The effect of local resource availability and clonal integration on ramet functional morphology in Hydrocotyle bonariensis

Summary Within a physiologically integrated clone, the structure and functioning of an individual ramet is determined by: 1) the response of that ramet to its local environment and 2) its response to resource integration within the clone. In a multifactorial experiment, Hydrocotyle bonariensis ramets were grown in limiting resource environments with and without the benefit of basipetal resource movement from another branch of the clone. Ramets were analyzed for their morphological responses to variation in local light, water and nitrogen availability and to the superimposed effect of resource integration on these conditions. The expression of ramet morphology, from induction to development, was highly plastic in response to variable local resource availability. Resource integration changed a ramet's local response in a variety of ways depending on the resource(s) being translocated and the character involved. Among leaf characteristics (leaf weight, petiole height, blade area), resource translocation into the shade resulted in an enhancement of the local response. Similarly, the translocation of nitrogen and water generally increased clonal proliferation and sexual reproduction among ramets. In contrast, the translocation of water reversed the effect of local low water conditions on ramets by inhibiting root production. Some characters such as internode distance and leaf allometry were unaffected by integration. The maintenance of connections between ramets as a Hydrocotyle clone expands allows for resource sharing among widely separated ramets and can result in an integrated morpological response to a resource environment that is patchy in time and space.     

Regulation of clonal growth and dynamics of Panicum virgatum (Poaceae) in tallgrass prairie: effects of neighbor removal and nutrient addition

Clones of the perennial grass Panicum virgatum were studied on the Konza Prairie in northeast Kansas to determine the effects of neighbors, nutrient availability, and physiological integration on ramet population dynamics and clonal growth and architecture. Opposite halves of established clones in the field were subjected differentially to treatments including neighbor removal and nitrogen addition, with intact or severed rhizome connections between halves. Neighborhood competition strongly influenced clone architecture and expansion rates. Removal of neighbors resulted in a >;95% increase in radial clone expansion, intraclonal ramet densities, ramet population growth rates, ramet biomass, and percent of stems flowering, averaged over a 4-year period relative to halves or clones with intact competitors. Plant responses suggest that effects of interclonal neighbors are mediated through alteration of the light environment in the clone canopy and water availability. Addition of nitrogen did not affect lateral spread or clone structure, but resulted in significant increases in ramet size, flowering, and seed production. ANOVA revealed no significant effect of rhizome severing or treatment × severing interactions, suggesting that the size of the integrated physiological unit is much smaller than clone size and/or that physiological integration had no effect on clone responses to environmental heterogeneity at the scale of the diameter of established clones.     

CLONAL INTEGRATION ACROSS A SALT GRADIENT BY A NONHALOPHYTE,HYDROCOTYLE BONARIENSIS (APIACEAE)

This study examined the benefits associated with resource sharing among interconnected ramets spanning a soil salinity gradient. Clones of Hydrocotyle bonariensis, a rhizomatous dune perennial, expand into salt marsh communities from surrounding upland dune systems in coastal North Carolina. In rhizome-severing experiments conducted under both field and laboratory conditions, Hydrocotyle was shown to proliferate ramets under saline conditions, provided that these ramets were connected to other ramets growing in nonsaline conditions. Ramets that benefited from resource integration did not appear to be affected by local salt exposure in that these ramets were morphologically similar to those grown under nonsaline conditions. Supporting ramets incurred no net cost in terms of biomass or ramet production, but there was an increased percent allocation to roots and rhizomes. Ramets grown in saline conditions without the benefit of clonal integration showed high mortality and produced little or no net clonal growth. It is likely that the acropetal movement of water allowed Hydrocotyle clones to ameliorate the heterogeneous saline conditions associated with coastal environments.     

