| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ecology |
2Department of Biological Sciences, University of Jordan, Amman, 11942 Jordan; 3Division of Biology, Kansas State University, Manhattan, Kansas 66506 USA
Received for publication January 3, 2002. Accepted for publication May 16, 2002.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key Words: Ambrosia ludoviciana • Aster ericoides • compensatory growth • forbs • grasses • grazing tolerance • Sorghastrum nutans • tallgrass prairie
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
). The American bison (Bos bison) is a keystone species in tallgrass prairie, having large effects on population and community dynamics and ecosystem function (Knapp et al., 1999
). Bison graze preferentially on dominant perennial grasses, releasing subordinate species from competition, and thus increasing plant species diversity (Collins et al., 1998
). In addition, nongrazing activities of bison such as wallowing and trampling create disturbed patches where ruderal species can colonize and persist (Knapp et al., 1999
).
Fire also has a large influence on vegetation and ecosystem dynamics in tallgrass prairie. Periodic removal of detritus by fire initiates a cascade of soil and microclimate changes that increase the growth, tillering, and reproduction of many grasses (Collins and Wallace, 1990
). Frequent burning increases the relative abundance of the dominant warm-season perennial grasses while decreasing the growth and relative abundance of many forbs (Gibson and Hulbert, 1987
; Hartnett, 1991
). Both the grazing patterns of ungulates and plant responses to grazing are also altered by fire (Vinton et al., 1993
).
In grasslands, where the convergent selection pressures of grazing and frequent fire are dominant, plants tend to tolerate rather than resist the loss of aboveground tissue, i.e., rather than developing strong defenses, they develop the ability to regrow rapidly under conditions of reduced canopy and increased nutrient supplies (Westoby, 1989
). Compensatory growth and reproduction following herbivory may occur via several mechanisms, including conservation of soil water and increased water use efficiency due to reduction of transpiring leaf surface, reduced self-shading, increased rates of photosynthesis, nutrient recycling from dung and urine deposition, and/or release of apical dominance and altered growth form (e.g., McNaughton, 1979
; Briske and Derner, 1998
). However, short-term compensatory growth may be offset by significant suppression of plant growth and reproduction with repeated defoliation over longer time scales (Vinton and Hartnett, 1992
).
A plant's capacity for compensation can be strongly influenced by the timing and frequency of defoliation, the type of herbivore, the plant parts consumed, or the competitive effects of neighbors (e.g., McNaughton and Chapin, 1985
; Vinton and Hartnett, 1992
). Hilbert et al. (1981)
hypothesized that slow-growing plants would generally show greater compensation because fast-growing plants are already growing at their maximum potential relative growth rates (RGR). By contrast, Maschinski and Whitham (1989)
hypothesized that fast-growing plants under resource-rich conditions will show the greatest compensation. Empirical studies testing these hypotheses have yielded equivocal results (Chapin and McNaughton, 1989
; Georgiadis et al., 1989
). Many studies examining plant responses to herbivory have been conducted under controlled greenhouse conditions using isolated plants. Thus they may provide insight into the physiological potential for compensation, but they are unlikely to accurately predict species responses to herbivory under natural conditions.
Persistent grazing pressure can also result in changes in plant population structure, genetic differentiation, and adaptive changes in plant physiology and growth form (e.g., Polley and Detling, 1990
; Pfeiffer and Hartnett, 1995
). For example, selection may favor genotypes that maintain larger pools of stored reserves to support rapid regrowth or growth forms that facilitate compensatory growth or reduce tissue accessibility to grazers (Westoby, 1989
). A long evolutionary history of grazing has had a large impact on the physiognomy of grassland communities and on their ability to support grazing (Mack and Thompson, 1982
).
One objective of this study was to assess the relationship between defoliation tolerance, plant growth rates, and ungulate grazing preference in three tallgrass prairie species, a dominant perennial grass and two common perennial forbs. Secondly, we sought to compare interspecific patterns in compensation measured under greenhouse conditions with patterns of grazing tolerance observed in frequently burned and unburned tallgrass prairie under field conditions to test hypotheses regarding the influence of plant growth rates and resource availability on grazing tolerance. Our third objective was to examine the effects of past history of bison grazing (previously grazed or ungrazed) on current plant responses to defoliation to assess the extent and patterns of differentiation in response to differential selective pressures associated with ungulate grazing.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
In October 1987, bison were introduced onto a 486-ha area of KPBS encompassing watershed units subjected to different fire treatments. In 1992, bison were introduced onto another 463 ha, extending the area grazed by bison to 949 ha. The animals are free to move among ten watershed units (average size = 95 ha) that are subjected to prescribed burning in April at 1, 2, 4, or 20 yr. Bison preferentially select frequently burned over infrequently burned watersheds during the grazing season (Knapp et al., 1999
), creating patches of repeatedly grazed vegetation within ungrazed areas (Vinton et al., 1993
).
