DOC 8T7P§2457 A B—1515 01515 November 1985 RESPONSE TO LIVESTOCK GRAZING AT THE TEXAS EXPERIMENTAL RANCH TEXAS EX PER! M ENTAL RA N C H ~ m COQPERATION WITH . - swan R. swsusou CATTLE co. TEXAS EXPR1ENTAL. RANC-H comm-via THE TEXAS AGRICULTURAL EXPERIMENT STATION! NOVmQ P. Clarke, Director/Tho Texas A8|M University Sylbm/ COIIQQQ Station, TQXII Contents FOREWORD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ii STUDY AREA . . . . . . . . . . . . . . . . . . . . . . . . . . i. . . .1 Climate Physiography and Soils Vegetation Grazing Treatments METHODS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1960-78 1978-82 Relationship Between Sampling Methods RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 VEGETATION TRENDS (1960-78) . . . . . . . . . . .4 Clay Loam Range Site Rocky Hills Range Site Clay Flat Range Site Loamy Bottomland Range Site Very Shallow Range Site Shallow Range Site Shallow Clay Range Site Deep Redland Range Site FREQUENCY (1981--82) . . . . . . . . . . . . . . . . . . . ..5 Clay Loam Range Site Rocky Hills Range Site Clay Flat Range Site Loamy Bottomland Range Site FORACE PRODUCTION . . . . . . . . . . . . . . . . . . . .7 Clay Loam Range Site Rocky Hills Range Site Loamy Bottomland Range Site DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 SUMMARY AND CONCLUSIONS . . . . . . . . . . ..9 Loamy Bottomland Clay Loam Rock Hills LITERATURE CITED . . . . . . . . . . . . . . . . . . . . ..9 APPENDICES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . IO Response of Vegetation t0 Livestock Grazing at the Texas Experimental Ranch . K. Heitschmidt, S. L. Dowhower, B. A. Gordon, and D. L. Price* Key Words: Grazing systems, stocking rate, forage production, species composition, range site. Foreword An understanding of the principles of grazing management is essential for the successful manage- ment of any ranching enterprise. Because these principles are based primarily on an understanding of the ecological response of the native forage to various grazing management practices, grazing management research often requires a longterm commitment of considerable resources. These com- mitments were made with the establishment of the Texas Experimental Ranch. The 7,000-acre ranch was created in 1959 to provide research scientists an opportunity to deter- mine both the short-term and long-term effects of various grazing management practices on both livestock production and the native vegetation. It was created from a cooperative agreement be- tween a group of interested ranch and businessmen, the Swenson Land and Cattle Company, and the Texas Agricultural Experiment Station. The businessmen formed the Texas Experimental Ranch Committee for the purpose of securing funds for construction of necessary improvements such as fences, watering tanks, and headquarter buildings. The committee has continued to function in this role over the past 25 years as well as act in an advisory capacity to the research scientists and ad- ministrators who are associated with the ranch. The Swenson Land and Cattle Company agreed to provide the necessary land and cattle needed to implement the research. The current owners, the Swen R. Swenson Cattle Company, continue to provide these valuable resources. The Texas Agricultural Experiment Station provides the necessary funds and personnel to maintain this uni- que research facility. The ranch and this publication are truly a product of the combined efforts of many individuals. Data included in this publication were collected over a 22-year period and reflect the dedicated work of many people. Specific appreciation is expressed to the many project leaders, research associates, technicians, and student workers who assisted in this endeavor. Specific thanks are ex- tended