A herpetological survey of three old-field sites at cooper lake in northeast Texas.
Quantitative ecological surveys have received increased attention in recent years (Ricklefs & Schluter 1993; Dodd & Franz 1993; Haila 1996). This has been prompted by general observations of noticeable declines among numerous taxa in a variety of ecological settings (Haila 1996; Haila & Margules 1996). In addition, recent political and public trends have created a mandate for the restoration of natural communities. Restoration practices require base-line ecological survey data to assess the nature of existing species assemblages, to assess the effectiveness of management practices, and to develop an understanding of how and why diversity varies among localities within the same ecosystem (Haila 1996). Unfortunately, restoration ecologists are chronically faced with situations in which natural, undisturbed communities are rare, extremely fragmented, or in many cases no longer exist. In the absence of suitable reference communities, it becomes necessary to assess local diversity among disturbed communities and to use these data to make inferences about the ecology and species diversity of pre-existing undisturbed communities (Haila 1996).
Texas is divided into several distinct ecoregions (Blair 1950; Gould 1975), one of which is the Blackland Prairie. This region, formerly dominated by tall-grass prairie, occupies a narrow belt from San Antonio to Paris and is one of the most endangered large ecosystems in North America. Of the original 4.3 million hectares of Blackland Prairie, less than 10 percent remains unplowed in small, scattered remnants whose vegetation composition has been altered by human activities and the termination of natural disturbances critical to the maintenance of prairie ecosystems (Diamond & Smeins 1993). This has created problems for those interested in restoring Blackland Prairie ecosystems. In particular, community associations within the Blackland Prairie ecoregion are poorly understood. While there is some information on basic floristic associations (Gould 1975; Diamond & Smeins 1993), knowledge of faunal associations is scanty, even for disturbed sites (Schmidly et al. 1993; Wilkins 1995). This is particularly true of herpetological associations. Thus, while Dixon (1987) has provided herpetological records for many Blackland Prairie counties, no quantitative survey data on reptile abundances and associations have been published for this region (Schmidly et al. 1993).
This study was undertaken to conduct a herpetological survey of larger species of terrestrial reptiles at three old-field sites in the northeastern portion of the Blackland Prairie ecoregion of Texas. These sites consisted of tracts of abandoned agricultural land, all of which contained soils of the alfisol group that originally supported Blackland Prairie vegetation (Hallmark 1993). The primary objective was simply to document the relative abundances of reptile species observed at these sites. To characterize these sites for future comparisons, and to better define the nature of the sites used, quantitative data on vegetation at each site is also presented. Collection methods incidentally captured a number of potential prey species. Since these species provide additional information on the ecological setting of the study sites and because they could be relevant to herpetological abundance and diversity, a brief description of these prey species is also included.
MATERIALS AND METHODS
Study areas. -- The three areas used in this study were located in the wildlife management areas along the north side of Cooper Lake in Delta Co., Texas (Fig. 1). All three sites contained soils of the Wilson/Normangee/Crocket Alphisol group (Ressel 1979) that formed under tall-grass prairie vegetation, most likely of the silveanus dropseed/meadsedge association (Diamond & Smeins 1993). The sites had been used for agricultural purposes until the early 1980's when the land was purchased by the U.S. Army Corps of Engineers and was set aside to be managed by the Texas Parks and Wildlife Department. Thus, at the time that the current study was conducted (1995), the sites had been idle for at least 10 years. An effort was made to select sites that were as similar as possible; however, there were some differences in topography and size. A brief description of the sites follows.
[FIGURE 1 OMITTED]
The Dogtown site was located at the northeast end of Cooper Lake, 4.8 km southwest of the town of Klondike in Delta County. The site consisted of 12 ha of open habitat. It was bounded on the north and east by lake shoreline and on the south and northwest by riparian forests dominated by hickory (Carya sp.) and oak (Quercus sp.). Along the west margin of this site was a railroad right of way lined with honey locust (Gleditsia triacanthos), hercules club (Zanthoxylum clavaherculis), bois d'arc (Maclura pomifera), hawthorne (Crataegus sp.), sugarberry (Celtis laevigata) and green ash (Fraxinus pennsylvanica). The Dogtown site was somewhat sloping and averaged about 6.2 m above the lake level. The vegetation was mainly grasses and forbs, but a few isolated small trees were scattered throughout the site, mainly bois d'arc, honey locust, plum (Prunus sp.), red cedar (Juniperus virginiana) and persimmon (Diospyros virginiana).
