitude the Cham- plain Valley was largely ice filled and only a narrow arm of the Coveville lake extended north between the ice and the edge of the uplands. During the retreat of the ice front from the Ausable to the Saranac, ice- marginal drainage flowed south along the ice edge where it rested against the uplands, cut channels in drift and in bedrock, removed the drift from large areas, and discharged into the lake, as described earlier under the Cadyville episode of deglaciation. The surface of the deltas at the mouths of the Sal- mon (Clinton County) and Saranac Rivers is irregu- lar, suggesting collapse and slumping after adjacent glacial ice had disappeared. That part of the Cove- ville stage which is represented by shore features in the Plattsburgh area probably lasted less than a hundred years. The delta east of Peasleeville at the mouth of the Salmon River (Clinton County) is large (141 million cubic yards) in relation to the length and size of the Salmon River (fig. 25). Perhaps the delta deposits came from the erosion of the large sandy kames in the valley upstream. The delta on the Saranac River east of Cadyville is small (48 million cubic yards), only about a third the size of the delta along the nearby and very much shorter Salmon River. The explanation may be that when the level of Lake Ver- mont dropped to that of the Fort Ann stage, the ice dam in the Champlain Valley had retreated to a line only a few miles north of the Saranac River. Thus, the river perhaps had time to build only a small delta into the Coveville lake. FORT ANN STAGE The level of Lake Vermont dropped nearly 100 feet, perhaps more than 150 feet, to that of the Fort Ann stage when the ice dam was a short distance north of the Saranac River, perhaps about at ice- front position 8 (pl. 1), that is, during the building of the moraine near Ellenburg Depot. Deltas began to be built at the lower lake level by the Ausable, Salmon, and Saranac Rivers, in part of material de- rived from older Coveville-stage deltas upstream. Along the Ausable River, the original volume of the Fort Ann delta at Keeseville is estimated at about 261 million cubic yards. Of this, 122 million cubic yards probably came from the erosion of the Coveville delta upstream near Clintonville (table 3), leaving about 139 million cubic yards of “new” sedi- ment to come from erosion in the drainage basin or 32 PLEISTOCENE GEOLOGY OF THE NORTHEAST ADIRONDACK REGION FIGURE Iii—Cobble and boulder grave] in Champlain Sea delta of the Saranac River. Massive topset beds exposed in borrow pit at apex of delta just east of Morrisonville. Clasts are subangular to slightly rounded; sand lenses suggest a faint horizontal stratification. from wasting glacial ice. The volume of “new” sedi- ment is about 2.5 times less than that deposited by the Ausable River in Coveville time. Presumably this smaller amount of sediment reflects a decrease in streamfiow (no contribution from the Saranac drain- age basin) and in volume of melting ice upstream. It is not clear whether this decrease in amount of sedi- ment argues for or against Craft’s (1969) suggestion that glacial ice was still present in the high peaks region during Lake Vermont time. The original volume of the Fort Ann delta of the Saranac River, 95 million cubic yards, is nearly twice that of its Coveville-stage delta. Both are smaller than the Fort Ann delta of the Ausable River. Because the volume of melting ice in the drainage basin was probably decreasing, perhaps to zero, in Fort Ann time, the increase in size of the Saranac delta suggests deposition over a much longer period of time; that is, the Fort Ann stage lasted twice and perhaps several times as long as that time interval during Which the Saranac River built its Coveville delta. The Saranac River delta is only slightly larger than that of the Salmon River (Clin- ton County). It is not clear why the Lake Vermont deltas of these two rivers, so different in length, are so similar in size. Lake level may have risen during a part of Fort Ann time. The presence of Fort Ann beaches on Flat Rock, as mentioned earlier, suggests a rise in lake level after the bedrock had been cleaned off. Along the Ausable River the top of the Fort Ann delta, at Keeseville, is about 60 feet below the highest stand of the Fort Ann stage (Denny, 1967) ; this delta was built, therefore, when lake level was below the maxi- mum level of the Fort Ann stage. Along the Salmon FIGURE 20.—Crossbedded pebbly sand capped by pebble to cobble gravel. Probably topset beds in Fort Ann delta of the Great Chazy River at Altona. Prominent soil tongues at top of bank. DEPOSITS AND SHORE FEATURES OF LATE-GLACIAL WATER BODIES 33 and Saranac Rivers the relation of the top of the Fort Ann deltas to the highest lake stand is not clearly defined. The northward retreat of the ice front in the Champlain Valley (from ice-position 8) uncovered the lower course of the Great Chazy River and its North Branch, and they began to build small deltas into the glacial lake. The tops of these deltas are also about 50 feet below the highest stand of the Fort Ann stage. Delta building by the Ausable and Great Chazy Rivers could have taken place during either rising or falling lake level or both. It is easier, however, to account for the absence of delta deposits at the maxi- mum stand and for the occurrence of beaches on Flat Rock by assuming that delta formation was largely completed before a 50-80-foot rise in lake level near the end of Fort Ann time. In profile, the Fort Ann beaches form a belt 50—75 feet wide, the top of which marks the maximum stand of the stage (Denny, 1967, 1970). Whether the beaches in the lower part of the belt were formed before the rise in lake level or during the subsequent fall from the highest stand is unknown. Presumably the rise was in response to a rise of the outlet of Lake Vermont near Fort Ann, NY. The extensive beaches at the maximum stand of the Fort Ann stage near Cannon Corners were formed of material washed off the adjacent areas of bare rock along English River (pl. 6). Perhaps the rise in the Fort Ann lake took place during the wash- ing of the drift from the bare-rock areas near Eng- lish River. In Franklin County, during Fort Ann time, deltas formed at the mouths of the Salmon, Trout, and Chateaugay Rivers where they entered glacial Lake Iroquois (pl. 1), which drained east by way of the gorge near Covey Hill. In time, the ice front retreated to the northeast off Covey Hill, and the level of Lake Iroquois dropped to that of Lake Vermont in the Champlain Valley. During the lowering of the Fort Ann lake prior to the incursion of the sea, there may have been a short interval when the lake in the Champlain Valley drained northeast between the retreating ice front and hills southeast of the St. Lawrence River (Wag- ner, 1969). Ultimately, drainage was opened to the Atlantic Ocean by way of the St. Lawrence Lowlands northeast of Montreal, and marine waters invaded the Champlain Valley. This invasion dates from about 12,000 years B.P. (McDonald, 1968; Prest, 1970). The duration of the Fort Ann stage is unknown, but it probably was only a few hundred years. Prest (1970) , in his reconstruction of glacial lake phases in the Great Lakes region and the St. Lawrence Low- lands, shows the Fort Ann stage beginning about FIGURE 21.——Foreset beds of medium well-sorted sand in delta of the Champlain Sea about 1 mile southwest of Port Kent. Exposure is about 15 feet high. Top of bank has been stripped. 34 44'40’ 35, indicates no data Mixed sand and gravel Coarse sand 0 1 2 MILES l___l—J Medium sand H a 6&3é3iiié'éEEé'3f Fine sand 0 a (2 Lake Vermont «"3 E c. g: .— au E ,2 ‘5 8;; a ————.._-..__ Very fine sand 3-; > g Fort Ann stage of to clay g 3 .: Lake Vermont .2» 3 2 m s Alluvium and Champlain Sea 44'30’ — PLEISTOCENE GEOLOGY OF THE NORTHEAST ADIRONDACK REGION 73°35’ . 45‘; _.r XX"; ' . “022:1” 3 an»: ‘1 A a: $57 \ L4 .9 _ Unpatterned area ground moraine -; TXII'B'IWfsVE’zB ,1!- “’3’ ~ {- w w A] .1!" {If 30' 73'25’ 1:34:13? _ , x J j 1,. . ’vj/ZRRCJIZ/Efi, saw an}? . ‘ ‘5; Island A FIGURE 22.—-Grain size and sorting of late-glacial lake, stream, and marine deposits in the Plattsburgh clay from Denny (1967, 1970). A, Areal distribution of DEPOSITS AND SHORE FEATURES OF LATE-GLACIAL WATER BODIES 73°35’ 30, 73925] 44-40' Burnt Hill':._ 35' — . Trask sortmg coefficient 1.6—1.8 >1.8 X Sample locality for mechanical analysis 44'30’ y— EXPLANATION \ Highest stands of Lake Vermont and Champlain Sea CW 1 Valcour Island 2 MILES Coveville stage of Lake Vermont Fort Ann stage of Lake Vermont Champlain Sea area. Based on rapid mechanical analysis B of sand made in the field. Distribution of gravel and silt and textural classes. B, Areal distribution of Trask sorting coefficient. 35 36 oo o l \J o I 01 o I 01 O I b O EXPLANATION ‘ (A) 0 GB Above Lake Vermont (16 analyses) _ O Coveville stage (15 analyses) FREQUENCY DISTRIBUTION, IN PERCENT N O G) Fort Ann stage (31 analyses) a. Champlain Sea (58 analyses) E Sand dunes (8 analyses) I I I I | I I I l l I I 0 0.2 0.4 0.6 0.8 ESTIMATED MEAN GRAIN SIZE, IN MILLIMETERS 1.0 \l o 8 8 I I I Ch 0 | 4s 0 I FREQUENCY DISTRIBUTION, IN PERCENT w 01 o o l I N O I ,_. O I 0_l_l I/ /I 1.20 1.40 I J I I l I I | I | I 1.60 TRASK SORTING COEFFICIENT B FIGURE 23.