Three perennial plant species that are all widely distributed and abundant on tallgrass prairie but have contrasting growth patterns and differ in their selectivity by ungulates were selected. The studied species included (1) Artemisia ludoviciana Nutt. (Asteraceae), white sage, a common rhizomatous perennial that flowers late in the season. It is widespread in North America, mainly in open upland prairies. (2) Aster ericoides L. (Asteraceae), heath aster, is an abundant late-season flowering composite, which can produce large single-stemmed plants with many branches or clusters of several smaller stems from rhizomes. It is widespread in open upland prairies. (3) Sorghastrum nutans (L.) Nash (Poaceae), Indian grass, is a tall, rhizomatous, warm-season perennial grass that flowers late in the season. It is widespread in open prairies and is often the dominant species or co-dominant with Andropogon gerardii (big bluestem). These three species, although co-occurring at the local and regional scale, vary in the frequency of their consumption by grazers in tallgrass prairie. The dominant grass S. nutans is a preferred forage species that is widely and extensively grazed by bison and cattle. The forb A. ericoides is occasionally grazed in the field, while A. ludoviciana is not consumed (Catchpole, 1996
).
Experimental design and data collection
Field experiment
Two ungrazed watersheds that are burned every 4 yr were selected. The 1997 experiment was designed as a randomized complete block design for each species, with the watershed unit as the block. A total of 75 plants (one ramet or tiller per genet) of each species was randomly selected from across the entire upland areas of each watershed and assigned to a clipping treatment, and another 75 plants were randomly selected across the watershed and assigned as unclipped controls. Sampled ramets were at least 2 m apart to insure that they represented multiple genets. Ramets assigned to the defoliation treatment were clipped early in the season (May), at which time 50% of the aboveground foliage was removed. These clippings were individually dried for 72 h at 80°C, and the initial (preclipping) aboveground biomass for the clipped ramets was estimated. (Analyses of a separate subsample of harvested ramets showed that 2x dry mass of the removed clippings provided an accurate estimate of total aboveground biomass.) The initial biomass of control (unclipped) ramets was also determined. Following a 4-wk regrowth period, a second clipping was conducted, again removing 50% of aboveground foliage. This experimental clipping regimen was selected to closely simulate both the timing and intensity (clipping height) of natural bison grazing measured in a previous study of grazed watersheds on KPBS (Vinton and Hartnett, 1992
), where the warm-season perennials typically experienced their first bout of bison grazing during the May green-up period and were most often grazed again at least once later in the season. In addition, an analysis of defoliation responses in two dominant perennial grasses (Andropogon gerardii and Panicum virgatum) in this previous study (Vinton and Hartnett, 1992
) showed no significant difference in relative growth rates, final biomass, or overall performance between naturally grazed plants and plants that received simulated herbivory (clipped at the same time and intensity). Thus, we are confident that our experimental clipping regimen provides a very good simulation of patterns and intensity of natural ungulate grazing.
Seven days following the second clipping, 25 ramets each of the clipped and unclipped treatments were harvested. After 30 d a second harvest of 25 ramets of each treatment was conducted, and a final harvest of 25 each was conducted after flowering. The harvested ramets were dried for 72 h at 80°C and aboveground biomass was measured to the nearest 0.01 g. At the final harvest, belowground components of the experimental ramets were also sampled. A 10.5-cm diameter soil core from around each ramet was taken to determine the number of belowground meristems (mature rhizomes and rhizome buds) produced per ramet.
Relative growth rate determinations
At each harvest date (and pre-treatment) 25 unclipped ramets of each species were randomly harvested, dried, weighed, and ranked (smallest to largest). For each interval between harvest dates, the difference between initial and final aboveground biomass of ramets with the same ranking in the size hierarchy at the beginning and end of the period were used to calculate RGR for each plant species (e.g., the ramet in a given position in the size hierarchy at one sample date was compared to the ramet in the same position at the previous sample date to estimate change in biomass and RGR). Several earlier studies have demonstrated that the position that an individual occupies within the size hierarchy of a plant population is largely determined in the very early stages of growth, and thus the rank order of size (biomass) of ramets within a local population does not change over time (Harper, 1977
). Thus, we assumed that ramets within each treatment group did not change their position in the size hierarchy over the course of the season in the pairing of ramets for initial and final biomass determinations and RGR calculations.