The Friendship site was 4.0 km northeast of the Dogtown site and 2.0 km south of the town of Klondike. It contained 24 ha of open habitat and was bounded to the east by Cooper Lake. Along the south and west margins were early second growth forests dominated by green ash, water oak (Quercus nigra), post oak (Quercus stellata), honey locust and sugarberry. To the north the site was bounded by a riparian forest of water oak and willow oak (Quercus phellos). The site was flat to slightly rolling, poorly drained, and averaged 3.0 m above the lake level. The vegetation was dominated by broomsedge bluestem and only a few isolated trees (red cedars) were evident.
The Liberty Grove site was 4.0 km east of the Friendship site and 5.0 km from the town of Klondike. It was the largest of the three sites and consisted of 36 ha of open habitat. Located on a peninsula, it was bounded by water to the north, east, and south. A hedgerow of cedar elm (Ulmus crassifolia) and sugarberry formed the western side of this site. The site itself was extremely flat, poorly drained, and averaged about 2.0 m above the lake level. The vegetation was dominated by grasses and forbs and numerous clumps of baccharus (Baccharus sp.) were evident.
Trapping methods. -- Large reptiles (primarily snakes) and some potential prey species were captured using drift fences with funnel traps. The drift fences and funnel traps were constructed of 0.6 cm mesh hardware cloth, the drift fences being 60 cm in height. The fences were sunken 8 cm into the ground and were laid out in a "Y" configuration. Each of the three radiating arms of the layout was 17 m long and held two 0.9 m funnel traps, one 3.8 m from the center and one 3.8 m from the end. Thus, each set of drift fences contained six funnel traps. To prevent harm to the animals, each funnel trap contained a dish of water and was shaded with a board. One set of drift fences was placed near the center of each study site.
The funnel traps were checked daily from 30 June 1995 through 30 September 1995, resulting in an overall trapping effort of 1674 trap-days, or 558 trap-days per site. The traps were checked either in the morning around 1000 h or in the evening around 1700 h. Each capture was identified to subspecies and marked by clipping a unique combination of tail scales just posterior to the cloaca. To minimize handling for safety reasons, specimens of the venomous western cottonmouth (Agkistrodon piscivorus leucostoma) and southern copperhead (Agkistrodon contortrix contortrix) were not marked. However, cottonmouths exhibited considerable variation in size and pattern and recaptures of this species were done by consulting field notes. Only one southern copperhead was captured; hence recaptures of this species did not factor into the analysis. Once the data had been recorded, individuals were released about 10 m away from the drift fences to avoid immediate recapture.
During the survey several potential prey species were also captured. These consisted of Woodhouse's toads (Bufo woodhousii woodhousii), southern leopard frogs (Rana sphenocephala), fulvous harvest mice (Reithrodontomys fulvescens), white-footed mice (Peromyscus sp.) and hispid cotton rats (Sigmodon hispidus). These captures were noted and the animals were released unmarked.
Plant sampling. -- To characterize each site, a survey of plant species found adjacent to the drift fence arrays was conducted. This was done by estimating coverage in each of ten randomly selected 1 [m.sup.2] quadrats at each site. Quadrat locations were determined by using the center of the drift fence array as an origin and by generating random compass directions (0 - 360[degrees]) and distances (0 - 50 m) using a random number generator.
For each quadrat a visual estimate was made of the overall percent cover of live grasses, dead vegetation, and live forbs. Percent cover was then estimated for the five most dominant species of grass and the five most dominant species of forbs (where dominance was assessed in terms of cover). Grass species were identified using the keys provided by Gould (1975) while forbs were identified using Ajilvsgi (1984), McGregor & Barkley (1986) and Hatch & Pluhar (1993).
Herpetological data. -- When recaptures were excluded, a total of 60 individuals representing 12 species (11 species of snakes and one species of lizard) were represented in the data (Table 1). An evaluation of the chronological sequence of captures indicated that individuals were captured at a relatively constant rate throughout the study period. However, 10 of the 12 species represented in the data (83.3%) were captured within the first 22 days of the 94 day study, suggesting that the sampling effort was sufficient to document most of the species present at the sites.