High levels of inter-ramet water translocation in two rhizomatous Carex species,as quantified by deuterium labelling

We studied water trnaslocation between interconnected mother and daughter ramets in two rhizomatous Carex species, using a newly developed quantitative method based on deuterium tracing. Under homogeneous conditions, in which both ramets were subjected either to wet or dry soil, little water was exchanged between the ramets. When the ramet pair was exposed to a heterogeneous water supply, water translocation became unidirectional and strongly increased to a level at which 30–60% of the water acquired by the wet ramet was exported towards the dry ramet. The quantity of water translocated was unrelated to the difference in water potential between the ramets, but highly correlated to the difference in leaf area. In both species, the transpiration of the entire plant was similar under heterogeneous and homogeneous wet conditions. This was a direct result of an increase in water uptake by the wet ramet in response to the dry conditions experienced by the interconnected ramet. In C. hirta, the costs and benefits of integration in terms of ramet biomass paralleled the responses of water consumption. This species achieved a similar whole-plant biomass in heterogeneous and homogeneous wet treatments, and water translocation was equally effective in the acropetal and basipetal directions. In C. flacca, responses of biomass and water consumption did not match and, under some conditions, water translocation imposed costs rather than benefits to the plants of this species. It is concluded that enhanced resource acquisition by donor ramets may be of critical importance for the net benefits of physiological integration in clonal plants.     

Effects of the loss of clonal integration on four sedges that differ in ramet aggregation

Although clonal growth is a dominant mode of plant growth in wetlands, the importance of clonal integration, resource sharing among ramets, to individual ramet generations (mother and daughter) and entire clones of coexisting species has not been well investigated. This study evaluated the significance of clonal integration in four sedge species of varying ramet aggregations, from clump-forming species (Clumpers –Carex sterilis, Eleocharis rostellata), with tightly aggregated ramets (rhizomes<1cm), to runner species (Runners –Schoenoplectus acutus, Cladium mariscoides), with loosely aggregated ramets. We manipulated clonal integration by either severing connections between target mother and daughter ramets or leaving connections intact, and then planted them in an intact neighborhood of a fen in Michigan, USA. We measured growth parameters of original and newly produced ramets over two growing seasons and conducted a final biomass harvest, to address four hypotheses. First, we expected integrated clones to accumulate more biomass than severed clones. However, final clone-level biomass and ramet production were the same for both treatments in all species although severing initially stimulated ramet production by Schoenoplectus and produced a more compact ramet aggregation in Cladium. Second, we hypothesized that mother ramets would experience a cost of integration, through reduced ramet or biomass production, while daughters would experience a benefit, through increased resource availability from mothers. Mother ramets of Cladium suffered a cost from integration, while Schoenoplectus mothers suffered a slight cost and Carex daughters saw a slight benefit. Finally, we hypothesized that integration would be more active in runner species than in clumper species. Indeed, we documented more active integration in runners than clumpers, but none of the study species were dependent upon integration for growth or survival once daughter ramets were established with their own roots and shoots. This study demonstrates that integration between established ramets may not be the most important advantage to clonal growth in this wetland field site. The loss of integration elicited varied responses among coexisting species in their natural habitat, somewhat but not completely related to their growth form, suggesting that a combination of plant life history traits contributes to the dependence upon clonal integration among established ramets of clonal species.     

Integration in the clonal plant <Emphasis Type="Italic">Eriophorum angustifolium</Emphasis>: an experiment with a three-member-clonal system in a patchy environment

A clonal plant in heterogeneous environments is usually expected to profit from resource exchange via a clonal network where ramets placed in contrasting environments can specialise so to acquire the most abundant resources. An experiment was designed using the three member clonal system of Eriophorum angustifolium, which consisted of one parent ramet growing in a resource poor environment and two offspring: one was limited in growth by nutrients while the other was light limited; the contrast in availability of limited resources between the offspring ramets was high, medium or none, with the system either connected or severed. The total resource availability was the same in all treatments. We proposed four possible scenarios for the system: offspring ramets will share resources via the deficient parent ramet, and the whole clone will profit from the contrasting environment (scenario 1); offspring ramets will support exclusively the parent ramet, and the whole clone will profit from a homogeneous environment (scenario 2); offspring ramets will stop the export of the limiting resource to the parent ramet, with split and connected treatments not differing (scenario 3); and offspring ramets will exhaust the carbon stored in the biomass of the parental ramet; offspring ramet will profit from connection (scenario 4). In the experiment, the limiting resources were sent to the strongest sink (scenario 2). The parent ramet growing in a deficient environment received the highest support in the treatment where both offspring ramets were growing in the same conditions (no-contrast treatment). Production of new shoots, but not biomass of whole clone, was supported in a homogenous environment. The experiment revealed that multiple stresses might prohibit free exchange of limiting resources via the clonal network and supports the idea that experimental studies on more complex clones are essential for understanding the costs and benefits of clonal growth.     