Ramet RGR was calculated based on incremental increases in aboveground ramet biomass between harvests (Hunt, 1978
). At each harvest, aboveground biomass and RGR were determined. In addition, the number of branches (leaves in case of S. nutans), number of mature rhizomes, rhizome buds, flowers, and dry masses of reproductive structures were measured, and sexual reproductive effort (SRE1 and SRE2) was calculated for the final harvest. Calculations of RGR, SRE1, and SRE2 were as follows:
The experiment was repeated again in 1998. However, one of the two watershed units sampled was burned in the spring of 1998. This provided a potential opportunity to assess whether a current-year spring fire influenced plant growth and reproductive responses to subsequent grazing. However, because replication of burned and unburned watersheds was not possible, the experiment was designed in a complete randomized design for each burn treatment (burned during 1998 [B 98] or not burned during 1998 [UB 98]), and each burn treatment was analyzed separately.
Greenhouse experiment
Four watersheds that are burned every 4 yr were selected. Two watersheds had historically been grazed by cattle for several decades and then grazed by bison for the past 10 yr, while the other two watershed units had been ungrazed for at least 25 yr. Using a soil corer (10.5-cm diameter), rhizomes from the selected plant species were randomly collected in April from 60 sampling points across each of the four watersheds. Again, cores were sampled from a large area across each watershed to insure the inclusion of many genets. Rhizomes were transported to the greenhouse, where they were washed free of soil and planted into steam-pasteurized sand. When the new shoots were approximately 10 cm in height, a total of 288 ramets of each species (72 ramets per watershed x 4 watersheds) were transplanted into plastic pots (6 x 25 cm) filled with approximately 600 g (dry mass) native prairie soil collected at KPBS. The initial biomass (fresh mass) of each ramet was recorded before transplanting. In addition, 60 ramets propagated from rhizomes from each contrasting site of origin (historically grazed vs. ungrazed) were weighed (fresh mass), dried for 72 h at 80°C, and reweighed for dry mass determination. Regression analysis of these data allowed for initial dry mass estimates of experimental and control ramets.
The experiment used a split-plot design for each species, with grazing history (site of origin) as the whole plot, and the assigned clipping treatment as the subplot. In the greenhouse, the ramets of each species were randomly divided into three groups (96 plants each), to be harvested at three separate dates throughout the growing season (June–September). In addition, half of the ramets in each harvest date group (48 ramets = 12 ramets per watershed) were randomly assigned to a clipping treatment, while the other half were assigned as unclipped controls. The ramets were maintained in a 20°–25°C greenhouse for 22 wk with no supplemental light provided. They were watered every other day and fertilized once a week (25 mL/pot of 3.44 mg/mL 15-30-15 N-P-K fertilizer).
Ramets assigned to the clipping treatment were allowed to acclimate for 21 d after transplanting, at which time 50% of the aboveground tissue was removed. Following a 4-wk regrowth period, a second clipping was performed, again removing 50% of aboveground tissue. Seven days following the second clipping, one-third of the ramets of each treatment were harvested. After 30 d of regrowth an additional third was harvested, with the final third harvested after flowering. Root and shoot dry masses, number of new ramets, number of rhizome buds, RGR, number of branches, number and dry masses of flowers, and sexual reproductive allocation were determined as described above for the field experiment. Due to the influence of 24-h supplemental light from an adjacent greenhouse in the 1997 experimental setup, a delay in flowering took place in most ramets, which made it impossible to assess the effects of the main treatments (grazing history and current clipping) on their reproductive responses. For this reason, rhizomes from each species were again collected in 1998 and the experiment was repeated. However, only one harvest was conducted at the end of the 1998 season to assess reproductive responses to clipping treatments.
An overall grazing tolerance index was calculated for each of the three species in both the field and greenhouse experiments. For each species, the percent reductions in growth (total dry biomass), sexual reproduction (SRE1), and vegetative reproduction (new rhizome meristems per ramet) measured in response to clipping were averaged to determine the overall species' grazing tolerance index, expressed as a mean percent reduction in overall plant performance.