When captures at all three sites were considered together (Table 1), the three most abundant species were the western cottonmouth (23% of all captures), the eastern yellow-bellied racer (Coluber constrictor flaviventris) (22% of all captures), and the Texas garter snake (Thamnophis sirtalis annectans) (13% of all captures). These three species accounted for 58% of all individuals captured during the survey. Most of the remaining nine species were present at intermediate abundances; however, three species, the eastern coachwhip (Masticophis flagellum flagellum), the blotched watersnake (Nerodia erythrogaster transversa) and the southern copperhead were represented by only a single individual.
There were no significant differences in the total number of individuals captured at each site (Goodness of Fit Test, [X.sup.2] = 0.30, df = 2, P > 0.90). Nine species were captured at both the Dogtown and Liberty Grove sites while six species were captured at the Friendship site. Thus, there were also no significant differences between sites in the total number of species captured (Goodness of Fit Test, [X.sup.2] = 0.75, df = 2, P = 0.75).
There were differences among sites in the relative dominance of particular species. The Dogtown site was dominated by western cottonmouths, which accounted for 49% of the individuals captured there. At Liberty Grove the dominant species was the yellow-bellied racer, which accounted for 37% of the captures at that site. A Fisher's Exact test confirmed that the relative abundances of these two species differed among the two sites (P = 0.0048). The Friendship site seemed to have the most even distribution of species abundances. At this site the most frequently captured species, the Texas garter snake, accounted for about 32% of the captures and its abundance was closely followed by that of western cottonmouths, eastern yellow-bellied racers and prairie kingsnakes (Lampropeltis calligaster calligaster).
Vegetation data. -- The dominant species of grass at all three sites was broomsedge bluestem (Table 2) and the percent cover provided by this species did not differ significantly among sites (Kruskal-Wallis Test, [X.sup.2] Approximation, [X.sup.2] = 0.1.71, df = 2, P = 0.425). The three sites did differ in the number of species of grass encountered (Kruskal-Wallis Test, [X.sup.2] Approximation, [X.sup.2] = 8.00, df = 2, P < 0.025). Fifteen species of grass occurred at the Dogtown site, nine species of grass were encountered at the Liberty Grove site.
The dominant forbs differed among sites (Table 3). The Dogtown site was dominated by sheepweed (Croton texensis), the Friendship site by ragweed (Ambrosia psilostachya) and the Liberty Grove site by late flowering boneset. The three sites did not differ in the number of species of forbs (Kruskal-Wallis Test, [X.sup.2] Approximation, [X.sup.2] = 0.608, df = 2, P > 0.100).
The overall cover provided by live grasses, dead vegetation, and forbs also differed among sites (Table 4; Kruskal-Wallis Tests, [X.sup.2] Approximation; Live Grasses, [X.sup.2] = 16.63, df = 2, P = 0.0002; Dead Vegetation, [X.sup.2] = 15.23, df = 2, P = 0.0005; Forbs, [X.sup.2] = 14.10, df = 2, P = 0.0009). Almost three-quarters of the cover at the Friendship site was in the form of live grasses. At the Liberty Grove site about half of the cover was in the form of dead vegetation while the remaining cover was about equally distributed among live grasses and forbs. About half of the coverage at the Dogtown site was live grass, but an additional 29.6% of the cover was in the form of forbs.
Potential prey species. -- Three varieties of small mammals and two species of anurans were regularly caught in the funnel traps. Small mammal captures were stratified into those greater than 50 g (hispid cotton rats) and those less than 50 g (fulvous harvest mice, Peromyscus sp.). Anurans, which included Woodhouse's toads and southern leopard frogs, were considered as a single prey category. The distribution of these captures is shown in Table 5. In general, more of these potential prey species were captured at the Liberty Grove site than at the Dogtown site. Mammals weighing less than 50 g were most abundant at the Friendship site and least abundant at the Dogtown site. Mammals greater than 50 g were most abundant at the Liberty Grove site and least abundant at the Dogtown site. Anurans were most abundant at the Dogtown site and least abundant at the Liberty Grove site. These differences were statistically significant (3 by 3 Contingency Table; [X.sup.2] = 151.75, df = 4, P = 0.0001).