-—Grain size and sorting of late—glacial sands. Cumulative curves showing the range in estimated mean size (A) and in Trask sorting coefficient (B) of sands of Lake Vermont and the Champlain Sea. Data from a few analyses of dune sand and of glacial sand above the highest stand of Lake Vermont are included for compari- son. 12,500 to 12,400 years B.P. (Prest, 1970, fig. XII— 16f). The ice sheet left Covey Hill, and the lakes in the St. Lawrence and Champlain Valleys became con- fluent shortly before 12,000 years B.P. (Belleville- PLEISTOCENE GEOLOGY OF THE NORTHEAST ADIRONDACK REGION Fort Ann phase, Prest, 1970, fig. XII—16h). The Champlain Sea invaded the St. Lawrence Lowlands and the Champlain Valley by about 12,000 years B.P. (Prest, 1970, fig. XII—16i). CHAMPLAIN SEA The deposits of the Champlain Sea are indistin- guishable lithogically from those of Lake Vermont. The marine deposits in places contain the remains of salt- or brackish-water organisms, largely mollusks. Localities where fossils have been found are shown on the published geologic quadrangle maps (Denny, 1967, 1970). Plate 1 shows that the marine deposits form a belt at lower altitudes than those of Lake Vermont. The principal streams built deltas into the Champlain Sea. Delta building continued along most of the large rivers until gradual rise of the land caused the sea to drain down to the level of present- day Lake Champlain. The Great Chazy River is an exception because it ceased to build a delta when sea level had dropped to a point about 100 feet above the present level of Lake Champlain. North of the Saranac River, beaches are common near the marine limit. Along the Saranac River north of Morrisonville and along the Salmon River (Frank- lin County) at Malone, the marine limit is repre- sented by steep bluffs cut in the sand of the older Fort Ann delta by wave action (MacClintock and Stewart, 1965, fig. 22b)._Thcse cliffs resemble the present-day blufi“ along the shore of Lake Champlain just south of the mouth of the Ausable River. In the area north of the Saranac River, beaches continued to form during the gradual lowering of the sea level until the shoreline reached a point about 200 feet above the modern lake. AGE The Champlain Sea lasted for about 1,500 years. It came into existence about 12,000 years B.P. This is the figure given by McDonald (1968) and by Elson (1969). Mott (1968) listed 27 Champlain Sea radio-- carbon dates; 25 of them are from shell material. The oldest date, 11,800: 160—180 years B.P., was meas- ured on samples from two localities (Geol. Survey Canada locs. GSC—505 and GSC—588). Prest, in his reconstruction, gives a range of 12,000 to 11,800 years B.P. for the sea’s birthday (Prest, 1970, fig. XII—16h, i). The Champlain Sea withdrew from the Champlain Valley about 10,500 years B.P., perhaps as late as 10,000 years B.P. Elson (1969) gives a figure of 10,000 years B.P. for the close of the Cham- plain Sea episode in Quebec Province, and Prest DEPOSITS AND SHORE FEATURES OF LATE-GLACIAL WATER BODIES 37 m 1.0 . l I E ' . E 0.5 -— x _ j x =, x o . X o . E O x . o z " . —_ e e W O u x o gt? @ e e N G a O O 0 g 0.1— ' — < a: 9 e (5 <2: 0'05_ EXPLANATION _ Lu 0 Delta of Saranac River 2 in Champlain Sea a a Delta of Satmon River '— in Lake Vermont < E at Delta of Ausable River 5 in Fort Ann stage Lu 1 of Lake Vermont 0.01 ' I ' ' O 1 2 3 4 5 DISTANCF FROM APEX OF DELTA, IN MILES FIGURE 24.—Grain size of delta sand and distance from apex. Semilogarithmic scatter diagram showing the relation between the estimated mean size of sand at sample sites on ancient deltas of the Saranac, Salmon, and Ausable Rivers and the distance of these sites from apex of delta. places the withdrawal of the sea from the Champlain Valley between 10,300 and 10,000 years B.P. Of the 27 Champlain Sea radiocarbon dates listed by Mott (1968), only five are less than 10,500 years .B.P. (10,200 to 10,450). A sample of shells from Champlain Sea deposits exposed in a borrow pit about 1.5 miles west of Chazy gave a radiocarbon age of 10,560:350 years B.P. (W—1109, Ives and others, 1964). The sample came from a pebbly sand at the base of about 10 feet of boulder and cobble gravel on the crest of the broad ridge named the Ingraham esker by Woodworth (1905a). Presumably these shells, largely mollusks, lived on the ridge during the decline in sea level. The shell locality is at an altitude of about 210 feet, nearly 250 feet below the marine limit. RATE OF DELTA FORMATION Using the estimates of volume of the Champlain Sea deltas and assuming that the embayment lasted for about 1,500 years, it is possible to calculate rates of delta formation for several rivers emptying into the Champlain Sea. The rates range from 53,000 cubic yards per year for the Great Chazy River and North Branch to 297,000 cubic yards per year for the Ausable River (table 3). In terms of the size of the drainage area upstream from the delta, the rate for the Great Chazy and North Branch is 0.169 acre-feet per square mile per year and for the Ausable River 0.367 acre-feet per square mile per year. ENVIRONMENT When the sedimentation rates have been calcu- lated, what do they mean in terms of the environ- ment of Champlain Sea time? Was the environment strictly periglacial, a time of increased bedload in the streams because of strong frost action and (or) sparse vegetation on slopes, and of increased stream- flow because of ice melting or high precipitation? Or was the environment nonglacial, much like the present? Schumm and Hadley (1961) found a good corre- lation between the rate of sedimentation in small reservoirs in the western United States and the relative relief of the drainage basin supplying the sediment. To compare the rate of formation of the Champlain Sea deltas with the rate of sedimentation in the small reservoirs, I have plotted (fig. 26) rate of delta formation against the relative relief of the drainage basin. In this comparison, I ignored the suspended load carried by the rivers into the deeper parts of the sea. The rate of delta formation, ex- pressed in acre-feet per square mile per year, in- creases with the relative relief of the drainage basin, expressed as maximum relief divided by main stream length (table 3), in the same manner but at a slight- ly faster rate than does the sedimentation rate in the small drainage basins of the western United States (Wyoming to Arizona). The increase in sedimenta- tion rate with relative relief is a common feature of drainage basins and is probably related to increased average slope. The deltas of the smaller streams have the higher rates and higher average slopes, and the unit rate of delta formation decreases with in- crease in drainage area of delta-building stream (fig. 27). In both examples, the unit rate of delta forma- tion for the Great Chazy and North Branch is small, suggesting again the influence of the Potsdam Sand- stone on the size and, therefore, on the rate of forma- tion of the deltas along the two streams. The comparison between the Champlain Sea deltas and the sediment in the small reservoirs suggests that delta formation was conditioned by the same 38 PLEISTOCENE GEOLOGY OF THE NORTHEAST ADIRONDACK REGION 1000 I I I I I I “ FA ‘ O / ~ \ — / s \ \ \ ‘ .\ x \ \ — / S \ \ \-\ 0 c / \ x \ s / \ / — \ ,4 / \ FA / / \ \ _/ / / / / ._ / _ / / / C / / / 3/ / (El S / / 8 \ S / n: 100 *3 C 5 — \ s / FA — U _ FA \\ <9 / _ a _ FA \ / _ LL _ _ O _ C _ 3 _ A C _ E I EXPLANATION 5 FA DELTA-BUILDlNG STREAMS <’ _ 8’ T c x A 0 ® Ausable River Trout River LL 0 — _ E E ' 3 Little Ausable Saranac River 9 River 2‘ ® g FA © Chateaugay River 3 10 A Salmon River _ O — (Clinton County) E _ — Great Chazy River — _ O _ . © _ Salmon and Trout RIvers _ _ (Franklin County) Great Chazy RIver _ and North Branch __ I A — FA Little Trout North Branch Great _ IX River Chazy River — 0 FA fl Coveville stage Fort Ann stage _ s _ Champlain Sea 1 0 l l l l l l ' 0 10 20 30 40 50 60 70 LENGTH OF DELTA-BUILDING STREAM, IN MILES FIGURE 25.—Up~per Pleistocene deltas and their source areas. Semilogarithmic scatter diagram showing the relation between the original volume of a delta and the length of the delta-building stream. Deltas of the Champlain Sea are surrounded by a dashed line. DEPOSITS AND SHORE FEATURES OF LATE-GLACIAL WATER BODIES 39 variables that control modern sedimentation in parts of western United States rather than by climatic factors operating during deglaciation. Insofar as such a comparison is valid, it also suggests that the environment of the Champlain Sea episode was more nonglacial than periglacial. But was the environment much like that of the present? The modern delta of the Ausable River suggests that sedimentation rates were much greater in Champlain Sea time than they are at present. AUSABLE RIVER DELTA IN LAKE CHAMPLAIN The Ausable River has built a crow’s-foot delta in Lake Champlain (fig. 28). If the lake has been in existence for nearly 10,000 years, an estimate of the volume of the modern delta Will yield a rate of delta formation for post-Champlain Sea (postglacial) time. On the basis of depth curves and soundings, the volume of the modern delta is estimated at about 212 million cubic yards (table 3). The rate of delta formation would thus be about 0.021 million cubic yards per year (0.026 acre-feet per square mile per year), suggesting that the bedload of the Ausable River is now very much less than it was in Cham- plain Sea time (figs. 26, 27). Delta building during Champlain Sea time may well have been much greater than at present. How- ever, the size of the modern Ausable River delta is not dependent on the length of post-Champlain Sea time; rather, the delta appears to be a young and unique feature. Several large streams empty into Lake Champlain: the Great Chazy, Saranac, Ausable, Winooski, Lamoille, and Missiquoi Rivers, yet the Ausable is the only one that has built a crow’s-foot delta graded to the modern lake. Even the Saranac River has no conspicuous delta at its mouth, in spite of the fact that the topography and geology of its drainage basin are similar to those of the Ausable River (table 3), and even though it discharges into Cumberland Bay, a shallow arm of Lake Champlain, whereas the Ausable River discharges into deep water. The modern delta of the Ausable River in Lake Champlain was not formed until the river’s course had been changed by piracy. The piracy could have taken place several thousand years after Lake Cham- plain came into existence. Thus, the modern delta may have been built rapidly. The history of the delta is interpreted to be as fol- lows. The Ausable River first built a delta into the Champlain Sea when the sea stood near its highest 1°-° _ I T I I I l ‘ a: < i _ J a: Lu Q. m =' 2 Eli) © < 1.0 D — _. O' (I) -— _ — o — .