*Department of Botany and Microbiology, University of Oklahoma, Norman, Oklahoma 73019
School of Biological Sciences, Oklahoma State University, Stillwater, Oklahoma 74078
Tall-grass prairie vegetation was analyzed in 15 Oklahoma counties. Andropogon scoparius Michx. was the most important grass in all geographic regions, but its percent composition was inversely related to the amount of precipitation. The percent composition of Andropogon Gerardi Vitman differed little between the western and central sections but increased markedly in the eastern section. Sorghastrum nutans (L.) Nash and Panicum virgatum L. had their highest percent composition in the central region with an intermediate amount of precipitation. Diversity and basal cover did not change from west to east.
Little (1) listed the prairie dominants of Muskogee county as Andropogon Gerardi, A. scoparius, and Koeleria cristata*. Nease (3) listed four dominants in a tall-grass prairie in south-central Oklahoma, A. scoparius, A. Hallii, Panicum virgatum, and Sorghastrum nutans. Kelting (4) described the dominants in a virgin prairie in McClain county as A. scoparius, P. virgatum, and S. nutans. Rice (5) listed the dominants in a tall-grass community in Marshall county as S. nutans, P. virgatum, and A. Gerardi.
Quantitative analyses of isolated tall-grass prairie stands have been reported for most geographic regions of Oklahoma, but no uniform procedures were followed and no accurate comparisons could be made of diversity, basal area, and percent composition (4, 5, 6, 7, 8, 9). In June, 1978, research was undertaken, therefore, to investigate the vegetative composition of 15 tall-grass prairie stands distributed throughout the tall-grass prairie region of Oklahoma (10).
Tall-grass prairie stands which varied in size from 3 to 80 acres were located in 15
counties (Figure 1). The physiographic provinces, average annual temperature, and average annual precipitation for each stand are listed in Table 1.
Areas for analysis were chosen after reconnaissance across the state. Information from landowners was used when available to select climax prairie stands; otherwise sites were chosen if they appeared to have climax tall-grass prairie vegetation, were unfenced, and showed no evidence of grazing damage. Nine of the 15 stands had been hay meadows, or had never been disturbed. The rest of the sites appeared to have climax tall-grass prairie vegetation, but past histories were not available. All sites had level to gently sloping topography. Parent materials and soil types varied considerably as is characteristic of the wide-ranging tall-grass prairie in Oklahoma (11).
The point contact method (12) was chosen as an appropriate method for grassland sampling because of its statistical reliability (13). A rectangular area of equal size was paced off in each prairie stand, and 30 point frames were analyzed along each of five equally spaced, parallel compass lines, running lengthwise inside the plot. Thus, a total of 1500 points was taken at each site, and a hit was recorded if a pin touched the lower one-half inch of any plant. Additionally, a list of plant species present in each stand was recorded. Crockett (7) found that 1500 points gave reliable and reproducible quantitative results in his sampling of the tall-grass prairie in the Wichita Mountains Wildlife Refuge, Oklahoma. One of us (E.L.R.) found that 1500 points gave reliable data when tested against weight-list determinations in several grassland areas in Oklahoma.
The basal area, determined by dividing the number of hits for each species by 1500 (the number of points), and the percent composition, determined by dividing the number of hits for each species by the total number of hits, were then calculated for each prairie stand. Diversity (number of species per unit area) was calculated by region, and similarity indices (14) were calculated between regions. Species with a percent composition of 20 or above were considered dominants.
Andropogon scoparius was the leading dominant in 11 of the 15 stands (Table 2). Hughes and Marshall counties, in south-central Oklahoma, had a sedge species and Sorghastrum nutans as leading dominants, respectively, Muskogee county had Andropogon virginicus as the leading dominant, and Haskell county had A. Gerardi as the leading dominant. According to our cri-
terion, four counties had two dominants and eleven counties had only one dominant.