Statistical analysis
Field experiment
In 1997, each of the ramet growth and reproductive responses to clipping was analyzed in a one-way ANOVA for each species using PROC MIXED to evaluate the effects of clipping on plant responses. The 1998 data were analyzed separately for each watershed (burned or unburned in the current year) by comparing ramet responses using PROC T-TEST (SAS, 1997
). The SRE1 data were arcsine transformed prior to analysis. The results are presented as the mean values for each clipping treatment (clipped vs. unclipped).
Greenhouse experiment
For each species, the data were analyzed in a two-way ANOVA using PROC MIXED to evaluate the effects of grazing history (site of origin) and current clipping on ramet responses (SAS, 1997
). The SRE1 data were arcsine transformed prior to analysis. Due to high correlation between root, shoot, and total ramet dry mass for all plant species studied (Spearman's 
ranged from 0.68 to 0.93, P < 0.01) only the total ramet dry mass will be discussed. For each grazing history treatment, the results are presented as mean performance of clipped relative to unclipped ramets (as a percentage).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
|
|
Concurrent with its effects on shoot RGR, defoliation also significantly reduced total aboveground biomass of A. ludoviciana in both the field and greenhouse (Fig. 2). In the 1997 field experiment, the biomass of clipped ramets was less than one-half that of unclipped controls at each harvest date (Fig. 2A). In 1998, aboveground biomass of unclipped control ramets was significantly greater in the watershed that had not burned that spring, and reductions in ramet biomass due to clipping were significantly greater in the unburned as compared to the burned watershed (Fig. 2B). In the greenhouse experiment, clipping significantly reduced ramet biomass, and ramet dry masses were significantly greater than those of plants in the field (Fig. 2C). Also, in the greenhouse experiment there was no significant effect of source population (grazing history) on growth responses to defoliation. However, there was a significant interaction between population grazing history and response to clipping for some other variables. For example, in ramets grown from rhizomes collected from the ungrazed sites, clipping resulted in a significant 14% reduction in root/shoot ratio by the end of the season (P = 0.05), but no significant change in root/shoot ratio was observed in ramets originating from the grazed sites. Also, A. ludoviciana ramets originating from the grazed sites showed a 42% increase in the number of rhizome buds in response to clipping (P = 0.05), but no such response occurred in ramets from the ungrazed sites (P = NS).
Aster ericoides
The patterns of responses to defoliation in A. ericoides were generally similar to patterns in A. ludoviciana. In the field experiment, clipping resulted in negative RGR, whereas unclipped control ramets maintained constant size or positive growth throughout each season (Fig. 3A, B). Total aboveground biomass was also significantly reduced by clipping and to a similar degree in the unburned and burned watersheds (Fig. 4A, B). Ramets grown in the greenhouse showed reduced biomass in response to clipping in 1997 (Fig. 4C), but clipping had no significant effect on final biomass in 1998. In general, ramets from the two source populations showed similar growth responses when defoliated. However, ramets from the historically grazed site, when defoliated, produced significantly lower mean aboveground biomass at final harvest (1.8 g) than clipped ramets from the historically ungrazed population (2.2 g) in 1997 (P = 0.05), whereas unclipped ramets from both source populations showed identical biomass (2.7 g). This suggests lower compensatory growth capacity in plants from historically ungrazed populations. A significant interaction between grazing history (source population) and responses to clipping was also observed for sexual reproductive effort. Only ramets collected from the historically grazed sites showed a significant reduction in SRE (P 
0.05) in response to clipping. The SRE of clipped ramets was 52% lower than that of unclipped controls. In ramets from the historically ungrazed population there was no significant effect of clipping on SRE.
|
|
|
|
0.05) and reduced SRE by >50% in both years (P 0.01). However, clipping had no significant effect on rhizome production or on SRE in ramets originating from the historically grazed sites. Detailed data and summary (means and standard errors) for each of the several above- and belowground response variables (e.g., root/shoot ratio, biomass, rhizome production, SRE) can be found in Damhoureyeh (1999)
.