— "‘ ‘ Lu Lu u. _ _ uh I O _ _ < X E Z‘ _ _ Q I'- < 2 n: E —‘ _. < @ EXPLANATION a DELTA-BUILDING STREAMS D l X 0 Ausable River IL" 0.1 _ E - Uttle Ausable River - _ . _. I- Saranuc Rive! '- ._ © _ Salmon River _ (Clinton County) .a 0 - Salmon and Trout Rivers — (Franklin County) _ AusabIe River deIu @ _ I" L‘“ Champlain Great Chazy River and North Bunch 0.02 I I l I | I 0.0 0.02 0.04 0.06 0.08 MAXIMUM RELIEF IN FEET RELATIVE RELIEF OF DRAINAGE BASIN _———— MAIN STREAM LENGTH IN MILES FIGURE 26.—Rate of formation of Champlain Sea deltas and relief of drainage basins. Semi-logarithmic scatter diagram showing the relation between the rate of delta building and relative relief of the drainage basin above apex of each delta. Champlain Sea assumed to have lasted about 1,500 years. Line representing a generalization from data on mean annual sediment yield for small drainage basins in western United States shown for comparison (from Schumm and Hadley, 1961, fig. 1). Point representing rate of formation of Ausable River delta in Lake Champ- plain assumes that the lake has been in existence for about 10,000 years. 40 PLEISTOCENE GEOLOGY OF THE NORTHEAST ADIRONDACK REGION 100. I IIIIIII I IIIIII D: — _ < m _ _ >— n: _ ._ Lu 0. LLI _ _ :‘ E 5 <3): © 0' 1 0 _ \ _ U) __ _ o: — _ LLJ o _ D. _ I- _' EXPLANATION — E u. — DELTA-BUILDING STREAMS + u'J II _ o + < AusabIe River . E ._ _ 2“ ° . <9 0 Saranac River E g E 0.1 — . D: ’— Salmon RIver _ .8 : (Clinton County) : E — O — d — Salmon and Trout Rivers —- D _ (Franklin County) _ LI. 0 E A bI R' d It — . . usa e Iver e a _ Iu—J LIttIe Ausable River In Lake Champlain < o: _ 9 + _I Great Chazy River and North Branch 001 I IIIIIII I IIIIIIl 10 100 1000 DRAINAGE AREA, IN SQUARE MILES FIGURE 27.—Rate of formation of Champlain Sea deltas and size of drainage basins. Logarithmic scatter diagram showing the relation between the rate of delta formation per square mile of drainage area and the drainage area upstream from the delta. Champlain Sea is assumed to have lasted about 1,500 years. Point representing Ausable River delta in Lake Champlain assumes that the lake has been in existence for about 10,000 years. point at an altitude of about 360 feet (fig. 28). Ausable Chasm was not yet in existence, for its rim ranges in altitude from about 250 to 350 feet. The surface of the delta above the chasm slopes toward the lake to an altitude of about 260 feet. Sea level gradually lowered. The river began to cut the upper part of the chasm and to erode part of its older delta. A new delta was built, and the chasm was eroded to a depth of as much as 60 feet. The new delta ranged in altitude from about 260 feet at its apex to about 180 feet near the present shoreline. The deposits of the younger delta overlap those of the older one; the two can be distinguished only on the basis of their topographic position and form. A further drop in water level, down to the present lake level or below, caused the river to cut down to an altitude of about 200 feet, where erosion was checked for a time by a bedrock sill about half a mile east of the mouth of the chasm (point W, fig. 28). Upstream from the sill, the river developed a meandering course and by cut and fill spread coarse sand and gravel on top of delta sand. Remnants of this meandering channel are still preserved upstream from the rock sill. At this time, the floor of the chasm was cut down to Within about 60 feet of its present level. The Champlain Sea drained from the Champlain Valley, and the modern lake came into existence. The Ausable River did not at once lower its channel to the new lake level because the river was held up by the rock sill (point W). Downstream from the sill, the river eroded a deep channel in the delta deposits, forming a series of rapids on bedrock, and flowed into a basin about half a mile Wide, now occupied by Wickham Marsh. A delta was built east of the marsh out into the lake. The outline of this delta below the surface of the lake is shown by the 12-foot-depth curve in figure 28. Presumably, the now-submerged delta near Wickham Marsh is built primarily of material eroded by the river downstream from the rock sill. The rate of sedimentation on the delta near Wick- ham Marsh decreased once the rapids below the rock sill had been formed. Waves and currents began to erode the lower end of the older Champlain Sea delta deposits and form a cliff that today is about 100 feet high and is perhaps a quarter of a mile west of its original position. The modern delta at the river’s mouth laps against the northern part of this cliff (near point X, fig. 28) and therefore that part of the modern delta is assumed to be younger than the cliff. Wave erosion formed the cliff and destroyed all of the subaerial part of the delta east of Wickham Marsh. The time required to erode this delta and form the cliff along the shore cannot be estimated on the basis of available information. Certainly some of the material was removed after the Ausable River had assumed its present course into the lake, and this diversion must have taken place before peat began to accumulate in Wickham Marsh. The diversion of the Ausable River from its aban- . doned channel at an altitude of about 190 feet to its present course was caused by piracy. Dry Mill Brook joins the Ausable River from the north at the head of the modern delta. The brook probably lowered its bed rapidly when Lake Champlain came into exist- ence. A gully on the south side of the brook, now destroyed, cut into the deposits of the Champlain Sea DEPOSITS AND SHORE FEATURES OF LATE-GLACIAL WATER BODIES 41 73.30 73°25’ 44°35’ \ I T \ \ lZ-foot-depth curve ><124 \ r X 7 vs 224 {A O “ X186 V 2?: X123 '1 r :> X171~ 6?. X20 2 _. ><25 f’ 169 l \ ><72 - APort Kent X127 5 | . : l / 1 l I MEAN LAKE LEVEL 95 FEET l X125 44‘30’ EXPLANATION E: Flood-plain and delta deposits of Ausable River Sand, gravel, silt, or clay Swamp deposits Till, bedrock, locally sand, gravel, silt, or clay D ELTAS Above Ausable Chasm Altitude at apex about 860feet. Built into Champlain Sea At month of chasm Apex at altitude of about 260 feet, in part bevelled by Aus- able River at altitude of about 200feetr Built into Champlain Sea 09°00 °° 0 0°: ”on°°n° o°°°o°g At mouth of river Built inw Lake Champlain 260— Contours drawn on restored surface of delta deposits of Champlain Sea. Contour interval 20feet _l_.l_.l—L—hl-— Inferred highest stand of Champlain Sea —> Abandoned channel of Ausable River Altitude about 190feet X127 Sounding, in feet Datum is mean low lake level, 92.5 feet W X Y Localities discussed in text FIGURE 28.—Ausable River delta in Lake Champlain. Geology generalized from Denny (1967). Base from U.S. Geo- logical Survey Plattsburgh 15-minute quadrangle, 1956. delta near the mouth of the chasm. As the gully into the gully and down Dry Mill Brook to the lake. deepened, its head approached the bank of the The river was then diverted into Dry Mill Brook and Ausable River (near point Y, fig. 28), where the cut down rapidly into the underlying sandy deposits, riverbed would have been about 100 feet above the as there was no rock sill at the point of capture to mouth of the gully. hinder erosion. The channel above Wickham Marsh The river, perhaps in time of flood, overtopped its was abandoned. north bank (near point Y) and flowed north down The river cut down rapidly into the older deltaic 42 PLEISTOCENE GEOLOGY OF THE NORTHEAST ADIRONDACK REGION deposits, and the chasm has been gradually deepened to its present level. The valley at the mouth of the chasm is about 60—80 feet deep and about a quarter of a mile wide. The material removed in the erosion of this valley was swept into the lake to form the modern delta. The volume of material removed is estimated to be about 144 million cubic yards, and the volume of the modern delta at the mouth of the river, about 133 million cubic yards. The delta of the Ausable River in Lake Champlain is unique. It owes its development to a stream diver— sion that caused rapid, perhaps catastrophic, erosion of older deltaic deposits and deposition of sand in the lake, in part on top of older delta sands. Near Ausable Point, the shore of the delta is now being eroded back (westward), and sand is piled by storm waves among the trees along the shore. This analysis of the modern delta of the Ausable River and the absence of similar deltas along the other large rivers emptying into Lake Champlain suggest that sedimentation during the Champlain Sea episode was greater than at present. Although the marine deltas appear to be the result of runoff from the drainage basin upstream, the rate of sedi- ment (bedload) transport was considerably greater than now. The environmental difl’erences that caused the change in rate of sedimentation are unknown. It is reasonable to suppose that the climate was still cooler than the present and that the forests had not yet completely