The percent composition of A. scoparius declined from west to east (Table 3). Andropogon Gerardi differed little between the west and central regions, but increased greatly in importance in the east. Sorghastrum nutans and P. virgatum were most prominent in the central part of the state.
There were no significant differences between regions in basal area or in diversity (Table 4). The total number of species in each region, however, increased slightly from east to west. The similarity indices were about the same for the west vs. central and central vs. east regions, but the index was much lower between the west and east regions (Table 4).
Rainfall apparently affects the percent composition of the prairie dominants in Oklahoma, as the composition of A. scoparius was inversely related to the amount of precipitation (Tables 1, 3), and A. Gerardi was most important in the eastern region, which has the highest precipitation. Sorghastrum nutans and P. virgatum had their highest percent composition in the central region, which has an intermediate amount of precipitation. This variation in species composition from east to west was reflected in the lowered similarity index between the east and west regions.
Apparently the varying climatic factors have little effect on diversity or basal cover of plants in our tall-grass prairie, since they did not differ from east to west. This is noteworthy in view of the fact that diversity in our upland forests increases greatly from west to east (15).
Andropogon scoparius was by far the most important grass in all geographic regions. This agreed with Bruner (16),
who stated that the dominant grass species in the true prairie of Oklahoma is A. scoparius. He stated further that A. scoparius is the leading dominant in the drier, more exposed portions even of the "subclimax prairie" in eastern Oklahoma, but that A. Gerardi is the most "characteristic dominant" in the eastern subclimax prairie occupying slight depressions or moist slopes. The high percent composition of S. nutans in Marshall county agreed with the results of Rice (5). The average percent composition of P. virgatum in each region was consistently lower than that of A. scoparius, A. Gerardi, or S. nutans. It never had an average composition over 5%.
This project was supported by a National Science Foundation Undergraduate Research Participation Grant (SPI-77-26243) to the Department of Botany and Microbiology, University of Oklahoma.
1. E. L. LITTLE, Am. Midl. Nat. 19: 559-572 (1938).
2. U. T. WATERFALL, Keys to the Flora of Oklahoma, 5th ed., Okla. State University, Stillwater, 1972.
3. F. R. NEASE, Range Deterioration in South-Central Oklahoma, Master's Thesis, University of Oklahoma, Norman, 1948.
4. R. W. KELTING, Ecology 35: 200-207 (1954).
5. E. L. RICE, Ecology 33: 112-116 (1952).
6. P. BUCK and R. W. KELTING, Southw. Nat. 7: 163-175 (1962).
7. J. J. CROCKETT, Ecology 45: 326-335 (1964).
8. H. L. HUTCHESON, Developmental Growth of Four Species of Range Grasses in North-Central Oklahoma. Master's Thesis, Oklahoma State University, Stillwater, 1963.
9. P. G. RISSER and R. K. KENNEDY, Herbage Dynamics of An Oklahoma Tall Grass Prairie, Osage Site, U.S. IBP Tech. Rept. No. 273, Grassland Biome, Colorado State University, Fort Collins, 1975.
10. L. G. DUCK and J. B. FLETCHER, A Game Type Map of Oklahoma, Okla. Game and Fish Dept., Oklahoma City, 1943.
11. F. GRAY and H. M. GALLOWAY, Soils of Oklahoma, Oklahoma State University Agr. Expt. Sta. Misc. Publ. 56, Stillwater, 1959.
12. E. B. LEVY and E. A. MADDEN, New Zealand J. Agr. 46: 267-279 (1933).
13. W. C. WHITMAN and E. I. SIGGEIRSSON, Ecology 35: 431-436 (1954).
14. E. P. ODUM, Fundamentals of Ecology, 3rd ed., W. B. Saunders Co., Philadelphia, 1971.
15. P. G. RISSER and E. L. RICE, Ecology 52: 876-880 (1971).
16. W. E. BRUNER, Ecol. Monogr. 1: 99-188 (1931).