| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ecology |
United States Department of Agriculture-Agricultural Research Service, Jornada Experimental Range, Box 30003, MSC 3JER, NMSU, Las Cruces, New Mexico 88003-0003 USA
Received for publication October 9, 2001. Accepted for publication May 7, 2002.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key Words: Bouteloua eriopoda • Bouteloua gracilis • Chihuahuan Desert grasslands • grass–shrub interactions • intra- and interspecific interactions • Poaceae • seed production • shortgrass steppe
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
) stoloniferous grass. Shortgrass steppe grasslands located along the Front Range of the Rocky Mountains are dominated by the long-lived (400 yr; Coffin and Lauenroth, 1990
) bunchgrass Bouteloua gracilis (H.B.K.) Lag. ex Griffiths. A transition zone between biomes dominated by these species occurs in central New Mexico. Controls on patterns in species dominance at this transition zone are unknown, although soil disturbances caused by bannertail kangaroo rats (Dipodomys spectabilis) result in an increase in localized dominance by B. eriopoda and a decrease in B. gracilis that cannot be explained by differences in vegetative spread alone (Fields, Coffin, and Gosz, 1999
). Thus, it was hypothesized that these small-scale dominance patterns are due to species-specific differences in recruitment. Specifically, seed production per plant was expected to be larger for the bunchgrass B. gracilis than for the stoloniferous B. eriopoda as a result of differences in energetic demands for vegetative propagules (Schmid et al., 1988
). By contrast, seed availability in the soil on a square meter basis, one measure of recruitment success, was expected to be greater for the short-lived B. eriopoda compared to the long-lived B. gracilis since populations consisting of short-lived genets must replace plants more frequently than populations where genets live indefinitely.
At the shortgrass steppe/Chihuahuan Desert transition, the landscape consists of a mosaic of grass patches that may be dominated by one or both Bouteloua species. Another patch type of particular importance is codominance of B. eriopoda with Larrea tridentata (DC.) Coville (Zygophyllaceae), a common invasive shrub in the Chihuahuan Desert (Buffington and Herbel, 1965
). Because patch types differ in microenvironmental conditions (Kröel-Dulay, Hochstrasser, and Coffin, 1997
; Kieft et al., 1998
), recruitment processes may also differ by patch type. Alternatively, one or more processes may be primarily constrained genetically, in which case plants growing in different patch types may exhibit similar responses. Effects of grasses on shrub seedling establishment have been studied (Brown, Scanlon, and McIvor, 1998
); however, little is known about the effects of shrubs such as L. tridentata on B. eriopoda recruitment and how these grass–shrub interactions may differ from intra- and interspecific interactions among grasses.
Both Bouteloua species have been well studied within the biome where each dominates (Gadzia, 1979
; Coffin and Lauenroth, 1989
, 1992
; Lauenroth et al., 1994
). However, it is not known whether studies conducted in different parts of the geographic distributions of these species are applicable at the biome transition zone (Minnick and Coffin, 1999
). Plants growing near the edge of their species range may be near the limits of their physiological tolerance (Arris and Eagleson, 1989
). Thus, recruitment potential may be lower at a biome transition zone compared with the core of a biome.
The three specific objectives for this study were: (1) to compare seed production and presence in the soil of B. gracilis and B. eriopoda at a biome transition zone; (2) to evaluate recruitment potential in different patch types for each Bouteloua species, either with or without the other Bouteloua species or L. tridentata; (3) to evaluate regional variation in recruitment potential for one species, B. gracilis, in cases for which similar data are available from a site located within the shortgrass steppe biome.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
). Annual temperatures averaged 14.1°C (SD = 0.7°) over the same time period. During the time of this study (November 1995–April 1997), temperatures were above average and summer precipitation was below average in 1995. Summer precipitation was above average in 1996 and 1997. Mean daily temperatures were above average in the summer of 1996 and fall of 1997 and below average in the winter of 1996–1997.