covered the landscape. The fossil record supports this interpretation (Terasmae, 1959; Davis, 1965; LaSalle, 1966; Terasmae and LaSalle, 1968) . For New England, Davis suggested that dur- ing the interval 12,000—10,500 years B.P. “. . . the vegetation may have resembled park-tundra, or, alternatively, spruce—oak woodland similar to modern vegetation near the prairie in Manitoba” (Davis, 1967, p. 11). The fossil assemblage in the Champlain Sea deposits that overlie the Ingraham esker (Wood- worth, 1905a) suggest shallow cold marine water (Denny, 1972). J. E. Hazel (written commun., 1971) feels that the ostracode assemblage indicates sub- frigid or frigid climatic conditions (that is, aver- aging about 10°C or colder in shallow water during the warmest month). TILT OF WATER PLANES The shorelines of the late-glacial water bodies in the Champlain Valley rise northward (Chapman, 1937 ; Farrand and Gajda, 1962; Wagner, 1969). In the northern part of the valley on the New York State side near the Ausable River (Denny, 1967, 1970), the Lake Vermont deposits range in altitude from about 490 feet to 670 feet, whereas near the Saranac River the range is from about 560 feet to 720 feet. The upper limit of the marine deposits ranges from about 360 feet near the Ausable River to about 510 feet near the International Boundary. The beaches and deltas of these two belts were formed in extensive water bodies, and the northward rise demonstrates that the region has been warped upward toward the north since the shore features were formed. As mentioned earlier, the correlation of the shore features throughout the Champlain basin is largely based on their location and altitude as shown on north-south profiles. The profiles for the areas mapped in detail show a line representing the in- ferred position of the highest stand of the Fort Ann stage of Lake Vermont and of the highest stand of the Champlain Sea. The inclination of these inferred water planes amounts to about 4.5 feet per mile for the top of the Fort Ann stage and about 4.9 feet per mile for the top of the Champlain Sea. Theoretically, the tilt of a marine-water plane cannot be greater than that of an older lake plane and might well be less. Thus, these values of tilt have a possible error of at least 0.2 foot per mile. The north-south profiles for the New York side of the northern Champlain basin furnish only a two- dimensional picture of the tilt. A three-dimensional model can be constructed by comparing the north- south profile showing the inferred water planes for New York with a similar north-south profile for Ver- mont (W. P. Wagner, written commun., 1970). A point at a given altitude on the New York profile is several miles south of the latitudinal position of a point at the same altitude on the Vermont profile. It is a simple problem in plane geometry to determine that the water planes have been tilted upward in a direction about N. 10°—15° W. along an axis trending at right angles thereto. The 10°—15° deviation of the true direction of tilt from the north-south orientation of the published profiles (Denny, 1967, 1970) increases the amount of tilt by only about 0.1—0.2 foot per mile, probably within the error of measurement of the individual shore features. SUMMARY AND CORRELATION NORTHEAST ADIRONDACK REGION The late-glacial history of the Northeast Adiron- dack region began with the removal of the ice sheet SUMMARY AND CORRELATION 43 from the southwest corner of the region and the building of massive outwash plains by southwest- flowing streams (ice-front position 1, pl. 1). There is no way to date this episode, but it probably followed the Luzerne readvance of the Hudson-Champlain ice lobe on the southeast side of the Adirondacks (fig. 14). The Luzerne readvance, originally described by Woodworth (1905b, p. 139), has been dated by Con- nally and Sirkin (1971) at about 13,200 years B.P. These authors believe that this readvance antedates the Quaker Springs and Coveville stages of Lake Vermont that were coextensive with lakes in the Hudson Valley (LaFleur, 1965). The Bridport read- vance (Connally, 1970) in the southern part of the Champlain basin entered into the Coveville-stage lake. The late-glacial history of the northeast Adiron- dack region can be said to end with the invasion by the Champlain Sea about 12,000 years B.P. I assume that the carbon-14 dates for the Champlain Sea de- posits are essentially correct. Although they are based largely on pelecypod shells, algae material (seaweed) from marine deposits near Ottawa gives a carbon-14 date consistent with that of shells found immediately above and below the algae material (Mott, 1968) . There are no carbon-14 dates from which to esti- mate the age or duration of any episode or stand in the northeast Adirondack region. The ages given here (table 1) are based merely on a uniform rate of ice-front retreat of about 600 feet per year, from the line of the Luzerne readvance north to a point in the St. Lawrence Lowlands about 25 miles south of Montreal, assumed to be equivalent to the Drum- mondville moraine (LaSalle, 1966, p. 98, fig. 3). Schafer (1968) suggested an average rate of retreat for northern New England of at least 1,000 feet per year. My reconstruction (fig. 14) makes the Loon Lake episode (ice-front position 1) contemporaneous with the Highland Front moraine (Gadd, 1964). Many workers believe that Lake Iroquois, in the Ontario basin, came into existence about 12,100— 12,000 years B.P. (Karrow and others, 1961; Gold- thwait and others, 1965; Calkin, 1970) and that it drained about 11,000 years B.P. Prest (1970, fig. XII—16f) suggested slightly earlier dates—that Lake Iroquois came into existence about 12,500 years B.P. and lasted, with gradually declining levels (“post- Iroquois lakes”), until shortly after 12,000 years B.P. My reconstruction (fig. 14) follows that of Prest, using the older dates for Lake Iroquois. This recon- struction avoids the problem of how to hold Lake Iroquois within the Ontario basin when marine waters were present in the St. Lawrence Lowlands to the northeast. The building of the outwash plains near Duane . Center and Loon Lake (ice-front position 1, pl. 1) began about 12,700 years B.P. The edge of the ice sheet gradually retreated to the northeast, ice- dammed lakes formed in north- and east-draining valleys, and streams were diverted across divides, washing debris from large areas of bedrock and carving extensive channels. The reconstruction of ice-front positions suggests that the retreat to posi- tion 2, extending from Owls Head southeast to Clintonville and including the diversion of the Sara- nac River into the Ausable River, represents a short interval in late Coveville time, perhaps about 12,600—12,500 years B.P. The ice front in the Cham- plain Valley retreated north of the Saranac River, Lake Vermont began to drain south by way of the outlet near Fort Ann, NY. (fig. 14), and lake level dropped to that of the Fort Ann stage. The highest stand of Lake Iroquois in Franklin County, east of Malone (pl. 1), would, under my reconstruction, date from about 12,500 to perhaps 12,200 years B.P. Because the stand is at an alti- tude of about 1,000 feet, it appears to have been graded to the Covey Hill channel and, therefore, may well date from the Covey Hill episode, that is from perhaps 12,400 to just prior to 12,200 years B.P. The north and northeast retreat of the ice front was probably interrupted by minor oscillations, dur- ing which the moraines near Ellenburg Depot and near Cadyville were built. The moraine building may have taken place about 12,500—12,400 years B.P. During possible later oscillations, drift was removed from areas northwest of West Chazy, and the level of Prest’s “post-Iroquois” lake was IOWered, prob- ably in several stages, from that of the Covey Hill channel (1,010 ft) to that of Lake Vermont (740 ft) ; these later oscillations were part of the Covey Hill episode. The retreat of the ice front from Covey Hill into the St. Lawrence Lowlands caused the final lowering of the level of Prest’s “post-Iroquois” lake to that of Lake Vermont in the Champlain Valley. The merging of the lake waters north of Covey Hill probably took place only a short time, perhaps only 200 years, before the lowering of Lake Vermont and the inflow of the marine waters of the Champlain Sea. Wagner (1969) suggested that there was a time, just before the marine invasion, when the Fort Ann outlet ceased operation and the ice-dammed lake in the Champlain basin drained northward. The north- ward-draining lake he named Lake New York. 44 PLEISTOCENE GEOLOGY OF THE NORTHEAST ADIRONDACK REGION The northeast trend of the moraines in southeast Quebec (fig. 14) and the dates for the marine inva- sion suggest, as noted by McDonald (1968, p. 675), that the front of the main mass of the ice sheet retreated northwest across the St. Lawrence Valley. The merger of the ice-dammed lakes in the St. Lawrence and Champlain Valleys probably occurred shortly before the advent of the Champlain Sea, that is, about 12,200 years B.P. (Prest, 1970, fig. XII— 16h). Carbon-14 dates suggest that the submergence lasted about 1,500 years. Uplift of the land closed the connection to the ocean, and Lake Champlain came into existence about 10,500—10,000 years ago (Prest, 1970, fig. XII—16o). ST. LAWRENCE LOWLANDS IN NEW YORK STATE The late-glacial history of the St. Lawrence Low- lands in New York State has been studied in detail by MacClintock and Stewart. In excavations for the St. Lawrence Seaway (fig. 14; MacClintock and Stewart, 1965, p. 81—95, figs. 25—33), a fivefold strati- graphic sequence was demonstrated, from oldest to youngest as