to Dr. W. I. (Dub) Waldrip, initial project leader at the ranch, for his contribution in develop- ing and initiating the long-term studies. Appreciation is also expressed to the many employees of the USDA Soil Conservation Service who assisted in this study and specifically to Mr. Clyde Lowther, USDA-SCS soil scientist, for the many hours he has spent at the ranch mapping soils and developing soil descriptions. We also acknowledge the contributions of the Swenson Land and Cattle Com- pany, the Swen R. Swenson Cattle Company, and past and present members of the Texas Experimen- tal Ranch Committee. M...” awafi$a Rob Brown, Co-chairman john Matthews, Co-chairman Texas Experimental Ranch Committee Texas Experimental Ranch Committee \ Response of Vegetation t0 Livestock Grazing at the Texas Experimental Ranch R. K. Heitsehmidt, S. L. Dowhower, R. A. Gordon, and D. L. Priee* The Texas Experimental Ranch is located on the eastern edge of the Rolling Plains resource area (Gould, 1975), and encompasses 2,900 hec- tares (ha). It was created in 1959 to provide research scientists with an extensive rangeland laboratory to evaluate the long-term effects of grazing management on cow/calf and vegetation performance. Since 1960, three grazing treatments have been continued in the same pastures without interruption: continuous grazing at heavy and moderate rates of stocking and a four-pasture, three- herd deferred rotation treatment stocked at a moderate rate. Previous studies have examined the long-term effects of these treatments on cow/ calf performance (Heitschmidt et al., 1982) and economic profit (Whitson et al. , 1982). The objective of this bulletin is to quantify the long-term effects of these grazing treatments on the native vegetation. Study Area Climate Climatic conditions are character- ized by warm, wet springs and falls, mild winters, and hot summers (Ap- pendices 1, 2, and 3). Average an- nual precipitation is 682 mm (Figure 1) and bimodally distributed. Mean daily minimum temperatures range from — 2°C in January to 22°C in July. Mean-j daily maximum temperatures range from 11°C in January to 36°C in July. The frost- free growing season generally ex- tends from March to November. Physiography and Soils The ranch is located in the Cen- tral Lowland Physiographic Pro- vince (Godfrey et al. , 1973). Topography is rolling and ranges from broad valleys and gentle slopes (1 to 3 percent) that lead to nearly level uplands ( < 1 percent), to nar- row secondary valleys with steep slopes ( > 5 percent). Range in eleva- tion is from 408 m to 463 m. Soils at the ranch range from the deep, well-drained clay and clay loams found in the valleys and on the gentle slopes and upland, to the shallow, stoney clay and clay loams found on the steeper slopes (Figure 2). Parent materials are limestone bedrock and overlayed clayey sediments deposited by wind and water from the mantle cover of the High Plains (Godfrey et al., 1973). The High Plains mantle consists of Rocky Mountain sandy to clayey sediments originally deposited dur- ing late Tertiary, Pliocene, and Quaternary. For a complete descrip- tion of dominant range sites and soils see Appendix 4. Vegetation Vegetation at the ranch is primarily a mixture of mid- and shortgrasses. Dominant midgrasses are sideoats grama, a warm-season perennial, Texas wintergrass, a cool- season perennial, and Japanese brome, a cool-season annuall. Dominant shortgrasses are buf- falograss and common curlymes- ‘See Appendix 5 for scientific name of all plant species. quite. Texas broomweed, a warm- season annual, is the dominant forb. Honey mesquite is the dominant woody plant, lotebush the dominant shrub, and brownspine pricklypear the dominant succulent. The entire ranch was aerially sprayed for con- trol