Resource sharing among ramets in the clonal herb,Fragaria chiloensis

Summary The herbaceous perennial, Fragaria chiloensis, reproduces vegetatively on coastal sand dunes in California by growth of stolons that bear rosettes. Movement of water and photosynthates through stolons integrates water and carbon metabolism of rosettes both before and after they root. New, unrooted rosettes import sufficient water and nitrogen to maintain levels near those of established rosettes; yet support of an unrooted rosette did not decrease growth of a connected, rooted sibling given abundant light, water, and soil nutrients. Under such conditions strings of unrooted rosettes with the associated stolon appeared self-sufficient for carbon; shade and drought induced import of photosynthates. New rosettes produced and maintained a limited root mass upon contact with dry sand, which could increase probability of establishment. Rooting did not induce senescence of stolons. Connection between two established rosettes prevented death by drought and shade, even when neither rosette could have survived singly. Results suggest that physiological integration of connected rosettes may increase total growth of clones of F. chiloensis through sharing of resources among ramets, especially when resource availability is changeable or patchy.     

Physiological integration among ramets of Lathyrus sylvestris L.

Summary Lathyrus sylvestris is a pioneer legume often found in disturbed habitats. Mainly reproduced through vegetative propagation, this clonal species presents a system of ramets that remain connected for several years. The existence of carbon transfer among ramets within a clone has been studied using ¹⁴C in situ. Assimilate translocation from primary to secondary ramets was observed in all clones when the primary ramet was exposed to ¹⁴CO₂. The amount of transfer ranged from trace up to 90% of the total ¹⁴C incorporated. However, in only half of the clones there was consistent enrichment of the secondary ramet (5 to 89%) suggesting that interramets transfer of carbon may be facultative. Furthermore, when significant export occurred from the primary ramet, it was always principally towards only one ramet even when the clone included more than one. The transfer of ¹⁴C from secondary to primary ramets was shown to be significant only when photosynthesis of the latter was decreased by shading. In this case import of carbon was never more than 60% of the incorporated ¹⁴C.No correlation was found between age or size of the ramets and the intensity of transfer. The shading effect let suppose that transfers are mainly driven by carbon limitation due to changing environmental conditions and not to the state of ramet maturity. The adaptative advantage of such facultative physiological integration between ramets of a clone is discussed.     

The interaction between water and nitrogen translocation in a rhizomatous sedge (Carex flacca)