follows: (1) A lower Malone Till de- posited by ice from the northeast as indicated by striae on bedrock beneath the till; (2) an upper Malone Till with a fabric indicative of ice from the northeast but interbedded with sand, gravel, silt, and varved clay; (3) Fort Covington Till with a fabric indicating ice movement from the northwest; in places this till overlies bedrock with striae indicat- ing ice movement from the northwest; (4) varved lake clay; and (5) fossiliferous marine clay, silt, and sand. These authors interpreted the sequence as follows: An advance of Malone ice to the southwest followed by an “. . . oscillatory waning of Malone ice while standing in waters of an ice-dammed lake” (Mac- Clintock and Stewart, 1965, p. 87). Next, the Fort Covington ice advanced toward the southeast. The waning of the Fort Covington ice was followed by an ice-dammed lake that later was replaced by the Champlain Sea. Terasmae (1965, p. 35, pl. 5) sug- gested that there may have been a minor post-Fort Covington advance from the north. In northern New York and also in the Champlain Valley in Vermont (Stewart and MacClintock, 1969, p. 177), these authors postulated a period of emer- gence between deposition of the varved clays (unit 4) and deposition of the marine clays (unit 5) ; that is, a period of dry land between the withdrawal of the fresh-water lake (Fort Ann stage of Lake Vermont) and the incursion of the sea. As far as I am aware, no evidence has been found elsewhere to suggest that the St. Lawrence Lowlands were completely drained of ice-dammed lakes prior to the invasion of the Champlain Sea (Terasmae, 1965, p. 34; McDon- ald, 1968, p. 674) . MacClintock and Stewart traced their twofold di- vision (Fort Covington and Malone) throughout the St. Lawrence Valley in New York State by studying the fabric of the tills (MacClintock and Stewart, 1965, p. 140; see also, MacClintock, 1959). The Fort Covington Till extends up the south slope of the St. Lawrence Lowlands almost to the foothills of the Adirondacks (fig. 14). Farther south, the drift is Malone Till. Two and a half miles northeast of Malone, these authors found gray Fort Covington Till separated by 2 feet of sand and silt from the underlying red-brown Malone Till. MacClintock and Stewart traced the southern limit of their Fort Covington Drift (fig. 14) northeast to Covey Hill and south into the Champlain basin. They suggested that the Ingraham esker (Woodworth, 1905a), a narrow ridge of glaciofluvial deposits ex- tending from Ingraham northward to a point west of Chazy (pl. 1), is actually a belt of ice-marginal kames built along the western edge of a lobe of Fort Cov- ington ice. I favor Woodworth’s interpretation of an ice-channel filling (Denny, 1972). In Vermont, the eastern part of the Fort Covington ice lobe, according to Stewart and MacClintock (1969, 1970), deposited their Burlington Till (fig. 14) that has a northwest fabric. My reconstruction of the ice-front positions near Covey Hill is not far different from that of MacClin- tock and Stewart (1965, fig. 19). Their Fort Coving- ton Drift border crosses Covey Hill along ice-front position 14 (pl. 1). The westerly trend of the streams that run diagonally down the north slope of Covey Hill suggests that the streams were initially ice- marginal channels. MacClintock and Stewart’s recon- struction calls for the Fort Covington Drift to extend as a lobe far up the Champlain Valley, but I believe that ice-front position 14 crossed to the Vermont side of the valley along a line only a few miles south of the International Boundary. I have not found evidence for more than one drift in the northeast Adirondack region. Perhaps the _ younger Fort Covington ice advance was only a minor feature and not the widespread advance postu- lated by MacClintock and Stewart. Differences in till fabric are their chief criteria for separating the two drifts outside the Seaway area. There is an alterna- tive explanation. Near the center of a large ice lobe, such as the one that advanced up the St. Lawrence REFERENCES CITED 45 Lowlands to the Ontario basin, the ice moves in the direction of the major lowland, in this case to the southwest. Near the margins of the lobe, however, the ice moves outward, perhaps nearly at right angles to the direction of movement of the lobe as a whole. This phenomenon is well shown in the Ontario basin where the general movement of the Erie lobe was to the southwest, whereas the margins of the lobe at times moved at right angles to the regional trend (Prest and others, 1968). Under this hypothe— sis, the Malone Drift was deposited by a southwest- moving ice lobe some distance back from the ice margin, whereas the Fort Covington Drift near the Seaway was deposited by southeast-moving ice near the margin of the same lobe. For example, the exposure near Malone (MacClin— tock and Stewart, 1965, p. 68—69) of gray till (Fort Covington) overlying reddish till (Malone) may re- cord the following sequence of events: Deposition of red drift, derived from adjacent red beds near base of Potsdam Sandstone, by southwest-moving ice some distance back from the front. Later, southeast- moving ice near the glacier front deposited gray drift derived from rocks that crop out to the northwest. CHAMPLAIN LOWLAND IN VERMONT The deglaciation of the northeast Adirondack re- gion encompasses a part of what Stewart and Mac- Clintock (1969, 1970) named the Burlington glacial stade. During this interval, they postulated an ice advance from the north-northwest that covered the entire Champlain lowland and crossed the Green Mountains to the east (fig. 14) . Deglaciation in- volved northerly retreat of the ice front by calving into the waters of glacial Lake Vermont. Recessional moraines were not formed. It is surprising that there should be such well-developed moraines as that near Cadyville on the west side of an ice lobe in the Cham- plain Valley and none on the east. Perhaps the exten— sive deposits of kame gravel along the east edge of the lowlands south of Burlington (Stewart and Mac- Clintock, 197 0) are ice-marginal features that mark the east side of a Champlain Valley ice lobe (Denny, 1966). Wagner believes (written commun., 1970) that minor oscillations of the ice front permitted several brief incursions of marine waters into the Champlain basin. APPALACHIAN REGION OF SOUTHEASTERN QUEBEC Deglaciation in the Appalachians east of the St. Lawrence Lowlands involved the northwest retreat of an active ice front that built prominent moraines. Gadd (1964) described what he called the Highland Front morainic system along the northwest flank of the Appalachian highlands (fig. 14), extending from Riviére—du-Loup, about 100 miles northeast of Que- bec City, to Granby, east of Montreal. The moraine is at altitudes ranging from about 375 to 700 feet. In part, it consists of a belt of kame moraine 5—6 miles Wide. The local relief is 100 feet or more. Gadd suggested that the ice sheet blocked both the St. Lawrence and Champlain Valleys when it stood along the moraine, damming Lake Vermont and preventing the sea from invading the St. Lawrence Lowlands. The northwest retreat of the ice edge let marine waters flood the St. Lawrence Lowlands and the Champlain Valley. The St. Narcisse moraine (fig. 14; LaSalle, 1966, 1970; Osborne, 1950; Karrow, 1959) on the northwest side of the St. Lawrence Valley was built by a readvance of the ice sheet into the Champlain Sea, perhaps about 11,000 years B.P: (Terasmae and LaSalle, 1968) . McDonald (1968), building on the work of Gadd and of Lee (1962, 1963), carried on detailed studies in southeastern Quebec near Sherbrooke (fig. 14). He believes that the Highland Front moraine east of Montreal was formed before 12,000 years B.P., pos- sibly about 12,600 years B.P. (McDonald, p. 675),. and near Riviére—du-Loup between 12,800 and 12,000 years B.P. The moraine apparently is not signifi- cantly time transgressive, and because there is at least local evidence of readvance, McDonald (p. 675) suggested that it may correlate with mappable fea- tures in New York State and regions farther west. On this basis, the moraine is approximately equiva- lent to the Loon Lake episode of northeast Adiron- dack region (ice-front position 1). The Drummondville moraine (Gadd, 1964) in the lowlands east of Montreal, at altitudes ranging from only about 200 to 300 feet, was built just before the invasion of the Champlain Sea, about 12,200 years B.P. (Prest, 1970, fig. XII—16g). REFERENCES CITED Azmon, Emanuel, 1961, Field method for sieve analysis of sand: Jour. Sed. Petrology, v. 31, no. 4, p. 631—633. Bostock, H. S., compiler, 1970, Physiographic regions of Canada: Canada Geol. Survey Map 1254A, scale 1:5,000,000. Broughton, J. G., Fisher, D. W., Isachen, Y. W., and Rickard, L. V., 1966, Geology of New York; a short account: New York Mus. and Sci. Service Educ. Leaflet 20, 45 p. Buddington, A. F., 1937, Geology of the Santa Clara quad- rangle N.Y.: New York State Mus. Bull. 309, 56 p. 46 PLEISTOCENE GEOLOGY OF THE NORTHEAST ADIRONDACK REGION 1953, Geology of the Saranac quadrangle, New York: New York State Mus. Bull. 346, 100 p. Buddington, A. F., and Leonard, B. F., 1962, Regional geology of the St. Lawrence County magnetite district, north- west Adirondacks, New York: U.S. Geol. Survey Prof. Paper 376, 145 p. Calkin, P. E., 1970, Strand lines and chronology of the gla- cial Great Lakes in northwestern New York: Ohio J our. Sci., v. 70, no. 2, p. 78-96. Chadwick, G. H., 1928, Adirondack eskers: Geol. Soc. America Bull., v. 39, no. 4, p. 923—929. Chapman, D. H., 1937, Late-glacial and postglacial history of the Champlain Valley: Am. Jour. Sci., 5th ser, v. 34, no. 200, p. 89—124. Coleman, A. P., 1937, Lake Iroquois: Ontario Dept. Mines 45th Ann. Rept. 1936, v. 45, pt. 7, p. 1—36. Connally, G. G., 1970, Surficial geology of the Brandon- Ticonderoga 15—minute quadrangle, Vermont: Vermont Geol. Survey, Studies in Vermont geology no. 2, 45 p. Connally, G. G., and Sirkin, L. A., 1971, Luzerne readvance near Glens Falls, New York: Geol. Soc. America Bull., v. 82, no. 4, p. 989—1008. Craft, J. L., 1969, Surficial geology and geomorphology of Whiteface Mountain and Keene Valley, in Barnett, S. G., ed., Guidebook to field excursions at the 4lst An- nual Meeting of the New York State Geological Associa- tion: Plattsburgh, N. Y., p. 135—137. Cushing, H. P., 1899, Report on the boundary between the Potsdam and pre-Cambrian rocks north of the Adiron- dacks: New York State Geologist, Ann. Rept. 16, p. 1—27. Daly, R. A., 1934, The changing world of the ice age: New Haven, Conn., Yale Univ. Press, 271 p. Davis, M. B., 1965, Phytogeography and palynology of north- eastern United States, in Wright, H. F., J r., and Frey, D. G., eds., The Quaternary of the United States: Prince- ton, N. J., Princeton Univ. Press, p. 377—401. 