of honey mesquite in 1964 and 1972 with 2,4,5-T [(2,4,5-trichloro- phenoxy) acetic acid] and in 1979 for control of both honey mesquite and pricklypear with a mixture of 2,4,5-T and picloram (4-amino-3,5,- 6-trichloropicolinic acid). A com- plete taxonomic list and ecological classification of all vascular plants found on the ranch is presented in Appendix 5. Grazing Treatments Three grazing treatments have been continued in the same pastures without interruption since 1959. These treatments are: (1) yearlong continuous grazing stocked at a heavy rate (heavy continuous-HC); (2) yearlong continuous grazing stocked at a moderate rate (moderate continuous-MC); and (3) a four- pasture, three-herd deferred rotation (deferred rotation-DR) stocked at a moderate rate. Four ungrazed ex- closures (EX) were also established in 1959. Rate of stocking in the two moderately stocked treatments was gradually increased from 1960 to 1978 to maintain relatively constant stocking pressures (Figure 1). Treatments are replicated across pastures. Size of the two heavy cen- tinuous and two moderate con- tinuous pastures is about 240 ha. Size of each of the four deferred rotation Agricultural Experiment Station, Vernon, Texas. i *Authors are associate professor, research associate, and former research associates, The Texas pastures is about 120 ha. The four small exclosures range insize from 1 to 8 ha. For a more complete de- scription of each grazing treatment, livestock management procedures, and livestock performance data, see Heitschmidt et al. (1982). Methods Vegetation sampling procedures varied among years. Initial sampling procedures (1960-78) were designed to monitor vegetation trends across time relative to both size and number of plants. Procedures since 1978 have focused primarily on quantifying vegetation changes after 2O years of treatment. I 960- 78 Vegetation was sampled using a 10-point, 45° angle frame. Twenty- four permanent frame locations per 8O mz-sample plot were read each summer from 1960 through 1969, and in 1971, 1973, 1976, and 1978. At each location the frame was lowered vertically into the vegeta- tion with all pins down until the legs of the frame and the pins touched the soil surface. The frame was then inclined to the 45° angle and each species and/or species groups touching each pin was recorded. Recorded species and species groups are presented in Table 1. Originally, 333 permanent plots were located on five range sites. A more detailed soil survey in 1978 in combination with the most recent range site classifications revealed the 333 plots were located on 18 soil series representing 10 range sites. An inadequate number of sample plots ( < 3) limited our data analyses to 14 soil series and 8 range sites. Number of plots per treatment by soil series and range sites is presented in Ap- pendix 6. Data were summarized by plot and statistically analyzed using least square linear regression models where Y equaled percent frequency of occurrence and X equaled number of years after initiating treatment. Frequency was calculated by dividing the number of hits for a given species by 240 which was the total number of pins read per plot. Models were developed only for- 2 10i- MC DR HC STOCKING RATE lHA/COW/YEARI w I 90- 80>- 70- PRECIPITATION ICMI N HC AVERAGE 65 YEAR Figure 1. 75 8O Total annual precipitation (cm), 23-year average, and annual rates of stocking for heavy continuous (HC), moderate continuous (MC), and de- ferred rotation (DR) grazing treatments. those species or species groups that had a frequency in either 1961 or 1978 of at least 5 percent. 