The selected method of capture utilized in this study provided relative abundance data for 12 species of reptiles. However, smaller species such as prairie ringneck snakes (Diadolphis punctatus arnyi), rough earth snakes (Virginia striatula), and Texas brown snakes (Storeria dekayi texana) which were observed near the capture sites were not represented in the sample. Use of a mesh size smaller than 0.6 cm in the drift fences and funnel traps would have significantly increased the likelihood of capture for these and other species, since small reptiles were capable, and even observed, to pass through the 0.6 cm mesh. Likewise, had a smaller mesh size been used, more individuals of the species represented in the data, in the form of hatchlings and small juveniles, would have been captured. The advantages and disadvantages associated with the use of funnel traps and drift fences are described in detail by Fitch (1992). Given the constraints inherent to this study, data was evaluated within the context of the ecology of the adult and subadult populations of the species surveyed.
None of the species encountered in this study were unexpected, since they have previously been documented either in or near Delta County (Dixon 1987). However, the herpetofauna of Delta County has been poorly documented. Thus, while speckled kingsnakes (Lampropeltis getulus holbrooki), eastern hognosed snakes (Heterodon platyrhinos), western slender glass lizards (Ophisaurus attenuatus attenuatus), eastern coachwhip snakes, blotched watersnakes, western ribbon snakes (Thamnophis proximus proximus), Texas garter snakes and southern copperheads were observed during this study, there are no official records for these species in Delta County (based on Dixon 1987 and distribution records cited in Herpetological Review from 1985 to 1998).
According to Schmidly et al. (1993), the fauna of the Blackland Prairies consists of a mixture of species with affinities to the eastern woodlands, central plains, southwestern deserts, and southeastern coastal plains. However, relative to herpetological fauna, 60% of Blackland species have faunal affinities in the eastern forests. At the species level, all twelve of the species captured during this survey have distributions that extend well east of the Blackland Prairies. However, four of the subspecies captured, the prairie kingsnake, the Texas garter snake, the eastern yellow-bellied racer and the blotched water snake are western subspecies, and could be considered as typical of plains and prairies. In particular, the garter snake, the racer, and the watersnake subspecies captured in this study are at the eastern limits of their geographic distributions (Dixon 1987). Categorized in this way, 67% of the species and subspecies captured in this study have eastern wood-land affinities, a figure similar to that cited by Schmidly et al. (1993).
Three species that were anticipated but were not observed in this study included the plains rat snake (Elaphe guttata emoryi), the dusty hognosed snake (Heterodon nasicus gloydi), and the bullsnake (Pitouphis melanoleucus sayi), since Delta County lies within the geographic limits of all three (Tennant 1985; Dixon 1987). However, these species seem to prefer dry, sandy or rocky prairies (Tennant 1984; Conant & Collins 1991) and the poorly drained clay soils found on the sites selected for this study may have precluded the occurrence of these species.
Although similar numbers of individual reptiles were captured at each site, the sites differed considerably in the number and types of species present. This was evident in terms of the dominant species at each site and in the types of species present at each site. These differences were somewhat surprising given that the three sites were all in the same local area and represented the same general habitat type. However there were significant differences in vegetation, particularly with respect to grass diversity and the dominant types of cover. These differences could correspond to differences in the type and distribution of microhabitats among sites. Thus, some of the differences among sites in the types and numbers of reptiles captured might be attributable to the observed differences in vegetation. For example, the Friendship site had a slightly lower number of reptile species. This site also appeared to be the most homogeneous relative to vegetation, being mostly covered by grass and containing few forbs. Thus, a lack of microhabitat diversity at this site might account for the slightly lower diversity of reptiles observed there. Similarly, it could be argued that any number of the observed vegetation parameters could be operating synergistically to cause the observed differences among sites in reptile species composition.
Some observations suggest that factors other than microhabitat differences were also responsible for the differences in reptile species observed among sites. The occurrence of some species at some sites might be tied to the nature and extent of surrounding habitats. For example, ribbon snakes, cottonmouths and watersnakes are more typically associated with aquatic habitats and would not be expected to dominate in dry upland areas. Thus, it was surprising that the cottonmouth was so common on all study sites and was even the dominant species at the Dogtown site. Undoubtedly, the proximity of these study sites to Cooper Lake facilitated the presence of this species. The Dogtown site, in particular, is adjacent to an extensive area of riparian woods, much of which is partially inundated either periodically or intermittently, and which provides excellent habitat for cottonmouths. The proximity of these woods to the Dogtown site likely makes the Dogtown site more prone to incursions by cottonmouths.