1967, Late-glacial climate in northern United States— A comparison of New England and the Great Lakes region, in Cushing, E. J., and Wright, H. E., Jr., eds., Quaternary paleoecology—Internat. Assoc. Quaternary Research, 7th Cong., 1965, Proc. v. 7: New Haven, Conn., Yale Univ. Press, p. 11—43. Denny, C. S., 1966, Surficial geology of the Plattsburgh area: Empire State Geogram, v. 4, no. 3, p. 6—10. 1967, Surficial geologic map of the Dannemora quad- rangle and part of the Plattsburgh quadrangle, New York: U.S. Geol. Survey Geol. Quad. Map GQ—635. 1970, Surficial geologic map of the Mooers quad- rangle and part of the Rouses Point quadrangle, Clin- ton County, New York: U.S. Geol. Survey Misc. Geol. Inv. Map I—630. 1972, The Ingraham esker, Chazy, New York: U.S. Geol. Survey Prof. Paper 800—B, p. B35—B41. Denny, C. S., and Goodlett, J. C., 1968, Tree-throw origin of patterned ground on beaches of the ancient Cham- plain Sea near Plattsburgh, New York: U.S. Geol. Sur- vey Prof. Paper 600—B, p. B157—B164. Denny, C. S., and Postel, A. W., 1964, Rapid method of esti- mating lithology of glacial drift of the Adirondack Mountains, New York: U.S. Geol. Survey Prof. Paper 501—B, p. B143—B145. Elson, J. A., 1969, Late Quaternary marine submergence of Quebec: Rev. Géographie Montréal, v. 23, no. 3, p. 247— 258. Emmons, Ebenezer, 1842, Geology of New York, Pt. 2, Com- prising the survey of the second geological district: Albany, N.Y., 437 p. Fairchild, H. L., 1912, The closing phase of glaciation in New York: New York State Mus. Bull. 158, p. 32—35. 1919, Pleistocene marine submergence of the Hudson, Champlain, and St. Lawrence valleys: New York State Mus. Bull. 209—210, 76 p. Farrand, W. R., and Gajda, R. T., 1962, Isobases on the Wisconsin marine limit in Canada: Canada Dept. Mines and Tech. Surveys, Geog. Br. Geog. Bull. 17, p. 5—22. Fisher, D. W., 1968, Geology of the Plattsburgh and Rouses Point, New York-Vermont, quadrangles: New York State Mus. and Sci. Service, Geol. Survey, Map and Chart Ser. 10, 51 p. Fisher, D. W., and others, 1962, Adirondack sheet, in New York State Mus. and Sci. Service, Geol. Survey, Geologic map of New York, 1961: New York State Mus. and Sci. Service, Geol. Survey Map and Chart Ser. 5. Gadd, N. R., 1964, Moraines in the Appalachian region of Quebec: Geol. Soc. America Bull., v. 75, no. 12, p. 1249- 1254. Goldthwait, J. W., 1913, The upper marine limit at Montreal; the upper marine limit at Covey Hill and vicinity: Inter- nat. Geol. Cong., 12th, Canada, 1912, Guide Book 3, p. 119—126. Goldthwait, R. P., Dreimanis, Aleksis, Forsyth, J. L., Kar- row, P. F., and White, G. W., 1965, Pleistocene deposits of the Erie lobe, in Wright, H. F., Jr., and Frey, D. G., eds., The Quaternary of the United States: Princeton, N.J., Princeton Univ. Press, p. 85—97. Ives, P. C., Levin, Betsy, Robinson, R. D., and Rubin, Meyer, 1964, U.S. Geological Survey radiocarbon dates VII: Radiocarbon, v. 6, p. 37—76. Karrow, P. F., 1959, Surficial geology, Grondines—Champlain, Portneuf, Lotbiniére, and Nicolet Counties, Quebec: Canada Geol. Survey Prelim. Ser. Map 41—1959. 1961, The Champlain Sea and its sediments, in Leg— gett, R. F., ed., Soils in Canada: geological, pedological, and engineering studies: Royal Soc. Canada Spec. Pub. 3, p. 97—108. Karrow, P. F., Clark, J. R., and Terasmae, Jaan, 1961, The age of Lake Iroquois and Lake Ontario: Jour. Geology, v. 69, no. 6, p. 659—667. Kemp, J. F., 1921, Geology of the Mount Marcy quadrangle, Essex County, New York: New York State Mus. Bull. 229—230, 86 p. La Fleur, R. G., 1965, Glacial geology of the Troy, N.Y., quadrangle: New York State Mus. and Sci. Service, Geol. Survey Map and Chart Ser. 7, 22 p. LaSalle, Pierre, 1966, Late Quaternary vegetation and glacial history in the St. Lawrence Lowlands, Canada: Leidse Geol. Meded., v. 38, p. 91—128. 1970, Notes on the St. Narcisse morainic system north of Quebec City: Canadian Jour. Earth Sci., v. 7, no. 2, pt. 1, p. 516—521. Lee, H. A., 1962, Surficial geology of Riviere-du-Loup—Trois Pistoles area, Quebec, 22 0/3, N/14, 21 N/13 east half: Canada Geol. Survey Paper 61—32, 2 p. REFERENCES CITED 1963, Pleistocene glacial-marine relations, Trois Pistoles, Quebec [abs]: Geol. Soc. America Spec. Paper 73, p. 195. MacClintock, Paul, 1959, A till-fabric rack: Jour. Geology, v. 67, no. 6, p. 709—710. MacClintock, Paul, and Stewart, D. P., 1965, Pleistocene geology of the St. Lawrence Lowland: New York State Mus. and Sci. Service Bull. 394, 152 p. MacClintock, Paul, and Terasmae, Jaan, 1960, Glacial history of Covey Hill: Jour. Geology, v. 68, no. 2, p. 232—241. McDonald, B. C., 1966, Surficial geology, Richmond-Dudswell, Quebec: Canada Geol. Survey Prelim. Ser. Map 4—1966. 1967, Surficial geology, Sherbrooke-Orford-Memphre- magog, Quebec: Canada Geol. Survey Prelim. Ser. Map 5—1966. 1968, Deglaciation and differential postglacial re- bound in the Appalachian region of southeastern Quebec: J our. Geology, v. 76, no. 6, p. 664—677. Miller, W. J ., 1919, Geology of the Lake Placid quadrangle: New York State Mus. Bull. 211—212, 106 p. Mirynech, Edward, 1967, Pleistocene and surficial geology of the Kingston-Cobourg-Tweed area, Ontario, in Jenness, S.E., ed., Geology of parts of eastern Ontario and west- ern Quebec—Geol. Assoc. Canada (and others), Joint Mtg., Kingston, Ontario, 1967, Guidebook: Toronto Ontario, Geol. Assoc. Canada, p. 183—198. Mott, R. J ., 1968, A radiocarbon-dated marine algal bed of the Champlain Sea episode near Ottawa, Ontario: Canadian Jour. Earth Sci., v. 5, no. 2, p. 319—324. Nelson, A. E., Wiesnet, D. R., Carswell, L. D., and Postel, A. W., 1956, Geologic map of the Chateaugay quadrangle, New York: U.S. Geol. Survey Misc. Geol. Inv. Map I— 168. Osborne, F. F., 1950, Marine crevasse fillings in the Lotbiniére region, Quebec: Am. Jour. Sci., v. 248, no. 12, p. 874—890. Postel, A. W., 1952, Geology of Clinton County magnetite district, New York: U.S. Geol. Survey Prof. Paper 237, 88 p. . P‘ostel, A. W., Dodson, C. L., and Carswell, L. D., 1956, Geology of the Loon Lake quadrangle, New York: U.S. Geol. Survey Geol. Quad. Map GQ—63. Postel, A. W., Wiesnet, D. R., and Nelson, A. E., 1956, Geologic map of the Malone quadrangle, New York: U.S. Geol. Survey Misc. Geol. Inv. Map I—167. 47 Prest, V. K., 1970, Quaternary geology of Canada, in Douglas, R. J. W., ed., Geology and economic minerals of Canada: Canada Geol. Survey Econ. Geology Rept. 1, 5th ed., p. 675—764. Prest, V. K., Grant, D. R., and Rampton, V. N., compilers, 1968, Glacial map of Canada: Canada Geol. Survey Map 1253A, scale 1:5,000,000. Schafer, J. P., 1968, Retreat of the last ice sheet in New England [abs.]: Geol. Soc. America Spec. Paper 115, p. 291. Schumm, S. A., and Hadley, R. F., 1961, Progress in the application of landform analysis in studies of semiarid erosion: U.S. Geol. Survey Circ. 437, 14 p. Stephens, N., and Synge, F. M., [1966], Pleistocene shore- lines, in Dury, G. H., ed., Essays in geomorphology: New York, Elsevier Pub. Co., p. 1—52. Stewart, D. P., and MacClintock, Paul, 1969, The surficial geology and Pleistocene history of Vermont: Vermont Geol. Survey Bull. 31, 251 p. 1970, Surficial geologic map of Vermont: Vermont Geol. Survey, scale 1.250,000. Taylor, F. B., 1924, Moraines of the St. Lawrence Valley: Jour. Geology, v. 32, no. 8, p. 641—667. Terasmae, Jaan, 1959, Notes on the Champlain Sea episode in the St. Lawrence lowlands, Quebec: Science, v. 130, no. 3371, p. 334—336. 1965, Surficial geology of the Cornwall and St. Law- rence Seaway project areas. Ontario: Canada Geol. Survey Bull. 121, 54 p. Terasmae, Jaan, and LaSalle, Pierre, 1968, Notes on the late-glacial palynology and geochronology at St. Hilaire, Quebec: Canadian Jour. Earth Sci., v. 5, no. 2, p. 249— 257. Wagner, W. P., 1969, The late Pleistocene of the Champlain Valley, Vermont, in Barnett, S. G., ed., Guidebook to field excursions at the 41st Annual Meeting of the New York State Geological Association; Plattsburgh, N.Y., p. 65—80. Woodworth, J. B., 1901, Pleistocene geology of portions of Nassau County and the Borough of Queens: New York State Mus. Bull. 48, p. 618—670. 1905a, Pleistocene geology of Mooers quadrangle [N.Y.]: New York State Mus. Bull. 83, p. 3—60. 1905b, Ancient water levels of the Champlain and Hudson Valleys: New York State Mus. Bull. 84, 265 p. Page A Abandoned stream channels ...... 7, 9, 10. 11 12, 15, 16, 19, 20. 