1 9 78-82 New vegetation sampling pro- cedures were initiated in 1978 to monitor, long-term vegetational changes induced by climatic fluctua- tions and livestock grazing. Method of sampling consisted of recording plant species present in 200 frequen- cy quadrats per sample plot. Fre- quency of occurrence was recorded in two sizes of quadrats. The smaller size quadrat (10 >< 1O cm) was located in the upper left corner of the larger quadrat (25 >< 5O cm). The smaller quadrat provided a more sensitive measure of changes in fre- quency of the more abundant annual species that occurred in the larger quadrat (Hyder et al., 1975). Twenty-four sample plots were selected as permanent plots from the 333 permanent plots originally sampled from 1960 through 1978. Sample plots were limited to the three dominant range sites found at the ranch. Each range site was sampled twice in each treatment. Range sites selected for study were rocky hills, clay loam, and loamy bottomland. In addition to frequency data, potential aboveground net primary production (ANPP) was estimated during1982 on the three range sites in the HC and MC pastures and the exclosures. At each location all aboveground standing crop was removed in February and four sets of paired quadrats were caged with wire exclosures. Total standing crop in one quadrat of each pair was harvested in June near the time of J \ L F Mollisols 1I Ir lnceptisols -—*-l Vertisol B ttom- |—--Cl Loam————1 Shallow Clay Deep Very Clay Rocky Shallow Clay o land ay Loam Redland Shallow I-Oam "m5 CIaY Flat L d Valera Rowden ue ers Throck Mereta Owefls Nuvalde Rowena B Fm; Nukrum A horizon . J I calcium l hofizon p0 ' ~11." 7."; “kick.” 1. l ~~-'Ls_e_dim l l l __ L ll IIJ-t-gfi 1 l I l limestone/shale i T l l F l I I T J_ I T Figure 2. Horizontal profile of major soils and associated range sites. TABLE I. ABBREVIATED CODES FOR SPECIES AND SPECIES GROUPS THAT WERE MONITORED FROM I961 THROUGH I978 IN PERMANENT FREQUENCY PLOTS Common Name Code Buffalograss and/or common curlymesquite Texas wintergrass Sideoats grama Annual grasses Red threeawn Sand dropseed Meadow dropseed Texas cupgrass Blue grama Vine-mesquite Western wheatgrass Big bluestem Miscellaneous grasses Texas broomweed Other forbs BUDA/HIBE STLE BOCU ANGR ARLO SPCR SPAS ERSE BOCR PAOB AGSM ANGE MISC XATE FORB maximum live biomass. The second quadrat was clipped in October near the end of the growing season. All standing crop was harvested by species at ground level and dried at 60°C prior to weighing. ANPP was estimated by species by summing Q peak standing crops. The ANPP data were subjected to standard analysis of variance pro- cedures (Little and Hills, 1978). Mean separation procedures follow Tukey Q procedures as outlined by Snedecor and Cochran (1967). Fre- “quency data were not statistically analyzed. Relationship Between Sampling Methods Knowledge of the relationship between sampling methods was con- sidered essential for proper inter- pretation of these data. The inclined point frame method utilized from 1960 through 1978 provided an in- tegrated measure of the frequency of occurrence of a species in both the horizontal and vertical planes. Fre- quency estimates obtained from this method were thus affected by not only the presence or absence of a species but also quantity of standing crop. The relationship betwen these frequency estimates and current standing crop was examined by Wied (1967). The vegetative sampling method utilized from 1978 through 1982 pro- vided an estimate of frequency of oc- currence of a given species in only the horizontal plane. The relation- ships between this and the point frame method were examined utiliz- ing standard correlation procedures (Draper and Smith, 1966) for estimates obtained in 1978 from 25 plots sampled with both methods. Likewise, similar correlation pro- cedures were used to examine the relationship between frequency and standing crop, and