A 2.5-km2 area on the McKenzie Flats (1650 m above sea level) was randomly selected containing patches that were either: (1) dominated by B. gracilis (blue grama); (2) codominated by B. gracilis and B. eriopoda; (3) dominated by B. eriopoda (black grama); or (4) codominated by B. eriopoda and Larrea tridentata (creosotebush). Patches codominated by B. gracilis and L. tridentata rarely occur. One patch of each type was randomly selected for sampling. Species composition and cover within each patch, measured annually since 1995, were compared with a large number of patches (60) to confirm the representativeness of each sampled patch to its type (Kröel-Dulay, Hochstrasser, and Coffin, 1997
; Peters, 2000b
). Soil sampling showed that sand contents in the upper 10 cm (83–86%) and 30 cm depths (73–75%) were similar for the three grass patches and higher than the B. eriopoda–L. tridentata patch (73% and 61%, respectively). Percentage clay content ranged from 5 to 9% (0–10 cm) and from 12 to 18% (0–30 cm) for grass-dominated patches, and from 11 to 12% for both depths in the grass–shrub patch.
Plant-level sampling of seed production and reproductive effort
Sampling was conducted in 1996 on 20 September (B. eriopoda) and 18 October (B. gracilis) to correspond to the peak of seed maturation, yet prior to dispersal. Sampling and experimental design were based upon a previous study of B. gracilis (Coffin and Lauenroth, 1992
) to allow comparisons between data sets. Three plants of each dominant species were randomly selected within each of five 10-m transects in each of three blocks for each patch; thus, a total of 45 plants was sampled for each species from each patch where it dominates or codominates. Individual plants of B. gracilis were defined as all tillers currently connected by a crown (Coffin and Lauenroth, 1988
). Because B. eriopoda is stoloniferous, individual clumps can be quite large (100 cm diameter) and diffuse with many loosely connected ramets (Peters, 2002
). Only small, well-defined clumps from each genet that were similar in size to B. gracilis plants (mean = 12.8 cm diameter) were selected.
Numbers of seeds and inflorescences and biomass (measured as grams of dry mass) of aboveground reproductive (inflorescences including seeds) and vegetative structures (nonreproductive tillers and crowns) were assessed for each plant by clipping material at the soil surface, separating it, then either counting or weighing it. Distribution of biomass to reproduction was calculated in two ways. First, reproductive biomass was divided by total biomass (reproductive + vegetative) as a measure of energy invested in reproductive structures. Second, the number of seeds produced was divided by total aboveground biomass as a measure of energy invested in seed availability for recruitment. Seed viability (as a percentage) for each species–patch type combination was determined using species-specific standardized procedures to break dormancy followed by a 28-d germination test and a tetrazolium test for viability. Fifty seeds were randomly selected for testing from a combined sample of all seeds collected by each species in each patch type. Mean seed mass (in milligrams) was calculated from the seeds tested for viability.
Analysis of variance (ANOVA) was used to evaluate differences between species and patch type with respect to indicators of reproductive effort. Plants were averaged within transects and the means were analyzed in a two-fold design with patch type, block within patch type, and transect within block as factors. The residuals were normally distributed for most variables. Response variables where residuals were not normally distributed were log-transformed prior to the analysis.
A priori contrasts were used to test for effects of species or patch type on the response variables (Sokal and Rohlf, 1998
). Five separate contrasts were conducted for each ANOVA if the overall F test for patch type was significant. The first contrast tested the significance of species for all patch types. The second contrast tested for the significance of the two patch types containing B. gracilis on responses of this species. The third contrast tested for the significance of the three patch types containing B. eriopoda on responses of this species; if this test was significant, then two additional contrasts were conducted to determine if responses for each pair of patches were different. A significance level of 
= 0.05 was used for all analyses.
Because sampling for seed production per plant is very time consuming, patch types were also compared using regression to analyze the relationship between seed production and one of several easily obtained measures per plant. Inflorescence density, basal area, and total aboveground biomass (reproductive and vegetative) were selected as the predictor variables. Separate regressions were first conducted for each species in each patch type. The influence of patch type on the slope and intercept of each relationship was then tested (Sugiyama and Bazzaz, 1998
). Slopes were compared pairwise using a Tukey-Kramer minimum significant difference procedure (Sokal and Rohlf, 1998
). ANOVA was used to test for differences in intercepts for those patch types where the slopes were not significantly different. Regressions for which the slope and intercept were not significantly different between patch types and within species resulted in a pooling of patch types.