In order to examine whether the translocation of water and nitrogen in clonal plants is interdependent, interramet translocation of these two resources was investigated in the greenhouse. Two-ramet systems of Carex flacca were imposed to different spatial patterns of water and nitrogen supply. The experimental design allowed to examine the effects of water heterogeneity on nitrogen sharing, and, vice versa, the effects of nitrogen heterogeneity on water sharing. Interramet translocation of both water and nitrogen was quantified by stable isotope labelling. If one of the ramets was deprived of water, nitrogen or both resources (parallel resource heterogeneity), resource translocation towards this ramet was markedly enhanced compared to a control treatment in which both ramets received ample water and ample nitrogen. Under these conditions, the amount of water or nitrogen translocated was not significantly affected by the pattern of heterogeneity of the other resource imposed on the two-ramet system. If one of the interconnected ramets was rooted in dry but nitrogen-rich soil and the other ramet was placed in nitrogen-deficient but well-watered soil (reciprocal resource heterogeneity), a significant amount of water was translocated towards the ramet in dry soil, while the low-N ramet hardly received any nitrogen. These results show that little nitrogen is translocated between ramets in a direction opposite to the transpiration stream within the rhizome. However, nitrogen may be translocated independently from water if both are transported in a similar direction within the clonal system. The effects of translocation on ramet performance (in terms of transpiration, nitrogen accumulation, and biomass) were assessed by comparing interconnected ramets with isolated (severed) ramets that were treated identically. Integration enhanced the performance of ramets deficient of one or both of the resources. In case of water translocation, the transpiration and growth of the water exporting (donor) ramets was similar to the transpiration and growth of their isolated counterparts. When nitrogen was heterogeneously supplied, however, nitrogen accumulation and growth of the donor ramet was reduced to the same extent as the performance of the nitrogen-deficient ramet was increased. Water translocation thus enhanced the performance of the whole plant, while nitrogen only reduced the differences in ramet performance within the plant. In the case of the reciprocal heterogeneity treatment, the benefits of translocation were strongly unidirectional towards the ramet in dry soil. The data for this treatment suggested that total nitrogen accumulation was enhanced by the acquisition of nitrogen from the dry pot as a result of “hydraulic lift” and water exudation in the dry soil. We conclude that nitrogen translocation in clonal plants, and the associated benefits in terms of resource utilization and growth, may strongly depend on the pattern of interramet water transport. The implications are discussed for studies of physiological integration in clonal plants and the patterns of interramet resource sharing in the field. Received: 2 November 1997 / Accepted: 9 April 1998     

Evolutionary impacts of the goldenrod ball gallmaker on Solidago altissima clones

Summary Three ramet clones of Solidago altissima were grown under greenhouse conditions to determine the effects of varying levels of attack by the goldenrod ball gallmaker (Eurosta solidaginis) on biomass allocation, leaf senescence rate and rhizome connections among ramets. The results, examined at both the individual ramet level and the clone level, showed that galled ramets became isolated from their clone through deterioration of rhizome connections. Gall effects were only observed at the ramet level although rhizome connection effects were detected at both the ramet and clone levels. The goldenrod ball gallmaker may therefore have little evolutionary impact on large clones but could appreciably affect newly established clones.     

EFFECTS OF FIRE ON CLONAL GROWTH AND DYNAMICS OF PITYOPSIS GRAMINIFOLIA (ASTERACEAE)

The rhizomatous perennial Pityopsis graminifolia was studied in a Florida sandhill community in an annually burned site, a periodically burned site, and a site that has been protected from fire since 1965. These different fire regimes significantly affected the demography and life histories of both plants and plant parts in this clonal species. Fires resulted in reductions in ramet biomass and height, and an increase in the (root + rhizome)/shoot biomass ratio. Burning also decreased the total number of flower heads and new rhizomes produced per ramet. However, the survivorship of initiated rhizomes was greater in burned sites and resulted in a larger number of established daughter ramets per clone. As a result, in burned sites there was a shift in clone structure toward larger numbers of smaller ramets, but there were no significant reductions in seed or rhizome production on a per genet basis. The results showed that the responses to fire in P. graminifolia are different when measured at the genet vs. ramet level and that the effects of fire on clones can be explained by demographic responses of plant parts. Population regeneration in the study sites was dependent on successful clonal ramet production because no seedling recruitment was observed. This suggests that disturbances other than fire are important for new genet recruitment in these clonal populations.     