between frequen- cy and ANPP estimates. Data util- ized in these analyses were from the 1982 sample plots located on the three range sites in the four grazing treatments. Correlation coefficients were developed for 14 species com- paring frequency and standing crop estimates in june, and for compar- ing frequencies in June with ANPP estimates for the entire growing season. Results Correlation coefficients between estimates derived by the three sampl- ing methods are presented in Table 3 2. Wied (1967) found strong rela- tionships existed between point frame frequency and standing crop of individual species but not between total frequency and total standing crop. He also found significant dif- ferences among species and between dates in the regression coefficients developed for predictive purposes. These differences were related primarily to differences in standing crop. Relationships between point frame frequencies and quadrat fre- quencies also varied among species. However, significant correlations were estimated for most of the more abundant species such as Texas wintergrass, red threeawn, sideoats grama, and buffalograss. Likewise, significance levels of correlation coefficients varied among species for comparisons between the two-di- mensional quadrat frequencies and standing crop estimates,’ and be- tween the two-dimensional quadrat frequencies and annual ANPP estimates. Significant \ correlations were established for most major species. A notable exception was sideoats grama. The absence of a significant correlation between fre- quency of sideoats grama in June of 1982 and estimated annual ANPP in 1982 may be related to the dynamic annual growth pattern of sideoats grama. In contrast to such species as Texas wintergrass, buffalograss, and various species of annual grasses, sideoats grama continued to grow throughout the 1982 growing season after the other species had completed most of their growth. Vegetation Trends (1960-78) The 1960 through 1978 point frame frequency data are presented in Appendix 1, illustrating the linear models developed for each treatment by soil series and plant species. On- ly statistically significant (P < 0.10) linear models are presented. Species identification codes for the linear models are presented in Table 1. Models whose intercept did not dif- fer. significantly (P > 0.10) from zero are depicted with’ a zero intercept. Models with an intercept value in ex- cess of 75 percent are presented with an intercept 10 percent less than ac- tual. Actual intercept values for 4 TABLE 2. SIMPLE CORRELATION COEFFICIENTS FOR VARIOUS PARAMETERS OB- TAINED FROM THREE SAMPLING METHODS Methods‘ Species/Species Group A vs. B’ A vs. B3 A vs. C B vs. C C vs. D Annual grasses — — 0.52 O.88** O.88** g Texas wintergrass 0.64** 0.79** O.76** O.66** 0.70** Western wheatgrass — — 0.37 — — Red threeawn — — 0.72** 0.41 0.34 Sideoats grama O.76** 0.91** 0.82** 0.59** 0.49 Hairy grama — — - 0.56 0.71* Silver bluestem — — — 0.65** 0.63** Texas grama — — — 0.60 0.58 Buffalograss 0.69** O.88** 0.73** 0.82** 0.90** Texas cupgrass — — 0.45 O.88** 0.94** Common curleymesquite —— — 0.91 ** — - f. Tumblegrass — -— — 0.71 * * 0.83** Meadow dropseed — — 0.49 0.85** 0.99** Sand dropseed — — 0.52* — — Vine-mesquite — — 0.14 — — \ Whited tridens -— — — 0.05 0.38 Other grasses 0.68** 0.67** — - — Broomweed — —— — 0.51 0.46 Forbs 0.64** 0.51** — — — Total 0.04 0.17* — — — ‘A=frequency estimates derived from point frame (1960-1978). B=standing crop. C=frequency estimates derived from 2-dimensional quadrat (1978-82). D=estimated annual ANPP. zjuly estimates from Wied (1967). 