Community-level sampling of seed production
Density of seeds produced on a square metre basis was calculated for each species and patch type to allow comparisons with the density of seeds found in the soil per square metre (see below). Density of seeds per unit of B. gracilis or B. eriopoda basal area was obtained using the basal area of each plant measured prior to clipping. Seed density per square meter was calculated using the average basal area of all plants (in square centimeters per square meter; N = 1500) within five 3 x 4 m quadrats in each patch type in September 1996 (Peters, 2002
). Because measures of reproductive effort obtained at the community level were estimated using one value of basal area for each patch type–species combination, statistical analyses were not possible and only general comparisons can be made when comparing seed production and seed presence in the soil.
Seed presence in the soil
Soil samples were collected three times over 18 mo to assess temporal variation in seed presence. Fifteen randomly located samples were collected from each patch type in November 1995 and 1996 and April 1997. The first date represents seed presence in the soil following a drought year when very few seeds were produced. The second date represents seed presence during a year with average precipitation, and the final sampling represents seeds available for establishment after overwintering. Each sample consisted of two 7.5 cm diameter soil cores taken to a depth of 5 cm (methods follow Coffin and Lauenroth, 1989
). Microsite effects on seed presence in the soil were accounted for in two ways. One core of each sample was obtained within a plant of B. gracilis or B. eriopoda, and one was taken in an adjacent bare area to account for differences due to plant presence (Coffin and Lauenroth, 1989
); cores were combined for the analysis. The importance of species identity was accounted for by collecting samples either under B. gracilis or under B. eriopoda for the patch type codominated by these species and either under B. eriopoda and L. tridentata for the patch type codominated by these species.
Samples were sieved using a 1-cm mesh screen to remove plant material, allowed to air dry, and stored for 7 d at 0°C to break dormancy (Coffin and Lauenroth, 1989
). Each sample was then spread in a 1-cm deep layer over a potting soil–sand mixture in plastic trays in a greenhouse maintained at suitable conditions for germination of both species. Samples were watered daily and one-third strength Hoagland's solution was applied once per week. Seedlings were counted at weekly intervals for 12 wk. Identified seedlings were removed from a tray. Sieving of soil after completion of the experiment resulted in no additional seeds of these species, thus all seeds in the samples were germinable.
The total number of seedlings per square meter found in the soil was calculated for each Bouteloua species on each date and averaged across microsite (plants of B. gracilis or B. eriopoda, bare interspaces) within each patch type. This pooling was necessary because of high inter-sample variation that precluded statistical comparison of patch types (Coffin and Lauenroth, 1989
). Regression analyses were not possible to predict seed presence in the soil since samples were not collected from the same plants where the predictor variables were obtained.
Regional variation in recruitment
Published data on B. gracilis collected from the Central Plains Experimental Range (40.8° N, 107.8° W) in north-central Colorado were used to represent the shortgrass steppe biome (SGS) and to allow comparisons with data collected at the SGS/CD site. Long-term (57 yr) mean annual precipitation was 326 mm (SD = 86 mm) with >75% occurring during the April through September growing season. Annual temperatures averaged 9.0°C (SD = 1.2°C) over the same time period. Plant and soil sampling are described in detail in Coffin and Lauenroth (1989
, 1992)
. Because the SGS/CD site has not been grazed by cattle since 1973, data on B. gracilis reproductive effort at the SGS were collected from five cattle exclosures that had not been grazed for at least 5 yr (Coffin and Lauenroth, 1992
); these data were averaged for the regional comparison. Soil samples were obtained from two moderately grazed pastures on different soils (sandy loam, sandy clay loam); the numbers of seeds in the soil were averaged for the regional analysis (Coffin and Lauenroth, 1989
). Sand (74%) and clay contents (15%) were similar to grass-dominated patches at the SGS/CD site for the 0–30 cm depth where data were available.