The abundance of prey items may also affect reptile distributions (Seigel & Collins 1993). The most common prey item reported for western cottonmouths are anurans (Tennant 1984; Ernst & Barbour 1989). This may also help explain why cottonmouths were so common at the Dogtown site where anurans were most abundant and why cottonmouths were least common at the Liberty Grove site where anurans were least frequent. Similarly, prairie kingsnakes, which prefer small mammals (Tennant 1984), were most frequent at the Friendship site where the most small mammals (< 50 g) were captured, while the larger Texas rat snake (Elaphe obsoleta lindheimeri), which also prefers warm-blooded prey (Tennant 1984; Ernst & Barbour 1989), was most common at the Liberty Grove site where the larger small mammals (> 50g) were most abundant.
These data show that reptile assemblages on former tall-grass prairie sites can vary considerably, but the data do not allow an adequate interpretation of these variations relative to microhabitat, prey availability or surrounding habitat. However, the fact that variations of this sort exist among local communities in similar habitats suggests that a variety of ecological processes are operating collectively to promote these variations. Existing knowledge of vegetation on the Blackland Prairies indicate that, at the landscape level, the Blackland Prairie is a complex mosaic of unique communities influenced by a dynamic interplay of disturbance factors, soil, and moisture regimes (Diamond & Smeins 1993). The results of this study suggest that the herpetological communities on the Blackland Prairies might reflect a similar complexity. For meaningful ecological restoration of tall-grass prairie ecosystems, it is imperative that restoration ecologists consider surrounding unmanaged ecosystems and the effects of management practices on potential prey populations and ecological heterogeneity as they, in turn, affect herpetological assemblages. This study, in addition to providing a baseline indicator of northern Blackland Prairie herpetological fauna, also reinforces the need for further, more comprehensive survey research, with emphasis on both local and landscape level ecological processes.
Table 1. Number of individuals of each species captured at each study site (recaptures excluded). Species Dogtown Friendship Liberty Grove Total Agkistrodon piscivorus leucostoma 9 4 1 14 Lampropeltis getulus holbrooki 3 0 1 4 Ophisaurus attenuatus attenuatus 3 0 0 3 Coluber constrictor flaviventris 2 4 7 13 Lampropeltis calligaster calligaster 1 3 1 5 Heterodon platyrhinos 1 1 1 3 Elaphe obsoleta lindheimeri 1 0 4 5 Thamnophis proximus proximus 1 0 1 2 Masticophis flagellum flagellum 1 0 0 1 Thamnophis sirtalis annectans 0 6 2 8 Nerodia erythrogaster transversa 0 1 0 1 Agkistrodon contortrix contortrix 0 0 1 1 Total Individuals 22 19 19 60 Total Species 9 6 9 12 Table 2. Grass species observed at each study site. Numbers indicate mean percent cover [+ or -] standard deviation. Grass Species Dogtown Friendship Andropogon virginicus 32.7 [+ or -] 27.8 41.1 [+ or -] 33.3 Dicanthelium sp. 17.4 [+ or -] 19.1 5.5 [+ or -] 9.0 Carex (sp. a) 4.0 [+ or -] 2.6 * Setaria geniculata 2.1 [+ or -] 6.3 12.8 [+ or -] 18.6 Sporobolus asper 1.6 [+ or -] 4.4 * Andropogon