3November estimates from Wied (1967). *Significant at P <0.05. “Significant at P <0.01. these models are printed in paren- theses on the associated regression line. Clay Loam Range Site Frequency of Texas wintergrass on the Valera soils declined at a rate of about 2 percent per year in both the HC and DR treatments while frequency of red threeawn increas- ed in both treatments at a rate near 0.5 percent per year. Frequency of buffalograss/common curlymesquite on the Throck soils declined at an an- nual rate of about 2 percent per year in both the MC and DR treatments and at a rate near 3 percent per year in the EX treatments. Response of buffalograssl common curlymesquite in the HC treatment was not predic- table. Frequency of miscellaneous grasses increased slightly over time in the HC and DR treatments while frequency of red threeawn increased slightly in all three grazed treat- ments. Frequency of Texas broom- weed increased slightly in the HC and DR treatments while frequency of other forbs increased in the MC and EX treatment. The MC treat- ment was the only treatment where frequency of sideoats grama in- creased. Frequency of Texas cupgrass increased slightly and fre- quency of sand dropseed decreased slightly in the EX treatment. Frequency of Texas wintergrass on the Nukrum soils declined in the HC and MC treatments and in- creased in the EX treatment. Annual rates of response were — 1.2, — 0.6, and 1.7 percent, respectively. Response of Texas wintergrass in the DR was not predictable while fre- quency of red threeawn increased slightly in all three grazed treatments. Frequency of buffalo- grass/common curlymesquite de- clined at a low annual rate in the DR treatment (— 1.3 percent) and MC treatments (— 1.6 percent), and at a rapid rate in the Ex treatment ( - 4.0 percent). Frequency of sideoats grama and other miscellaneous grass species increased slightly in the MC treatment. Frequency of Texas cupgrass and forbs increased slight- ly in both the MC and EX treatments while frequency of sand dropseed in the exclosures decreased. l Frequency of Texas wintergrass\ on the Rowena soils declined over . _ redictable. \ “falograss/common curlymesquite time at a moderate rate in both the DR and the EX treatments. The response of Texas wintergrass in the HC and MC treatments was not Frequency of buf- declined in the MC, DR, and EX treatments at rates of — 1.9, — 1.0, and — 1.6 percent per year, respec- tively. Frequency of red threeawn increased slightly over time regardless of grazing treatment. Fre- quency of Texas broomweed in- creased slightly in both the MC and DR treatments. Frequency of Texas wintergrass on the Nuvalde soils declined at a moderate rate in all three grazed treatments. Response of Texas wintergrass in the exclosures was not predictable. Texas broomweed in- creased slightly in the DR treatment while frequency of other forbs in- creased at a moderate rate in the MC and DR treatments and at a rapid rate in the exclosures. Red threeawn increased slightly in the exclosures. Frequency of buffalograss/ common curlymesquite declined at an annual rate of -3.3 percent in the ex- closure. Response of buf- falograss/common curlymesquite was unpredictable in the grazed treatments. Rocky Hills Range Site Trends in vegetation response on the Throck soils were only predic- table in the MC treatment. Both sideoats grama and buffalo- grass/common curlymesquite de- clined at a moderate rate while fre- quency of other miscellaneous grasses increased at a moderate rate. Clay Flat Range Site Frequency of Texas wintergrass declined over time in the HC, DR, ' and EX treatments with rate of decline most rapid in the exclosures. Likewise, frequency of buffalo- grass/common curlymesquite de- Q clined in the DR and EX treatments but at a more rapid rate in the ex- closures than the DR treatment. Response of buffalograss/common curlymesquite was not predictable in the