Because seed presence in the soil (1985) and seed production data (1989) were collected in different years at the SGS, it was not possible to compare inputs and losses of seeds from the soil. Because the data were collected in different years at the SGS and SGS/CD, between-site comparisons should be treated with caution, although the sites do not necessarily experience similar weather patterns even in the same year.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
|
|
; SGS/CD: 86–90%). Most measures of reproductive potential were similar among grass-dominated patches for both species. The exceptions for B. gracilis were the larger allocation to reproductive biomass and distribution of biomass to seeds in the patch codominated by B. eriopoda (Fig. 1c and d). The exception for B. eriopoda was the heavier seeds found in the patch dominated by this species compared with the patch codominated by B. gracilis (Table 1). By contrast, most measures for B. eriopoda were significantly less in patches codominated by L. tridentata compared with grass-dominated patches. The exceptions were the biomass of inflorescences and all reproductive structures that were similar for all patches. Smallest allocation of B. eriopoda biomass to reproductive biomass in the patch codominated by L. tridentata was due to greater vegetative biomass of these plants compared to those in grass-dominated patches.
Patch type based on presence or absence of shrubs was also important to the relationships between seed production per plant and inflorescence density, aboveground biomass, or basal area for B. eriopoda (Fig. 2). All grass patches had similar regression coefficients; however, these coefficients were significantly different than coefficients for patches codominated by L. tridentata. Thus, two groups were formed for the final analysis for B. eriopoda: one group consisted of the two grass-dominated patches pooled, and the second group consisted of the B. eriopoda–L. tridentata patch only. For B. gracilis, patch type was not important to the slopes and intercepts for relationships between seed production and biomass or culm density; thus data were pooled across patch type for the final analysis.
|
Seed presence in soil
Both Bouteloua species had variable numbers of germinable seeds in the soil through time that were not predictably related to the timing of seed production and dispersal in September and October (Table 2). Because sampling was not conducted in spring 1996 at the SGS/CD site, it is not possible to distinguish the persistent component of the seed bank (Thompson and Grime, 1979
). However, the lack of B. eriopoda seeds and low numbers of B. gracilis seeds in three of four patch types (
53 seeds/m2) in November 1995 following a year-long drought suggests that both species have transient seed banks.
|
Dispersal of seeds among patch types is indicated for both species. No B. gracilis plants were found in the patch codominated by B. eriopoda and L. tridentata (Table 1), although more germinable seeds of this species than of B. eriopoda were found in the soil (Table 2). Small numbers of germinable B. eriopoda seeds were found in the B. gracilis-dominated patch type where B. eriopoda plants did not occur (Table 1). Similar numbers of germinable B. gracilis seeds were found at the SGS (0–174) and SGS/CD sites (0–548), except for the large number of seeds from November 1996 in the patch dominated by this species (1065).
Comparison of seed production and seed presence in the soil
For most patch types at the SGS/CD, higher viability by B. gracilis (Table 1) offset the lower number of seeds produced to result in a similar production of viable seeds as B. eriopoda on a square meter basis (Fig. 3a). The exception is the large number of viable seeds produced by B. eriopoda in the patch type dominated by this species that was primarily due to its high cover (Table 1). Seed production by B. gracilis was similar between patch types at the SGS/CD and higher than at the SGS.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
; Goldberg, 1985
; Breck and Jenkins, 1997
). In this study, recruitment potential differed between two perennial grass species with different growth forms, with important implications for patterns in species dominance and recovery after disturbance. In contrast to expectations, seed production per plant was greater and seed presence in the soil was less for the short-lived, stoloniferous species B. eriopoda compared with the long-lived bunchgrass B. gracilis. High seed production per plant, yet low seed viability by B. eriopoda was accompanied by few germinable seeds in the soil, whereas B. gracilis produced fewer seeds per plant with higher viability, and a greater percentage of seeds produced in the fall were found in the soil the following spring. These results suggest that recruitment processes are complex and not easily explained by energetic models or life history traits. Recent simulation model analyses of grass-dominated patches at the SGS/CD that assumed seeds were available found higher mean probabilities of B. eriopoda seedling establishment (0.14 seedlings/yr) than for B. gracilis (0.07 seedlings/yr) over the past 65 yr (Peters, 2000a
). A consideration of all three processes (seed production, presence in the soil, and establishment) suggests that recruitment by B. eriopoda may be more limited by the availability of germinable seeds in the soil, and B. gracilis may be constrained more by seedling establishment. Long-term studies that examine multiple processes are needed under a variety of climatic and soil conditions to support this hypothesis. Furthermore, the key process in any given year will likely depend upon the specific conditions of that year (Fair, Lauenroth, and Coffin, 1999
).