Interspecific patterns of grazing tolerance
The patterns of grazing tolerance of each of the three species and their overall grazing tolerance indices (estimated as the average percentage reduction in growth and reproduction in response to clipping) under greenhouse conditions are shown in Table 1. Based on the overall grazing tolerance index, A. ericoides ranks highest among the three species in grazing tolerance (compensatory capacity). It showed the smallest reductions in biomass, leaf production, number of rhizome buds, and new ramet production in response to clipping relative to the other two species. By contrast, A. ludoviciana showed the lowest overall compensatory capacity, and the greatest reductions in biomass, reproductive effort, branching, and rhizome production in response to defoliation among the three species. The overall grazing tolerance index calculated for each species showed that A. ericoides experienced a 19% reduction in overall performance in response to clipping, as compared to a 25% reduction in mean performance for S. nutans and a 40% reduction in performance for A. ludoviciana. Thus, based on responses to clipping under greenhouse conditions, A. ericoides, S. nutans, and A. ludoviciana rank first, second, and third, respectively in grazing tolerance (Table 1). In this greenhouse experiment there was no clear relationship between inherent growth rate (as measured by RGR of unclipped control ramets) and grazing tolerance or compensatory capacity. Among the three species, S. nutans showed the highest inherent growth rate but was intermediate in grazing tolerance, and the two forbs showed almost identical growth rates but differed significantly in compensatory capacity (Table 1).
|
Under natural field conditions, the interspecific patterns of grazing tolerance differed considerably (Table 2). Sorghastrum nutans showed the highest grazing tolerance, as evidenced by much smaller reductions in biomass, growth rates, and reproductive effort, and greater overall grazing tolerance index than the two forb species (Table 2). The two forb species were very similar in their grazing tolerance and significantly lower in compensatory capacity than the grass S. nutans. In this field study, the species that showed the highest inherent growth rate (S. nutans) also showed the highest grazing tolerance, and the two forbs with lower growth rates also showed lower grazing tolerance (Table 2). The dominant grass, S. nutans, showed similar growth rates in the field and in the greenhouse, but slightly lower grazing tolerance under field conditions. By contrast, the inherent growth rates of both forbs in the greenhouse were almost three times greater than their growth rates observed in the field, and their grazing tolerance indices were also much higher in the greenhouse than in the field (Tables 1, 2).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
15–25% reductions) relative to the amount of tissue lost (50% of foliage removed twice). Under field conditions, the plants exhibited less compensation and greater reductions in growth and reproduction in response to defoliation. This pattern is consistent with other studies indicating that, under natural field conditions, stresses associated with competition and limited resource availability may reduce compensatory responses and intensify the negative effects of defoliation (e.g., Harper, 1977
; Maschinski and Whitham, 1989
; Vinton and Hartnett, 1992
).
Despite the identical defoliation regimes imposed, there were clear interspecific differences in responses to simulated grazing under both field and greenhouse conditions. In the greenhouse, the forb A. ericoides showed the highest overall grazing tolerance, the grass S. nutans was intermediate, and A. ludoviciana showed the lowest tolerance. In the field experiment, however, S. nutans clearly displayed the highest grazing tolerance or regrowth capacity, followed by the two forb species, which showed very similar responses. These interspecific differences in grazing tolerance are consistent with predictions based on the species' inherent growth rates and their patterns of consumption by ungulate grazers on tallgrass prairie. In this study, the species with the highest growth rate (S. nutans) also displayed the greatest compensatory growth capacity in the field, and the slower-growing forb species displayed the lowest grazing tolerance. Furthermore, all three species showed greater compensatory growth under the resource-rich environment of the greenhouse compared to the field. These results support the hypothesis of Maschinsky and Whitham (1989)
that compensatory growth would be greatest for rapidly growing plants under resource-rich conditions. Our results do not support the alternative hypothesis of Hilbert et al. (1981)
who predicted that rapidly growing plants in resource-rich environments would show low compensatory growth because they are already growing at their physiological maximum rates. The differences in interspecific patterns of defoliation responses observed between our field and greenhouse experiments also underscore the caveat that plant responses to herbivory are context dependent, and effects of real or simulated herbivory measured under greenhouse conditions may not accurately predict responses to herbivory in natural communities.