ternarius 1.5 [+ or -] 4.7 * Eragrostis intermedia 0.6 [+ or -] 1.6 * Aristida longispica 0.4 [+ or -] 1.0 * Carex (sp. b) 0.3 [+ or -] 0.5 1.5 [+ or -] 4.7 Coelorachis cylindrica 0.2 [+ or -] 0.6 * Tridens flavus 0.2 [+ or -] 0.6 * Digitaria sanguinalis 0.2 [+ or -] 0.4 * Paspalum floridanum 0.1 [+ or -] 0.3 * Aristida oligantha 0.1 [+ or -] 0.3 * Bothriochloa saccharoides 0.1 [+ or -] 0.3 * Panicum hians * 9.6 [+ or -] 24.8 Unknown Grass * 1.5 [+ or -] 4.7 Agrostis hiemalis * 1.1 [+ or -] 3.1 Cynodon dactylon * 0.9 [+ or -] 1.5 Grass Species Liberty Grove Andropogon virginicus 20.6 [+ or -] 11.5 Dicanthelium sp. * Carex (sp. a) * Setaria geniculata 0.2 [+ or -] 0.6 Sporobolus asper * Andropogon ternarius * Eragrostis intermedia * Aristida longispica * Carex (sp. b) * Coelorachis cylindrica * Tridens flavus * Digitaria sanguinalis * Paspalum floridanum * Aristida oligantha * Bothriochloa saccharoides * Panicum hians * Unknown Grass 3.0 [+ or -] 3.5 Agrostis hiemalis * Cynodon dactylon * Table 3. Species of forbs observed at each study site. Numbers indicate mean percent cover [+ or -] standard deviation. Forb Species Dogtown Friendship Croton texensis 8.7 [+ or -] 14.2 * Artemisia sp. 7.8 [+ or -] 24.3 2.1 [+ or -] 6.3 Ambrosia psilostachya 7.3 [+ or -] 13.4 0.2 [+ or -] 0.4 Rubus sp. 6.9 [+ or -] 21.8 0.1 [+ or -] 0.3 Trifolium sp. 2.5 [+ or -] 6.2 0.6 [+ or -] 0.7 Neptunia lutea 0.8 [+ or -] 1.5 * Unknown Forb sp. a 0.3 [+ or -] 0.7 0.2 [+ or -] 0.4 Aster ericoides 0.1 [+ or -] 0.3 * Asclepias viridis 0.1 [+ or -] 0.3 * Eupatorium serotinum * 1.1 [+ or -] 3.1 Solidago sp. * * Sida sp. * * Verbena sp. * * Agalinis purpurea * * Forb Species Liberty Grove Croton texensis * Artemisia sp. * Ambrosia psilostachya * Rubus sp. 0.5 [+ or -] 1.6 Trifolium sp. * Neptunia lutea * Unknown Forb sp. a 3.6 [+ or -] 3.8 Aster ericoides 0.3 [+ or -] 0.9 Asclepias viridis * Eupatorium serotinum 14.3 [+ or -] 12.6 Solidago sp. 2.5 [+ or -] 5.0 Sida sp. 0.1 [+ or -] 0.3 Verbena sp. 3.9 [+ or -] 6.3 Agalinis purpurea 1.1 [+ or -] 3.1 Table 4. Percent coverage of grasses and forbs observed at each site. Numbers indicate mean percent cover [+ or -] standard deviation. Dogtown Friendship Liberty Grove Live Grass 51.2 [+ or -] 24.0 72.3 [+ or -] 10.4 24.2 [+ or -] 11.1 Dead Grass 19.3 [+ or -] 16.3 24.3 [+ or -] 8.9 49.6 [+ or -] 15.3 Forbs 29.6 [+ or -] 26.2 3.5 [+ or -] 6.5 26.2 [+ or -] 12.4 Table 5. Number of individuals of each type of potential prey species captured at each study site. Prey Type Dogtown Friendship Liberty Grove Mammals < 50 g 5 34 14 Mammals < 50 g 14 65 103 Anurans 42 10 1 Total 61 109 118
We thank the U.S. Army Corps of Engineers and The Texas Parks and Wildlife Department for permission to work on the public lands surrounding Cooper Lake. In particular we thank David Sierra and Larry Lebeau, both of TPWD, for assistance in establishing study sites. Ken Steigman of the Heard Museum in McKinney, Texas, is thanked for his valuable advice and for lending us some of his equipment. Financial assistance was provided by a Texas A & M University-Commerce Mini Grant awarded to JGK.
Ajilvsgi, G. 1984. Wildflowers of Texas. Shearer Publishing, Bryan, Texas. xvii + 414 pp.
Blair, W. F. 1950. The biotic provinces of Texas. Texas J. Sci., 2(1):93-117.