The importance of vegetative spread relative to seedling recruitment may also differ between the two species. Heavier seeds, larger reproductive biomass allocation per plant, and higher seed viability by B. gracilis compared with B. eriopoda suggests that more energy is allocated to seedling recruitment than vegetative spread. Although tillering rates have not been measured at the SGS/CD site, B. gracilis spreads very slowly through the production of tillers in the northern mixed-grass prairie (<2 cm/yr; Samuel, 1985
), and most recovery following disturbance in the shortgrass steppe is through seedling establishment (Coffin and Lauenroth, 1990
). By contrast, a greater energetic commitment to vegetative compared with reproductive biomass was found for B. eriopoda compared with B. gracilis. Large numbers of inflorescences (not shown) and high vegetative biomass per plant would result in high rates of tiller and stolon production since inflorescences in B. eriopoda often become rooted and function as stolons (Gadzia, 1979
). These results are supported by previous studies in the Chihuahuan Desert where B. eriopoda can spread rapidly through the production of stolons (Nelson, 1934
). Because B. eriopoda is short-lived, replacement after plant mortality through seedling establishment must also occur, even if only at low frequencies. Thus, response to disturbance by B. eriopoda at the shortgrass steppe/Chihuahuan Desert transition zone likely includes both reproductive and vegetative modes of reproduction, whereas B. gracilis most likely responds through seedling establishment.
Significant regressions between seed production and inflorescence density or aboveground biomass per plant provide easily obtained metrics to estimate total seed production for B. gracilis or B. eriopoda plants in patches dominated by these species. Between-plant variation in viability needs to be determined before regressions with viable seed production can be evaluated. Positive relationships between aboveground biomass and production of seeds have been found for other annual and perennial grasses and forbs (Mack and Harper, 1977
; Escarré and Thompson, 1991
). The lack of relationship between plant basal area and seed production is not surprising because basal area consists primarily of perennial organs (tillers) that reflect environmental conditions over a number of growing seasons (Hyder et al., 1975
).
Patch type comparisons
Most measures of reproductive potential were not affected by different types of grass patches for either species. The exception for B. gracilis was its larger allocation of biomass to reproductive structures and greater distribution of biomass to seeds in patches codominated by B. eriopoda that suggests a shift towards a more reproductive strategy compared to plants located in B. gracilis patches. For B. eriopoda, patch type was most important when comparing grass- and shrub-dominated patches. Bouteloua eriopoda plants growing in patches codominated by L. tridentata produced fewer seeds per plant with lower viability, and fewer seeds were found in the soil compared to grass-dominated patches. This shift in carbon allocation from sexual to vegetative biomass as a result of the presence of shrubs has not been observed previously for B. eriopoda, but may be an important factor contributing to its continued decline as shrubs invade (Buffington and Herbel, 1965
). A number of factors are important to shrub invasion, including drought, grazing by cattle, and lack of fire (Humphrey, 1958
; Archer, Schimel, and Holland, 1995
). Current results suggest that a negative feedback exists between the presence of shrubs and recruitment by B. eriopoda that would further reduce the ability of this grass species to persist through time. These negative effects are likely due to biotic interactions between L. tridentata and B. eriopoda (Knipe and Herbel, 1966
) because soil texture differences were small between patch types.