The interspecific differences in grazing tolerance in the field also match predictions based on natural patterns of consumption of the three species in tallgrass prairie. Sorghastrum nutans, along with other dominant perennial C4 grasses, is selectively grazed by bison and cattle. Both of these ungulates show strong preference for graminoids, which comprise >80% of their diets (Catchpole, 1996
; Hartnett, Steuter, and Hickman, 1997
). Sorghastrum nutans is the most frequently consumed of the three species studied, and it showed the highest grazing tolerance. Conversely, A. ludoviciana, which shows no evidence of consumption by ungulate grazers in the field (Catchpole, 1996
), showed the lowest compensatory capacity. It is likely that these interspecific patterns represent a trade-off between defense and tolerance as alternative strategies for coping with herbivory (van der Meijden, Wijn, and Verkaar, 1988
; Westoby, 1989
; Herms and Mattson, 1992
; Rosenthal and Kotanen, 1994
). Forbs such as A. ludoviciana are avoided by ungulate herbivores, most likely due to low palatability and/or other chemical defenses, but they have low tolerance or compensatory growth capacity. By contrast, the dominant grasses, such as S. nutans, are not well defended and are readily consumed, but have high compensatory growth capacity and are able to tolerate significant levels of foliage removal.
Significant interactions between responses to defoliation (current clipping) and source population (historically grazed or ungrazed for the past 25 yr) under common greenhouse conditions for some response variables provided evidence of some genetic differentiation between populations and adaptation to selection imposed by ungulate grazers. For example, in both S. nutans and A. ludoviciana, ramets originating from the population on the historically grazed watersheds showed a significant increase in belowground rhizome bud production in response to clipping, whereas plants originating from the ungrazed watersheds did not show this response. In A. ericoides, ramets from the historically ungrazed populations showed significantly less compensatory growth than did plants from the historically grazed population, when both were grown and clipped under identical greenhouse conditions. In S. nutans, ramets from the ungrazed population showed no change in root/shoot ratios in response to clipping, whereas clipped ramets from the historically grazed watershed showed a significant decrease in root/shoot ratio toward the end of the season, perhaps reflecting their reallocation of stored belowground reserves to support aboveground compensatory growth. These differences between plants from the two sites when grown in a common greenhouse environment suggest some genetic differentiation of populations in response to a relatively short (25 yr) history of divergent selective pressures associated with grazing or no grazing. However, because the plants were propagated from rhizomes collected from the two sites, the carryover of maternal effects from the source populations may influence the response differences observed.
In conclusion, these complementary greenhouse and field experiments show that co-occurring tallgrass prairie species vary considerably in their response to grazing and in their patterns of compensatory growth following defoliation. Effects of defoliation also differ among life history stages, and sexual reproduction is reduced to a much greater extent than growth or vegetative reproduction in these prairie perennials. The differences among traits of the three species in relation to their patterns of consumption by large grazers suggest a trade-off between defense and tolerance (regrowth) as alternative strategies for coping with herbivory. Furthermore, differences in response to defoliation under field vs. greenhouse conditions clearly indicate that the capacity for compensation, and differential species responses to herbivory, are context dependent and are strongly influenced by patterns of resource availability and neighborhood competition in the field. The interspecific patterns observed also support the hypothesis that compensatory growth capacity is greater for species with high growth rate in resource-rich environments than for slow-growing species under resource-limited conditions. Finally, our results suggest the potential for relatively rapid differentiation among tallgrass prairie plant populations in response to the selective pressures associated with grazing, and they support several recent studies indicating that patterns of grazing tolerance may be both genetically based and also environmentally dependent (Strauss and Agrawal, 1999
). The ability to rapidly adapt may be expected in tallgrass prairie species due to their long history of convergent selection pressures associated with grazing by large grazers, fire, and periodic drought in these grasslands (Milchunas, Sala, and Lauenroth, 1988
).