An analysis of the costs and benefits of physiological integration between ramets in the clonal perennial herb Glechoma hederacea

Summary The costs and benefits, measured in terms of dry weight, of physiological integration between clonal ramets, were analysed in two experiments conducted on the clonal herb Glechoma hederacea. Firstly, integration between consecutively-produced ramets was examined in an experiment in which stolons grew from one set of growing conditions (either unshaded or shaded and either nutrient-rich or nutrient-poor) into conditions in which light or nutrient level was altered. Comparisons were made between the dry weight of the parts of the clones produced before and after growing conditions were changed, and the dry weights of the corresponding part of control clones subjected to constant growing conditions. In a second experiment, integration between two distinct parts of G. hederacea clones was investigated. In this experiment clones were grown from two connected parent ramets and the parts of the clone produced by each parent ramet were subjected independently to either nutrient-rich or nutrient-poor conditions. Ramets in resource-rich conditions provided considerable physiological support to those in resource-poor conditions. This was measured as a dry weight gain compared with the weight of the corresponding part of the control clones growing in resource-poor conditions. However, when stolons grew from resource-poor conditions into resource-rich conditions, there was no similar evidence of the resourcepoor ramtes receiving support from resource-rich ramets. Physiological integration did not result in dry weight gains when this would have necessitated basipetal translocation of resources.Because of the predominantly acropedal direction of movement of translocates in G. hederacea, the structure of the clone was important in determining the effectiveness of integration between ramets. Where physiological integration was effective, the cost to the supporting ramets in terms of dry weight was insignificant. Physiological integration allows clones to maintain a presence in less favourable sites with insignificant cost to ramets in favourable sites, thereby reducing the probability of invasion by other plants, and providing the potential for rapid clonal growth if conditions improve. Integrated support of ramets in unfavourable conditions also enables the clone to grow through unfavourable sites, thus increasing the probability of encountering more favourable conditions by wider foraging.     

Environmental heterogeneity and clonal growth: a study of the capacity for reciprocal translocation in Glechoma hederacea L.

Clonal fragments of Glechoma hederacea L. (Lamiaceae) were subjected to environments in which light and nutrients were supplied with a strictly negative association in space, i.e. when one of these resources was in ample supply the other was scarce. Treatments were chosen to simulate environments in which clones grew either within homogeneous conditions or across patch types (heterogeneous conditions). The hypothesis was tested that reciprocal translocation (i.e. exchange of both nutrients and assimilates) between connected groups of ramets would increase biomass production of clones growing under heterogeneous conditions compared to that of clones growing in homogeneous conditions. A cost-benefit analysis was carried out to test this hypothesis. Results suggested that reciprocal translocation did not occur at the structural scale considered in this experiment; no evidence was found for a significant effect on whole clone biomass of assimilate and/or nutrient translocation between clone parts experiencing contrasting levels of resource supply. It is suggested that predominantly acropetal movement of resources and the pattern of integrated physiological unit formation in G. hederacea are the main properties responsible for the lack of mutual physiological support between connected clonal fragments growing in differing habitat conditions. These properties are expected to promote clonal expansion and the exploitation of new territory, rather than sustaining clone parts in sub-optimal patches of habitat for prolonged periods of time.     

Habitat selection in spatially heterogeneous environments: a test of foraging behaviour in the clonal submerged macrophyte Vallisneria spiralis

1. To test whether clonal macrophytes can select favourable habitats in heterogeneous environments, clonal fragments of the stoloniferous submerged macrophyte Vallisneria spiralis were subjected to conditions in which light intensity and substratum nutrients were patchily distributed. The allocation of biomass accumulation and ramet production of clones to the different patches was examined. 2. The proportion of both biomass and ramet number of clones allocated to rich patches was significantly higher than in poor patches. The greatest values of both clone and leaf biomass were produced in the heterogeneous light treatment, in which clones originally grew from light‐rich to light‐poor patches, while clones produced the most offspring ramets in the treatments with heterogeneous substratum nutrients. Similarly, root biomass had the highest values in nutrient‐rich patches when clones grew from nutrient‐rich to nutrient‐poor patches. 3. The quality of patches in which parent ramets established significantly influenced the foraging pattern. When previously established in rich patches, a higher proportion of biomass was allocated to rich patches, whereas a higher proportion of ramet number was allocated to rich patches when previously established in poor patches. 4. Results demonstrate that the clonal macrophyte V. spiralis can exhibit foraging in submerged heterogeneous environments: when established under resource‐rich conditions V. spiralis remained in favourable patches, whereas if established in adverse conditions it could escape by allocating more ramets to favourable patches.     