Because both Bouteloua species had seeds in the soil of patches where they had very little aboveground cover, seed dispersal was prevalent throughout the study area. Large numbers of B. gracilis seeds found in the soil of patches dominated by B. eriopoda and L. tridentata were surprising and supports the contention that B. gracilis is primarily constrained by seedling establishment rather than by seed availability.
Regional comparisons
The greater magnitude of response by B. gracilis for most variables at the transitional site in New Mexico compared to a site within the shortgrass steppe is in contrast to predicted responses based upon physiological constraints (Arris and Eagleson, 1989
). One explanation for these results is the single year of sampling; high interannual variation in seed production has been observed for B. gracilis at both sites. However, annual censuses of this species at the SGS (1989–1997) indicate that seed production in 1989 was above average (D. P. C. Peters and W. K. Lauenroth, unpublished data); seed production at the SGS/CD in 1996 was also above average (D. P. C. Peters, unpublished data). Another possibility is that the production and storage of B. gracilis seeds are favored by warmer temperatures found in central New Mexico compared with northern Colorado.
Summary of conclusions
In contrast to expectations, seed production per plant was greater and seed presence was lower in the short-lived stoloniferous species (B. eriopoda) compared with the long-lived bunch grass (B. gracilis) at a shortgrass steppe/Chihuahuan Desert transition zone. However, most measures of reproductive potential were larger for B. gracilis than B. eriopoda plants. These results, combined with published rates of seedling establishment, suggest that the key process limiting recruitment differs for these species. Seedling establishment is expected to constrain successful establishment by B. gracilis whereas the availability of germinable seeds in the soil may be more important for B. eriopoda. Successful recruitment is affected by a suite of biotic processes interacting with the environment of a plant. A subset of these processes was examined here. A long-term study conducted under variable environmental conditions that includes the complex of processes is needed to fully elucidate the key process or processes limiting recruitment for these species.
|
FOOTNOTES |
|---|
2 Phone: 505 646 2777; FAX: 505 646 5889; debpeter{at}nmsu.edu
. Previous name and address: Debra P. Coffin, Natural Resource Ecology Laboratory, Department of Rangeland Ecosystem Science, and Graduate Degree Program in Ecology, Colorado State University, Fort Collins, Colorado 80523 USA
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Arris L. L. P. S. Eagleson 1989 Evidence of a physiological basis for the boreal-deciduous forest ecotone in North America. Vegetatio 82: 55-58[ISI]
Breck S. W. S. H. Jenkins 1997 Use of an ecotone to test the effects of soil and desert rodents on the distribution of Indian ricegrass. Ecography 20: 253-263[CrossRef][ISI]
Brown J. R. J. C. Scanlon J. G. McIvor 1998 Competition by herbs as a limiting factor in shrub invasion in grassland: a test with different growth forms. Journal of Vegetation Science 9: 829-836
Buffington L. C. C. H. Herbel 1965 Vegetation changes on a semidesert grassland range from 1858 to 1963. Ecological Monographs 35: 139-164
Coffin D. P. W. K. Lauenroth 1988 The effects of disturbance size and frequency on a shortgrass plant community. Ecology 69: 1609-1617[CrossRef][ISI]
Coffin D. P. W. K. Lauenroth 1989 The spatial and temporal variability in the seed bank of a semiarid grassland. American Journal of Botany 76: 53-58[CrossRef][ISI]
Coffin D. P. W. K. Lauenroth 1990 A gap dynamics simulation model of succession in the shortgrass steppe. Ecological Modelling 49: 229-266
Coffin D. P. W. K. Lauenroth 1992 Spatial variability in seed production of the perennial bunchgrass Bouteloua gracilis (H.B.K.) Lag. ex Griffiths. American Journal of Botany 79: 347-353[CrossRef][ISI]