HC treatment. Frequency of iscellaneous grasses increased r slightly in the HC and EX treat- ments as did sideoats grama in the DR treatment and sideoats grama, forbs, and red threeawn in the EX treatment. Loamy Bottomland Range Site Frequency of Texas wintergrass on the F rio soils increased at a moderate rate in the MC treatment, declined at a slow rate in the DR treatment and at a moderate rate in the EX treatment. Frequency of buf- falograss/common curlymesquite declined at annual rates of —2.4, — 0.9, and — 2.8 percent in the MC, DR, and EX treatments, respective- ly. Sideoats grama increased slight- ly in the HC and DR treatments. Frequency of miscellaneous grasses increased slightly in the DR and EX treatments. Frequency of forbs in- creased at a moderate rate in the EX treatment (2.2 percent) while fre- quency of sand dropseed declined at a rate of — 1.1 percent per year. The response of the vegetation on Spur soils was only monitored in the HC treatment. Frequency of buf- falograss/common curlymesquite in- creased at an annual rate of 1.4 per- cent. Frequency of miscellaneous grasses decreased slightly while fre- quency of red threeawn increased slightly. Very Shallow Range Site Frequency of buffalograss/com- mon curlymesquite declined slight- ly on the Leuders soils in all three grazed treatments. Frequency of Texas broomweed increased slight- ly in the HC and DR and treatments. Frequency of miscellaneous grasses also increased slightly in the HC treatment. Shallow Range Site Frequency of buffalograss/com- mon curlymesquite declined at a moderate rate on the Mereta soils in the MC, DR, and EX treatments as did the frequency of Texas wintergrass in the DR and EX treatments. Red threeawn increased at a moderate rate in the MC and DR treatments. Likewise, frequen- cy of forbs and Texas cupgrass in- creased in the EX treatment, fre- quency of Texas broomweed in- creased in the DR treatment, and frequency of sideoats grama increas- ed in the MC treatment. Shallow Clay Range Site Frequency of buffalograss/com- mon curlymesquite declined on this series at a moderate rate in all four treatments. Frequency of Texas wintergrass also declined slightly in the HC treatment. Response of Texas wintergrass was not predic- table in the MC, DR, and EX treatments. Frequency of miscellan- eous grasses increased slightly in the three grazed treatments while fre- quency of Texas broomweed increas- ed in the DR treatment and frequen- cy of forbs increased in the EX treatment. Deep Redland Range Site Frequency of red threeawn on the Rowden soils increased slightly in both the HC and EX treatments while frequency of Texas winter- grass, buffalograss/common curly- mesquite, and sand dropseed de- clined in the Ex treatment. Frequen- cy of forbs increased at an annual rate of 1.7 percent in the EX treatment. Frequency (1981-82) Clay Loam Range Site Dominant grasses (greater than 5O percent frequency) on the Throck soils in the HC treatment were Texas wintergrass and buffalograss (Table 3). Subdominants (greater than 25 percent frequency) were red threeawn and tumblegrass. The dominant forb (greater than 25 per- cent frequency) was redseed plan- tain. The dominant grasses in the MC treatment were Japanese brome and Texas wintergrass with buf- falograss a subdominant. The domi- nant forbs were redseed plantain and annual broomweeds. The dominant grass in the DR treatment was sideoats grama. Subdominants were Texas wintergrass, red threeawn, and buffalograss. No forbs had a fre- quency greater than 25 percent. The dominant grass in the EX treatment was Texas cupgrass. Subdominants were Texas wintergrass, sideoats grama and slim tridens. The domi- nant forb was heath aster. On the Nukrum soils, the domi- nant grasses in the HC treatment were Texas wintergrass, buf- falograss, and tumblegrass (Table 3). 