22 Alder Bend Road ....... -- 9 Alder Brook ---- _- 4 Alluvial fan ____________ __ 17 Alluvium _____________________________ 4 Appalachian region, southeastern Quebec, summary and correlation of late- glacial history ............ 45 Ausable Chasm - 30. 40, 41, 42 Ausable Forks _______________ __ 3, 16 Ausable Point _- _____________ 42 Ausable River ............ 2, 15, 16, 17, 27. 28, 29, 30. 31, 32. 33. 36. 37, 39 40. 41. 42, 43 Aussble River delta, Champlain Sea ............ 30, 31, 36, 37. 42 Lake Champlain _________________ 31, 89 Lake Vermont during Coveville stage .......... 16, 17, 29, 30, 32 Lake Vermont during Fort Ann stage ..... 27. 28, 29, 30, 31. 32. 33 B Bare-rock areas ________ 2, 6. 7, 11, 16, 19, 20, 22, 23. 24, 26, 26, 31, 32, 33, 43 Beaches ................... 2, 9, 12, 13. 20, 23, 24, 25, 80, 31, 32, 33, 86, 42 Bear Hollow ......................... 12, 23 _ Bedrock .............. a. 4, 5, 6, 20, 29, 30. 40 See also Bare-rocks areas. Bedrock knoila .................... 11, 13. 24 Belleville beach ....................... 24 Belleville-Fort Ann phase of post-Iroquois lakes ....... 36 Black Brook ................... 17 Blackman’s Rock ..................... 12 Blake Broo ......................... 16 9.. ‘ beds 26 Bridport readvance o! the Hudson— Champlain ice lobe ....... 43 Burlington glacial stade ........ __ 45 Burlington Till ............... 44 G Cadyville ............ 6, 7, 19, 20, 22, 31. 43. 45 Canadian border 2, 7, 10, 12, 15, 19, 20, 22, 24, 30 See also International Boundary. finnon Corners ............ 12, 13, 24, 25, 33 Champlain ............................ 5 Champlain basin, Vermont, summary and correlation of late- glacial history ............ 45 Champlain Sm ...... 2. 20, 25. 26, 27, 28, 30, 86. 39, 40, 42, 43, 44, 45 age .............................. 36, 44 environment ______________________ .97, 42 Champlain Sea deltas, rate of formation ................ 87 volume of ......... - 30 Chateaugay .......................... 9 INDEX [Italic page numbers indicate major references] Page Chateaugay-Cadyville episode of deglaciation .............. 19, 31 Chateaugay Lake _____________________ 17 Chateaugay River .................... 28 Chateaugay River delta, Lake Iroquois during Coveville stage of Lake Vermont ___________ 30 Lake Iroquois during Fort Ann stage of Lake Vermont --_ 31,33 Chazy _______________________________ 37, 44 Chumbusco ..................... 9, 20, 21, 22 Clinton Mills _______________ 5, 7, 9, 20, 22, 23 Clintonville 16, 17, 29, 30, 31, 43 Cobblestone Hill ............... 11, 12. 23, 26 Cold Brook ___________________________ 12. 23 Cols ______________________ 9, 15, 16, 20. 23, 24 Coveville outlet Lake Vermont ........ 25 Coveville stage, Lake Vermont __ 16, 17, 25, 28, 29, 80, 43 Covey Hill _____ 9, 10, 11. 13, 15, 16, 20, 21, 22 23,24, 25, 33, 36, 43, 44 Covey Hill channel ..... ll, 13, 15, 16, 20, 22, 23, 24, 33, 43 Covey Hill episode of deglaciation ..... 22. 43 Crossbedding ......................... 3. 23 D Deformation of deltaic deposits ........ 9, 28 Deglaciation, history - -- 2, 4, 5, 15, 42 DeKalb moraine -__- ............. 6 Deltas ................... 2, 7, 9, 16, 17, 19. 20, 22, 25, 26‘, 30, 31, 32, 33. 36, 3’1. .19. 42 Drift -. .................... 2, a, 4, 5, 7, 12. 13 15, 20, 26, 44 Drumlins ................ 5 Drummondville moraine - 43, 45 Dry Mill Brook ................. 40, 41 Duane Center .................. 43 Duquette Road ....................... 6 E Ellenburg Depot ---- 4, 5, 7, 9, 20. 22, 23. 31, 43 Ellenburg Mountain .................. 5 Ellenburg phase, post-Iroquois lakes -n- 22 Elsinore .............................. 6 End moraine ......................... 21. 22 English River ............ 8, 13, 23, 24, 25, 33 English River—North Branch Great Ch'azy River divide .......... 13, 23, 24 Erie lobe _______________ 45 Exfoliation domes ____________________ 3 F Flat Rock. near Altona __- 11, 12, 22, 23, 26, 32, 33 Fluted drift __________________________ 5 Foreset beds _________________________ 11, 28 Fort Ann outlet, Lake Vermont 23, 25, 33, 43 Fort Ann stage. Lake Vermont---- 12, 23, 24, 25, 26. 27, 28, 29, 30, 81, 42, 44 Fort Covington Till __________________ 21, 44 Frontenac phase, post-Iroquois lakes _- 22 Page G Giaciofluvial deposits _- 7, 16. 19, 26, 28, 30, 44 Grain size in shoreline deposits ________ 28 Great Chazy River __________ 7. 9, 16. 20, 22, 26, 28, 29, 30, 33. 36, 39 Great Chazy Delta, Lake Vermont ' during Fort Ann stage- --31, 33 Great Chazy River-North Branch Great Chazy River delta, Champlain Sea ........ 30, 36, 3'! Great Chazy River-North Branch Great Chazy River divide _______ 20 Green Mountains .................... 45 Grooved drift ________________________ 5 H Hardscrabble Road ................... 6 Harkness _- ___________________________ 17 Hawkeye Granite Gneiss ______________ 5 Highland Front moraine .- Hudson River ........................ 25 Humbug Mountain ................... 16 I Ice-channel fillings ................... 4 Ice-dammed lake .............. 2, 7. 9, 15, 16 19. 20, 22, 23, 24, 25, 43, 44 Ice-marginal streams ......... 2, 9. 12, 15, 17. 19. 20, 22, 23, 24,31, 44 ice movement, direction .............. 5 [ngnaham eIker __ 20, 37, 42. 43 Internal stzucture of deltas ........... 28 International Boundary ___________ 3, 16, 20, 28, 25. 42. 44 Introduction .......................... 2 J Jack pine ........................... 11 Johnson Mountain ................... 22 K Kame moraine ....................... 45 Kaine terraces _ ____________________ 2, 15 _ 4, 6, 7, 9, 17, 19, 31, 44, 45 Keeseville _-_ ..... 17, 2’7, 28, 29, 30, 31, 32 Kettles _______________________________ 24 L Lake Champlain ............... 26. 27, 36, 44 Lake deposits -.- 9 Lake Frontenac _ 22 Lake Iroquois ..................... 2, 11, 15 16. 20, 21, 22, 23. 24, 33, 43 Lake New York _____________________ 43 Lake Ontario basin - _ 20, 22, 24, 25, 43, 45 Lake outlets ________________________ 2, 26 See also Coveville outlet, Covey Hill channel, Fort Ann outlet, and Rome outlet. Lake Vermont ............ 2, 7, 12, 13, 16, 20, 22, 23, 24. 25, 27, 28, 30, 36, 42, 43, 45 49 50 Page Lake Vermont deltas, volume of ...... 30 Late_ 0 0 2y ; m a: ‘ . 1% S E < * ’We 8 a 2 Q» ‘3 ~§ ( g QVC ______ E a“: g Sand and gravel Inferred highest stand 1:: D O GLACIOFLUVIAL DEPOSITS Largely ice—contact Stratified sand and gravel MORAINAL DEPOSITS ng Recessional moraine Ground moraine ¥ O Ridges and knolls of till, sand, and gravel Till, includes small bodies of sand and gravel and alluvium of Holocene age J /0/ Large area of abundant outcrop Delta deposits Striae, drumlin, and grooved of Potsdam Sandstone drift 30; I, i k k ; v: I! (7 _ ‘ ,’/. ‘ . ,,,,, ' ., ~ (It: « - “I OI / Abandoned stream channel or col ———8 Ice-front position Dashed where inferred. Number refers to position in sequence; 1 is oldest _____ 3E_____ Divide between Champlain and St. Lawrence Lowlands showing lo- cation of abandoned stream chan- nel or col crossing the divide Contact Area of this report O . h lions “ f ‘ , i/ ' MK @c gev L _ :M: {14015, L_-fi-_\_*\_::.‘ 44°‘i5' A ~ m INTERIOR—GEOLOGICAL SURVEY. WASHINGTON. D.C. 73 °15’ 1974—6723168 INDEX MAP Base from U.S. Geological Survey, Lake Champlain, 1947,- SCALE 1:250 000 Compiled by c. S. Denny largelyfrom Buddington (1937), OF 'Ogdensburg, 1943; Canada Dept. of Energy, Mines and 5 o 5 10 15 MILES Denny(1967, 1970), MacCIintock and Stewart (1965), NEW YORK Resources Mines, Ottawa, 1953, and Montreal, 1965 H '—4 l—‘—* I Nelson and others (1956), Postel (1952), and Postel and others (1956) 5 O 5 10 15 KILOMETERS , . l CONTOUR INTERVAL 100 FEET DATUM IS MEAN SEA LEVEL SURFICIAL GEOLOGIC MAP OF THE INORTHEAST ADIRONDACK REGION, NEW YORK PROFESSIONAL PAPER 786 UNITED STATES DEPARTMENT OF THE INTERIOR 35015: PLATE 2 A GEOLOGICAL SURVEY 30. -cr ~ _. "fl __ —LTR%€.§TETE? o I 45' .g-n—I - S; , 45000; M‘” I 5' QUEBEC .m- \74 00 - ‘ 2 L . 45‘ 00 m - ’ I! /w . *» I, '- "' (-3131? - m , m 0 I \ .‘3‘ Font 95 . w} “I“ :0“ 3._%!WW‘ :2?» {in ,. 7 ‘3» “t f! mm 3 PM A“) A. 535‘ A I/ ‘ rd“ ” :‘c 8”” ~ am “3% ‘ M *3 <5 II an !e / 0 am Corners A“ : I _ :63 IE; ‘95 t ’ 9 Wes! ‘ 3 3/? MW/ I /_ ... a" Center ’ 4: W,» Coveytown F» at, A/ V,“ _ , g h) I ”Mg/rum m MW Cont?) ATM/M” «a. ‘1 i a $9 2 j Censta Tm S” Then (T * A \T h n g’ Point. F0! _. flan ’ " e 0‘" 9’ ‘ e r Qd “ r, to EXPLANATION .. \.,_ h o ‘5. i I K 0 E ,4 N\\ ’0‘ a I. I $9 ' I ' g < st I“ ' s g : IQ" Qd z A 9 Sc 2 0 Chazy o *9 3 [I T(\ “” Chat , o u (I) ‘ "O” i 3 *9 Law" 2° 3 3 129 Q 1' . Glacial and alluvial deposits [ll—J ‘ w“ I 191 S 2’ ~31 ? ‘ Q '2 __ g 5“: Shown only where underlying bedrock < Miner 19 13' 5 Q t 5: a u “ ‘ geology is unknown 3 ake _ " 0° 00 6 z k: - O . I i: e , *’ 9 .3 II 2 fir,» JV \ I V v PLS _] O Q _ a 5k V: .24 my ~ 8 " 3 9 , d 5 < N :5 a 1 ‘ 0 Bay EL 0 i . I o Sedimentary rocks 5 «03¢ Q 5: Cha . \ Point av Roche é % 1 9: Hero % Q . Qd w p€u ngn pea ‘7 . Jericho 3:; E <21: :51 ‘53? Q I). 2/ z Crystalline rocks, Granite g'neiss Anorthosite E ”"3 x: t a . ’ we n N. undifferentiated Chiefly Lyon Moun- Map umtam Egg" m ‘ i M a 4537;» " - Q 3’ ’ All map units of Pre- WW Granite (met-‘33Y and others (19 ) 2 I , ‘3“ T35 if: g Cb 3 t G Si V The cambrian age in includes smegran— <1: I \ J ’ ~ N e g Q Q Gut Fisher and others ”9 and related 8 ~ - f - 45’ (1962) except a , 7001‘s 0: ° a: ' ,, 0“ Dhg, and hbg with- Map units phg and [L ”a. r" S a I i CUWEM :0 0 out overprint hbg without over- ..: Plat , h Head \ print in Fisher and o ' (PT ‘1: 0° ‘9' 5.6 others (1962) d 0 P‘s ‘ It” ' o - 5, e I 1 2 1 t : _ ’° 3) , Piansb g I“ f sm“ t1: ’ Cumberla ; l ‘ “and N /\y I M i 36 3 ay x‘ 31,}: “:“a 35 ta ‘ / £\’ ‘3" mg“ \ “~ Contact / 3 93" m, 37 {Q I M 0‘ 1:: P15 T Q?) if ———40—— ————20—--— 1 $0 0 22 b ‘7 “two . J - we; " V ‘1‘ “ Granite gneiss Anorthosne _ IT g M 0 _ er Isopleths showing percentage of clasts 1n till \T ,— " {22 Ch" Havun Contour interval 10 percent ONT “T #1 P1 ‘5‘ h i burg g? I . ‘ s 1 Q 3‘ / , ' * ‘ > _ 5 . . z» (’3‘) Q7 A 2 x E: V , t, / ti \ a Iver I“ Q: c U / 2 “I of /_/ \fi 33“” ”.4 (\ 2:3! {8 C X:— G ' ”V J ‘ ~ \ 2 b 9 C? a "‘ N 0’ V27 N‘» \ Qd & akaw Island P id ° . . \E Q / // ”If: ”<8 f \' ale Ighd Sample site in tlll T" VAT/ / ° Upper number indicates percentage of \ T clasts larger than about 0.5 in. that are granite gneiss. Lower number indicates m \ percentage of clasts larger than about 1 I‘ I in. that are anorthosite “2‘43 0 _ A \ O 3 ‘ ‘ c ptl'n / t. CLINTON COUNT; - 1 1 —O— 1 f 1a.. E5156? 