Escarré J. J. D. Thompson 1991 The effects of successional habitat variation and time of flowering on seed production in Rumex acetosella. Journal of Ecology 79: 1099-1112[CrossRef][ISI]
Fair J. F. W. K. Lauenroth D. P. Coffin 1999 Demography of Bouteloua gracilis in a mixed prairie: analysis of genets and individuals. Journal of Ecology 87: 233-243
Fields M. J. D. P. Coffin J. R. Gosz 1999 Burrowing activities of kangaroo rats and patterns in plant species dominance at a shortgrass steppe-desert grassland ecotone. Journal of Vegetation Science 10: 123-130
Gadzia K. L. 1979 Growth and development patterns of black grama (Bouteloua eriopoda [Torr.] Torr.) in southern New Mexico. M.S. thesis, New Mexico State University, Las Cruces, New Mexico, USA
Goldberg D. E. 1985 Effects of soil pH, competition, and seed predation on the distributions of two tree species. Ecology 66: 503-511[CrossRef][ISI]
Humphrey R. R. 1958 The desert grassland: a history of vegetational change and analysis of causes. Botanical Review 24: 193-252
Hyder D. N. R. E. Bement E. E. Remmenga D. F. Hervey 1975 Ecological responses of native plants and guidelines for management of shortgrass range. United States Department of Agriculture, Agricultural Research Service Technical Bulletin 1503
Kieft T. L. C. S. White S. R. Loftin R. Aguilar J. A. Craig D. A. Skaar 1998 Temporal dynamics in soil carbon and nitrogen resources at a grassland–shrubland ecotone. Ecology 79: 671-683[CrossRef][ISI]
Knipe D. C. H. Herbel 1966 Germination and growth of some semidesert grassland species treated with aqueous extract from creosotebush. Ecology 47: 775-781[CrossRef][ISI]
Kröel-Dulay G. T. Hochstrasser D. P. Coffin 1997 Compositional comparison of grass-dominated patches at a semiarid-arid grassland ecotone. Bulletin of the Ecological Society of America 78: 272
Lauenroth W. K. O. E. Sala D. P. Coffin T. B. Kirchner 1994 The importance of soil water in the recruitment of Bouteloua gracilis in the shortgrass steppe. Ecological Applications 4: 741-749
Mack R. N. J. L. Harper 1977 Interference in dune annuals: spatial pattern and neighborhood effects. Journal of Ecology 65: 345-363[CrossRef][ISI]
Minnick T. J. D. P. Coffin 1999 Geographic patterns of simulated recruitment of two Bouteloua species: implications for distributions of dominants and ecotones. Journal of Vegetation Science 10: 343-356
Neilson R. P. L. H. Wullstein 1983 Biogeography of two southwest American oaks in relation to atmospheric dynamics. Journal of Biogeography 10: 275-297
Nelson E. W. 1934 The influence of precipitation and grazing upon black grama grass range. U.S. Department of Agriculture Technical Bulletin 409
Peters D. P. C. 2000a Climatic variation and simulated patterns in seedling establishment of two dominant grasses at an ecotone. Journal of Vegetation Science 11: 493-504
Peters D. P. C. 2000b Plant species dominance and disturbance at a grassland-shrubland ecotone. Bulletin of the Ecological Society of America 81: 175
Peters D. P. C. 2002 Plant species dominance at a grassland-shrubland ecotone: an individual-based gap dynamics model of herbaceous and woody species. Ecological Modelling 152: 5-32
Samuel M. J. 1985 Growth parameter differences between populations of blue grama. Journal of Range Management 38: 339-342
Schmid B. G. M. Puttick K. H. Burgess F. A. Bazzaz 1988 Correlations between genet architecture and some life history features in three species of Solidago. Oecologia 75: 459-464[CrossRef][ISI]
Sokal R. R. F. J. Rohlf 1998 Biometry: the principles and practice of statistics in biological research. W.H. Freeman, New York, New York, USA
Sugiyama S. F. A. Bazzaz 1998 Size dependency of reproductive allocation: the influence of resource availability, competition, and genetic identity. Functional Ecology 12: 280-288
Thompson K. J. P. Grime 1979 Seasonal variation in the seed banks of herbaceous species in ten contrasting habitats. Journal of Ecology 67: 893-921[CrossRef]
Wright R. G. G. M. Van Dyne 1976 Environmental factors influencing semidesert grassland perennial grass demography. Southwestern Naturalist 21: 259-274
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