5 éozfio. 4VCQQUHMQ Zwiuuwamw.awciuoawo 33. BSwUOE vcm >392 3 mENSw ”£9PE2I’“"»’$II2" “I**°P 3 |_ xm. om >m mm P20~E w? £62m Jfizzmzmm 2O $6253 OZKP mo» mU<~E>< $2 £3<> 2625mm”: .3232» I < I E I P lllll II IDPIP m ~°2" mO UPZ UI _=: >2u¢¢ P ._. 3 m Pm N N ._.. ¢N @|-OO3(N@(\I \_@ ‘—fi'ffi\—o\ CW | | | I"- Pfi|\(\l P-Lfi \—x— v— lml m ._. x0 moo| N xm 9 "MI IPI @Lfi R Pm RE2’°EI+ |m@:2*$|»~ N N 9 R mP OON=—® |Pfi x0 Lfi \—(\l Nf\l<1'@ PW v- © P wfl“ ||— 2 Li’) I@N|”PS||'I” |Q§Pl““I|”lP“ 9. 0P mo UPZ OI xw 1:302 >35 2. ll II- |©mmmmrw |r_ ‘Ooomchmm F'1\O (NW3 v- IQQ II-“Q I |I—"\I°°5" mo OE UI _Emo._ >20 P 2. v m: NP k0 N v- I£€"+ ’3I*»"IQ xw msoscrcou MCOTMQ>HUE UCN UIM . Q ”“llP“l" 2 "”IlP“II NP mm mm IN <1’ |<2'<"\N~— C“? |E~o._ c5nwwoc, QEEU Q3216? Pobcemwm Evcwczwmv Etomvoo? >>o=w> Bozusm .@,:_$§~m P6200 .23 EmwI 535E223 >>o._ zficcomi vwwBEQEQ 3x2. m_3:cc< mwtwm zcm EwCxDZN amtwm zom x. PP MEOQUQ£ UCOEEDHD .N .P commwm CCG>> flfiCCy-Un- \ Q3216? PEQEEEQ cfitzfiq uwwmvwz 353w Ema wwEP x3>w Ewm>cw<< ww3=c 3xwP 6:8 Pzwmwzztzom m_3:=:< .N .P co3wm _ooU .< mp5» .__ wzswmwE oE> 32:: E=m 26E: $E>> vwwWQQu vcmm @3305 32322 mmflww QPCDP EauEw £31 N w.~_:vmwE>_.5u coEEoU mwfimaau @382. 32:: EEI QoEozou mco~c< mmfiwofitsm E932 6>=m 3E2 3x2 362w 5.31 382w flmowEm c3332: vwm EQGQDE m5 mEccwbwm m_m:¢:< .~ .P co3wm 2:32P .m mmfiwwqzn 3x2. 3.362.; 3x2. mmflmumwg? c$$w>> xflccmtwm @382 $_ww>>x_m 5:3 2P5 wEEQ owwcmamP m_m:cc< N .P co3mm _ooU .< 333.0 ._ ~25 P Z_ Ea P 2O 3563mm”. P ._.m> $01» 5m? OmOD._UZ_ 863% .m_2<~: EU m» 32¢»: PZU oP >m oP EONE $2 862m 4 O>>._. $05,‘ mfi1 ZO £63m ._.Z AJNHHUBEA zoHP<»om Qwmmwlwm m5; m2: 5.2 ‘ . ma: - 82 3% E 8 l? i‘? E?» 3 E (Z) AJNHHUEIHA m:Q:zHHz@Q >>30 5mm m~2m> 320A >30 5mm x0025. 2> .. m€.-..s~.m7¢‘.£.m7. $2 g ‘ . R. - Sm l a. . E - 8 L E P (Z) A3N3flU38d (Z) AJNHHUHHA $2 . E 8 . sm . 3 . E . am .. E. mmDmQ4@Xw . m . R2 . 5.2 . ma: E9 um, em .5“ B8 . am . sm . s.‘ . sm . 8 . Q“ mpozzfikzou wH 85¢“ . mwmfi cg: - s amass’ mamvmcufi . @@m_ SSSQSS §CDLDQCYIGI§ m:Q:zH~zQQ >>20 3mm E55? Z SEGA M . m~mH . JA . . @§@_ . m@@_ . @@m_ z O|E< QZZ< . E ea» w::\§sa 1 am ism s... Sm Sm 5 wm3m@4QXw l m $2 SSSSSSS l\(Dl.D<'C77(\lv-4 mD@3zHHz@@ wFqu . H5 z@H»<»Qm Qwmmwgmm . I 4 1 4 J S S S 3 S S S l\ (D If) i’ (T) (U v-l m:Q:zHHzoQ >>30 QQwBoM 2- 4 .$.2 . - + +82: ea .5 4 SN .-% w... .9 $5 Q. .@ E U8 .5 mzpmodvm L R2 82%;? $2 82 mm» $032228 MZWGQQ: $SSSSSS I\(DLDQ('O('\Iv-¢ 15> RE. - . gas. . - Lmfl m 3 m M U w zQCkEm Qmmmfiwg m |J w m M w m. EQOJ >.£U oEm>sZ S204 > m 1\\ . m§mH . . @>m_ . . . . mmwfi . . . . swmfi . 1 4 m>@H . ‘ , . s»@_ . . . J mm@_ . . =@m_ | a8 1 SN mu dam Sm mm . Sm :4 . av mm 18¢ |A , Sm m 4 am aw - 8 E. . E. m:Q:zH~zQ@ wk>§ mhmfi . s~m_ . “UH: . uzzaa m:o:zHFzou >>30 8% OZ2 A3N3flU38d 7% S l\ EaoA >30 >255 x825. 31mm 550m mum mm=w@4QXw ads SQSSSSS §CDLDQ'CYJC\IP-l (Z) A3N3flU3Hd >20 >m~oo1~ .59; E30 m~mH . . @~mH . . . mwmfi . m5 z@H~> W- . m5" . 1 5.9 . . . < £2 32 E .3 um; . E ud . m Sm nu mu 3 § IA ; w E . E l E. zQHH<»Qm Qwmmmmwg . E2 E JJ SN “a . mm n mm m . mm aw m/w Sm . so 1 E. m:Q:zHFzQ@ >>20 mama/O W>oqq20 cow>>om QZ>20 5mm 2236A >>oqq>20 QHPBZ >>o44< 1.8 + 32) . . . . . . . .. Degrees Fahrenheit a 4v!’ nruumuuahaaehmhaummnuzhm.zras ‘a [Blank Page in Original Bulletin] ' 5 ‘s: Q Mention of a trademark or a proprietary product does not constitute a guarantee or a warranty of the product by the Texas Agricultural Experiment Station and does not imply its approval to the exclusion of other products that also may be suitable. All programs and information of The Texas Agricultural Experiment Station are available to everyone without regard to race, color, religion, sex, age, handicap, or national origin. 2M—1-86