Conant, R. & J. T. Collins. 1991. A field guide to the reptiles and amphibians of eastern and central North America. Houghton Mifflin Company, Boston, xx + 450 pp.
Diamond, D. D. & F. E. Smeins. 1993. The native plant communities of the Blackland Prairie, Pp. 66-81, in The Texas Blackland Prairie: Land, History, and Culture (M. R. Sharpless & J. C. Yelderman, Jr., eds.), Baylor Univ. Program for Regional Studies, Waco, Texas, xvi + 369 pp.
Dixon, J. R. 1987. Amphibians and reptiles of Texas. Texas A & M University Press. College Station, Texas, xii + 434 pp.
Dodd, K. C., Jr. & R. Franz. 1993. The need for status information on common herpetofaunal species. Herpetol. Rev., 24(2):47-49.
Ernst, C. H. & R. W. Barbour. 1989. Snakes of Eastern North America. George Mason University Press, Fairfax, Virginia, 282 pp.
Fitch, H. S. 1992. Methods of sampling snake populations and their relative success. Herpetol. Rev., 23(1): 17-19.
Gould, F. W. 1975. The grasses of Texas. Texas A & M University Press, College Station, Texas, viii + 653 pp.
Hatch, S. L. & Pluhar. 1993. Texas range plants. Texas A & M University Press, College Station, Texas, 326 pp.
Haila, Y. 1996. Quantitative surveys in biodiversity research. Ecography, 19(3):321-322.
Haila, Y. & C. R. Margules. 1996. Survey research in conservation biology. Ecography, 19(3):323-331.
Hallmark, C. T. 1993. The nature and origin of the Blacklands soils, Pp. 41-64 in The Texas Blackland Prairie: Land, History, and Culture (M. R. Sharpless & J. C. Yelderman, Jr., eds.), Baylor Univ. Program for Regional Studies, Waco, Texas, xvi + 369 pp.
McGregor, R. L., II & T. M. Barkley. 1986. Flora of the great plains. Univ. Press of Kansas, Lawrence, Kansas, vii + 1392 pp.
Ressel, D. 1979. Soil survey of Lamar and Delta Counties, Texas. United Stated Department of Agriculture, Soil Conservation Service, in cooperation with Texas Agricultural Station, 148 pp.
Ricklefs, R. E. & D. Schluter. 1993. Species diversity: an introduction to the problem, Pp. 1-10 in Species diversity in ecological communities, historical and geographic perspectives (R. E. Ricklefs & D. Schluter, eds.), University of Chicago Press, Chicago, 414 pp.
Schmidly, D. J., D. L. Scarbrough & M. A. Horner. 1993. Wildlife diversity in the Blackland Prairies, Pp. 82-95 in The Texas Blackland Prairie: Land, History, and Culture (M. R. Sharpless & J. C. Yelderman, Jr., eds.), Baylor Univ. Program for Regional Studies, Waco, Texas, xvi + 369 pp.
Seigel, R. A. & J. T. Collins. 1993. Snakes: Ecology and Behavior. McGraw-Hill Inc., New York, xvi + 414 pp.
Tennant, A. 1984. The snakes of Texas. Texas Monthly Press, Austin, Texas, 561 pp.
Tennant, A. 1985. A field guide to Texas snakes. Texas Monthly Press, Austin, Texas, 260 pp.
Wilkins, K. T. 1995. The rodent community and associated vegetation in a tallgrass Blackland Prairie in Texas. Texas J. Sci., 47(4):243-262.
Jeffrey G. Kopachena and Cynthia S. Kollar*
Department of Biological and Earth Sciences
Texas A & M University-Commerce
Commerce, Texas 75429-3011
Department of Agriculture and Wildlife Science
Texas A & M University-Kingsville
Kingsville, Texas 78363
JGK at: firstname.lastname@example.org
|Printer friendly Cite/link Email Feedback|
|Author:||Kopachena, Jeffrey G.; Kollar, Cynthia S.|
|Publication:||The Texas Journal of Science|
|Date:||Aug 1, 1999|
|Previous Article:||Changes in the fish assemblage of an intermittent prairie stream upstream from a Texas impoundment.|
|Next Article:||Changes in understory woody vegetation in red-cockaded woodpecker (Picoides borealis) nesting clusters.|