90‘1”" \ a Strlae, drumhns, or grooved drift \ Lx‘k Port Kent \ “‘“x e 73”,.” ' I} T‘fg’ffilog e i e @ SCIWY’" “~\t":z,,y 30, a Q / Maud ~\\~ - ,7 P15 c; ‘~\\ 0 I.) on Bougiass r“ a ' 1.0 w m «I 2 X13 I o 52 ’ T I Q :\‘\\“\g \ 95 .l 8 B ‘I ,9) g \ {A L ‘ \ / Wilhboro Pom! 9* I} Show. n: a ‘ $3 Four Brothers PW” 9: a ° 0 ' sboro Point $3 0 H A M P"L A Mr. ” F C; h "on? \\, a _\\ I A / VI at“ fl - 44015' 74 000' 45’ I N’lNTERIOR—GEOLOGICAL sunvav, WASHINGTON, D.C.—1974— 572363 73 “15’ Base from U~S- Geological Survey SCALE 13250 000 Geology generalized from Fisher and others (1962). Per- Lake Champlain, 1947, and Ogdensburg, 1943 5 0 5 10 15 MILES centage of clasts estimated by rapid method described - - ,_. ,_, g I by Denny and Postel (1964) 5 O 5 10 15 KILOMETERS I—--I I—I I—-I J CONTOUR INTERVAL 100 FEET DATUM IS MEAN SEA LEVEL MAP SHOWING BEDROCK GEOLOGY AND PERCENTAGE OF CLASTS OF GRANITE GNEISS AND OF ANORTHOSITE IN TILL, NORTHEAST ADIRONDACK REGION, NEW YORK UNITED STATES DEPARTMENT OF THE INTERIOR ~ GEOLOGICAL SURVEY ,M ‘17 NW!” 3 74°03}; r’“ , Iv i“ ” 4/ DD at; mm/ M} & Q .30, 9 Cl 5 Hemmlngford "/V/ {7 P15 f q f ~29"? 5:) [mkwm/N CD "MN‘x/V O W“; »\ ”A r W w no i 74:21“ / [\IS / \ aman M ’ ‘ M» l~~;/~~«wr‘~:t“~\ a Mr ,W/T xv / 6 Es ,WQEN; \ / 6H,A»~W—-—w~t‘.:~\\\x C vey ‘li t ,, 5"" 2 0 max \ 73’13) ”‘ N CANADA We, ”mm—u.— a— 45000: Mai) ’F‘” are m L} \1 .. a T UNITED sfiires ;TW*- Trou i ‘ Qafiic -; fl 33 0 _ I ‘ — 45"“00' (:7 m 40 l y K RA mu 3? a 90'} ’ 7' ‘ ,1 . it “:8; x A ,~ { M use m :5 Po; .. . > r at tiff“ I \ 50° ’5 i‘ MM “VT/1,} ‘ )A‘ l g M ii l H o s n Mb ._. EQSlA 1.33 h! 8g )3” arl is! > W ON. 3"”ch Corners : k y: ¢ W ( was! . e 3 g p/ , fl 9 , Center 3 a “89 "' r/ l) we COYEYW’W j’K ’” "f m (m. abuser) { x ,, ; , m Cont?> , \ \ " N” 4' CT“ xflx ya / J lie ‘3 A r , C - 9 I I,» b- 1 /~ . e Canals 7}» «J The \, 0 p ‘ ”T “ _ 2‘ / I "\ n . 3 e C "J i W, , g; y < Age » E" Point am For _ Ce ° 9 , 0y ‘3 r. a» \\ 50 i l .. w,” . 21f”) ,. a '"‘\ A/ _ A ' N 1% m \._...f/ 5‘8 W Cj \\ (3 I 8 w/ J‘ Qd (v 1 , 52 \ ff ,_ “a9 Nix r De ‘1 1 3:67, ’1 K ,. ‘9 ‘3 53:. 6‘ (A; ””11””. V/r/ ’1‘ //\ i i, we ,. / “V”: ‘53:». Fa x” d m \ a- fig V k?) 8 I \N‘) x“ {9 ‘9 M Q C "MM” 5 w ‘ >4; HO/ 3v95f\ t? 212 CM 91 . e ‘1, w W CW? 0 8 .z? a “ 5mm é? o \ 3/“- «waJw > A _ Landm Q w 0 we (,1. | 0‘. It“ 5, ska/U]; \ x “>99\‘ ‘ 3 gay; a”? Q 3.: w, . . * A. “ a.“ "w ‘ I , , O I’fls‘V”"\§II rifiéyg e ‘9, , \; \\ M’W’-, x V ' y ,, J3 g .: isée a v : é;- ‘ 8 8 r if N l 1 f if": b.) O i 9 ,0 g E WWWWW g” z 5 a» 5k 3 my a (9 Bag 0 Vein? an Roche 3 h Hero ingrah __ ’ i P/ ' Qd ”*5 o \ Z ”N. ’2 ' /W N‘ l n I <9 ‘3: The K; Cu! ‘2 QM ..... 4 E)’ > ' Cumberland 0 rv\\ ,, , V2.31 : _\’l "Qua z u? h {3 P15 7 “if I: ’15: use [:36 , Plaflsb r .9 f ”swag. \\ i _ ‘1‘ (‘uméen’an {I mom; - ”y, 3‘ 3 0y ‘rxfr’q r N Ia x Jan w ! ~ “\ :03“ Rx 37 E} l/ a; \5 Q) 3 5/» “N \‘rx /,5 _ k. , ‘2- Ofia J R» x w a, 9 :74 ”Max \ ‘ Q “a? \‘~ :3 er / a} '\\ :2 59 partway.» ml”; “Pl shurgl } é: \ / if; s ‘ M / a I I g E ” W155 ‘2 ‘3 s z ) , \ Qd } iv ‘3 1:0 <9 ( .. (I \ alc »' alcourixland \vaidenc o ‘ V ; . hand / / G \ \ \ 'O- c plain rm ’ climax cou ’ ESSEX COUNTY \ \, \ fl. \\ Port Kent " ‘ we 8 , . “wing” r’ c: ’ “Na 2 , / Schuyler N‘x Ffi’ly 3H" //¢ it!!!“ ‘ \N‘x‘ ( If“ A as A ‘81 E x10 0 E 0 95 l 3 \ {2s Willsbom Fain? Shell’s/me Four mother: PW" (50 6 a 5mm F‘oint «2' IN ‘0 6 Mix CHAMP'LA c—u/ I d M h llons ", ii» We?“ ’ 44015' 30‘ INTERlOR—GEOLOGICAL SURVEY. WASHINGTON. D.c.-—l974—G72368 71 i, 1 5), 3,»; 2/, Base from US. Geological Surve L L ' - . I). Ogdensburg, 1943; Canada Dtptél; (EinaeTngall‘rd'irijfzird SCALE 1.250 000 Geology generalized from Fisher and others (1962). Per- R . ' 5 o 5 10 1 MILES centage of clasts estimated by rapid method described esources Mines, Ottawa, 1953, and Montreal, 1965 I—l l - l—-——-—l : . ,5 by Denny and Postel (1964) 5 O 5 10 15 KlLOMETERS i2? I—I l—l ' J CONTOUR INTERVAL 100 FEET DATUM IS MEAN SEA LEVEL MAP SHOWING BEDROCK GEOLOGY AND PERCENTAGE OF CLASTS OF PALEOZOIC SEDIMENTARY ROCK IN TILL, NORTHEAST ADIRONDACK REGION, NEW YORK PROFESSIONAL PAPER 786 PLATE 5 EXPLANATION Qd Glacial and alluvial deposits Shown only where underlying bedrock geology is unknown Dolostone, limestone, sandstone, and shale Largely of Ordovician age [Z Potsdam Sandstone Quartz sandstone, arkose, and shale PALEOZOIC QUATERNARY L__\F__JL_W__J CAMBRIAN Z I S pCu Lu 0: 0: in Crystalline rocks n‘ E 0 Contact 20 Isopleth showing approximate percentage of Paleozoic sedimentary rock clasts in till Contour interval 10 percent X95 Sample site in till Value indicates percentage of clasts larger than about 0.5 in. in diameter that are Paleozoic sedimentary rocks _0— Striae, drumlins, or grooved drift % >K_<0 7. n a lkj R s E m M a P 4. m m L E _ s .m A o N m m. m m P o m % 1 m E g w G F Q m 0 G R P o g Dashed where gradatioml E X P LA N AT | O N Qal Alluv1um 0 Large area of abundant outcrop of Potsdam Sandstone 0 Moralne near Cadyv1lle Also includes some sand and gravel Contact, approximately located Highest stand of the Fort Ann stage of Lake Vermont Highest stand of the Coveville stage of Lake Vermont Abandoned stream channel 5 Sand and gravel m r Q Recessional moraine Narrow ridges and knolls of till 5 Sand 1969: WASH! NGTON, D. C —-I974—G72368 Geology compiled by C. S. Denny, based largely on Denny (1967) ‘30” INTERIOR—GEOLOG|CAL SURVEY. 73°37'30" 7303?: 1 MILE 1 KILOMETER SCALE 1:24 000 O O DATUM IS MEAN SEA LEVEL SURFICIAL GEOLOGIC MAP OF THE MORAINE NEAR CADYVILLE, CONTOUR INTERVALS 10 AND 20 FEET .5 I—II-HI—lt—lI—I 73040' NORTHEAST ADIRONDACK REGION, NEW YORK UNITED STATES DEPARTMENT OF THE INTERIOR GEOLOGICAL SURVEY Base from U S Geologlcal Survey Dannenora, 1966, and Morrisonwlle, 1966 UNITED STATES DEPARTMENT OF THE INTERIOR GEOLOGICAL SURVEY 44°52’30“i " Base from US. Geological Survey Ellenburg Depot, 1964 INTERIOR—GEOLOGICAL SURVEY. WASHINGTON, D.C. —l974—G72368 73°47’30” Geology by C. S. Denny, 1969, based on scattered field observations, interpretation of aerial photographs, and MacClintock and Stewart (1965, pl. 18) SCALE 1:24 000 1,5 1 MILE I , x x l 1 KILOMETER CONTOUR INTERVAL 10 FEET DATUM IS MEAN SEA LEVEL PROFESSIONAL PAPER 786 PLATE 5 EXPLANATION Alluvium Sand and gravel V QUATERNARY Recessional moraine Ground moraine J Known bedrock outcrops Moraine near Ellenburg Depot Contact, approximately located Dashed where gradational <— Glacial striation Point of observation at tip of arrow x." Sand pit Letter indicates locality referred’to in text —0— Drumlin or grooved drift 44°52’30” SURFICIAL GEOLOGIC MAP OF THE MORAINE NEAR ELLENBURG DEPOT ALONG NORTH BRANCH GREAT CHAZY RIVER, NORTHEAST ADIRONDACK REGION, NEW YORK PROFESSIONAL PAPER 786 PLATE 6 UNITED STATES DEPARTMENT OF THE INTERIOR GEOLOGICAL SURVEY 73°50 CANADA 47,30" UNTINQDON £0.“ 68.2. 7-, fvfr ' .finmfiafi-n02 -,« U ° ‘\\ 45¢oo'f EXPLANATION Sand and gravel QUATERNARY Recessional moraine Ground moraine Narrow ridges and krwlls of till Large areas of abundant outcrop of Potsdam Sandstone CAMBRIAN o o O Moraine near Ellenburg Depot Contact, approximately located Dashed where gradational Highest stand of the Fort Ann stage of Lake Vermont <——————— Glacial striation Point of observation at tip of arrow <2 Abandoned stream channel Beach ridge or spit XA Locality referred to in text 44°57'30” 44°5?%0“§‘ INTERIOR—GEOLOGICAL SURVEY, WASHINGTON. D.C.—1974—G72368 73 “45‘ Geology compiled by C. S. Denny, 1969, from scattered field observations, aerial photographs, and from Denny (1970), MacClintock and Stewart (1965, pl. 18), and Postel (1952, pl. 1) Base from U.S. Geological Survey Ellenburg Depot, 1964, and Altona, 1966 1 MILE 1 5 O l KlLOMETER l—li—li—ii—ii—q' CONTOUR INTERVAL 10 FEET DATUM lS MEAN SEA LEVEL SURFICIAL GEOLOGIC MAP OF THE AREA AT THE HEADWATERS OF ENGLISH RIVER WEST OF CANNON CORNERS, NORTHEAST ADIRONDACK REGION, NEW YORK UNITED STATES DEPARTMENT OF THE INTERIOR GEOLOGICAL SURVEY 35’ 73°32’30” 5M 353‘ .~ “x, a l“- . .. ' \ v£ , x , . , w g-i'kil av 35; INTERIOR—GEOLOGICAL SURVEY. WASHINGTON. D.C —1974—G72368 73 032730” Base from U.$. Geological Survey SCALE 1:24 000 West Chazy, 1966 0 5g 1 MILE Geology by C. S. Denny, 1969; based largely on Denny (1970); bedrock attitudes from D. R. Wiesnet (written commun., 1965) 0 .5 1K|LOMETER CONTOUR INTERVAL 10 FEET DATUM IS MEAN SEA LEVEL MAP SHOWING BEACHES ON COBBLESTONE HILL AND PART OF FLAT ROCK NEAR ALTONA, NORTHEAST ADIRONDACK REGION, NEW YORK if'x 4405230" °.-.',-o. ‘a'o'...°‘ "93° ' 3.93.2». Sand and grave] Gravel Recessional moraine / W 44 “50’ PLATE 7 / I EXPLANATION 379%“? Ground moraine Potsdam Sandstone In part concealed by vegetation Contact, approximately located Dashed where gradatioml 2 ._L Inclined Horizontal Strike and dip of bedding Highest stand of Fort Ann stage of Lake Vermont Highest stand of Champlain Sea Beach ridge or spit <: Abandoned stream channel PROFESSIONAL PAPER 786 QUATERNARY CAMBRIAN