|
FOOTNOTES |
|---|
4 Author for reprint requests (dchart{at}ksu.edu
)
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Briske D. D. J. D. Derner 1998 Clonal biology of caespitose grasses. In G. P. Cheplick [ed.], Population biology of grasses, 106–135. Cambridge University Press, Cambridge, UK
Catchpole F. B. 1996 The dynamics of bison (Bos bison) grazing patches in tallgrass prairie. Master's thesis, Kansas State University, Manhattan, Kansas, USA
Chapin F. S. S. J. McNaughton 1989 Lack of compensatory growth under phosphorus deficiency in grazing adapted grasses from the Serengeti Plains. Oecologia 79: 558-562[CrossRef][ISI]
Collins S. L. A. K. Knapp J. M. Briggs J. M. Blair E. M. Steinauer 1998 Modulation of diversity by grazing and mowing in native tallgrass prairie. Science 280: 745-747
Collins S. L. L. L. Wallace 1990 Fire in North American tallgrass prairies. University of Oklahoma Press, Norman, Oklahoma, USA
Damhoureyeh S. A. 1999 Variation in grazing tolerance among tallgrass prairie plants. Ph.D. dissertation, Kansas State University, Manhattan, Kansas, USA
Georgiadis N. J. R. W. Ruess S. J. McNaughton D. Western 1989 Ecological conditions that determine when grazing stimulates grass production. Oecologia 81: 316-322[ISI]
Gibson D. J. L. C. Hulbert 1987 Effects of fire, topography and year-to-year climatic variation on species composition in tallgrass prairie. Vegetatio 72: 175-185[ISI]
Harper J. L. 1977 Population biology of plants. Academic Press, Oxford, UK
Hartnett D. C. 1991 Effects of fire in tallgrass prairie on growth and reproduction of prairie coneflower (Ratibida columnifera: Asteraceae). American Journal of Botany 78: 429-435[CrossRef][ISI]
Hartnett D. C. A. A. Steuter K. R. Hickman 1997 Comparative ecology of native and introduced ungulates. In F. L. Knopf and F. B. Samson [eds.], Ecology and conservation of Great Plains vertebrates, 72–101. Springer-Verlag, New York, New York, USA
Herms D. A. W. J. Mattson 1992 The dilemna of plants: to grow or defend. Quarterly Review of Biology 67: 283-335
Hilbert D. W. D. M. Swift J. K. Detling M. I. Dyer 1981 Relative growth rates and the grazing optimization hypothesis. Oecologia 51: 14-18[CrossRef][ISI]
Hunt R. 1978 Plant growth analysis. Edward Arnold, London, UK
Knapp A. K. J. M. Blair J. M. Briggs S. L. Collins D. C. Hartnett L. C. Johnson E. G. Towne 1999 The keystone role of bison in North American tallgrass prairie. BioScience 49: 39-50[CrossRef][ISI]
Mack R. N. J. N. Thompson 1982 Evolution in steppe with few large hooved mammals. American Naturalist 119: 757-773[CrossRef][ISI]
Maschinski J. T. G. Whitham 1989 The continuum of plant responses to herbivory: the influence of plant association, nutrient availability, and timing. American Naturalist 134: 1-19[CrossRef][ISI]
McNaughton S. J. 1979 Grazing as an optimization process: grass–ungulate relationship in the Serengeti. American Naturalist 113: 691-703[CrossRef][ISI]
McNaughton S. J. F. S. Chapin III 1985 Effects of phosphorous nutrition and defoliation on C4 graminoids from the Serengeti plains. Ecology 66: 1617-1629[CrossRef][ISI]
Milchunas D. G. O. E. Sala W. K. Lauenroth 1988 A generalized model of the effects of grazing by large herbivores on grassland community structure. American Naturalist 132: 87-106[CrossRef][ISI]
Pfeiffer K. E. D. C. Hartnett 1995 Bison selectivity and grazing response of little bluestem in tallgrass prairie. Journal of Range Management 48: 26-31
Polley H. W. J. K. Detling 1990 Grazing-mediated differentiation in Agropyron smithii: evidence from populations with different grazing histories. Oikos 57: 326-332[CrossRef][ISI]
Rosenthal J. P. P. M. Kotanen 1994 Terrestrial plant tolerance to herbivory. Trends in Ecology and Evolution 9: 145-148
SAS. 1997 SAS/STAT software: changes and enhancements, release 6.12. SAS Institute, Cary, North Carolina, USA
Strauss S. Y. A. A. Agrawal 1999 The ecology and evolution of plant tolerance to herbivory. Trends in Ecology and Evolution 14: 179-185
van der Meijden E. M. Wijn H. J. Verkaar 1988 Defence and regrowth, alternative plant strategies in the struggle against herbivores. Oikos 51: 355-363
Vinton M. A. D. C. Hartnett 1992 Effect of bison grazing on Andropogon gerardii and Panicum virgatum in burned and unburned tallgrass prairie. Oecologia 90: 374-382[CrossRef][ISI]
Vinton M. A. D. C. Hartnett E. J. Finck J. M. Briggs 1993 Effect of fire, bison (Bison bison) grazing and plant community composition in tallgrass prairie. American Midland Naturalist 129: 10-18
Westoby M. 1989 Selective forces exerted by vertebrate herbivores on plants. Trends in Ecology and Evolution 4: 115-117
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