Effects of Glomus fasciculatum and isolated rhizosphere microorganisms on growth and phosphate uptake of Plantago major ssp. pleiosperma

An experiment was set up in order to study 1) the relationship between net P uptake and dry matter production in mycorrhizal and non-mycorrhizal plants and 2) the effects of isolated rhizosphere bacteria and fungi on net P uptake and growth of P. major ssp. pleiosperma. A similar relationship between net P uptake and dry matter production was found for both mycorrhizal and non-mycorrhizal plants, although the regression lines differed in intercept.Compared to non-inoculated treatments, inoculation with bacteria slightly decreased dry matter production and P uptake of P. major, whereas inoculation with fungi or bacteria + fungi showed no effect. The results are discussed in terms of competition for available P and host photosynthates between host plant and rhizosphere microorganisms.     

Nitrogen translocation in a clonal dune perennial,Hydrocotyle bonariensis

Summary Hydrocotyle bonariensis, a common rhizomatous perennial of coastal North Carolina, forms extensive clones in dune systems characterized by a patchy nitrogen distribution. An experiment was conducted in which Hydrocotyle clones were grown across artificially created soil nitrogen gradients to determine: (1) the effect of soil nitrogen availability and nitrogen translocation on clonal structure and (2) the costs versus benefits of nitrogen translocation as measured by sexual and clonal reproduction. Acropetal translocation of nitrogen resulted in highly significant benefits to clones growing from areas of high N to areas of low N. Limited basipetal translocation was also demonstrated. Hydrocotyle ramets responded to increased nitrogen availability, from either intraclonal translocation or immediate uptake from the soil, by producing branches. Nitrogen level, however, had no effect on internode distances. Clonal integration of nitrogen, in tandem with a plastic morphology, allow Hydrocotyle clones to expand across a nitrogen-limited dune environment and to locally exploit nitrogen patches when they are encountered.     

Effects of pollen and nitrogen availability on reproduction in a woodland herb,Lysimachia quadrifolia

Summary Relationships between pollen loads, resource availability, and fruit and seed production were determined for Lysimachia quadrifolia ramets in two adjacent sites (the scrub site and the open site) in 1982, 1983, and 1984. Pollen loads limited % fruit set and seed production in the open site in 1982 and 1983. Reproduction in the scrub site was resource limited in 1982, as shown by an increase in % fruit set when one-half of the flowers on a ramet were removed prior to fruit initiation. In the scrub site in 1983, pollination of one-half of the flowers on a ramet decreased the % fruit set of the remaining, unpollinated flowers. Fruit production in the same site was limited by pollen in 1984. Addition of nitrogen fertilizer to the scrub site in 1984 had no effect on fruit and seed production. There was more variation in fruit set between sites than between years. There was no trend to greater fruit set or number of seeds/fruit on early flowers compared to late flowers on the same ramet. Correlations between measures of reproduction were positive or insignificant. These results demonstrate year-and site-specific variation in the factors that limit plant reproduction.     

Effect of population spatial structure on pollination and seed set of a clonal distylous plant,Persicaria japonica (Polygonaceae)

In order to assess the mechanisms through which the spatial structure of the population influences female reproductive success, spatial distribution of clones, degrees of limitation of legitimate (inter-morph) pollination, type and abundance of pollen loaded on the stigmas, and seed set were measured for many clones of two natural populations of the distylous clonal plant,Persicaria japonica. Within the populations, according to the spatial relation to the nearest opposite morph clone, individual clones were assorted into two spatial types,i.e., clones that congregated with clones of the opposite morph (congregating clones), and clones that occurred singly at a considerable distance from the nearest opposite-morph clone (single clones). The pollination success,i.e., the proportion of legitimately pollinated flowers, and seed set were severely limited in the single clones compared to the congregating clones. Since artificial legitimate pollination improved the seed set in single clones, at least to some degree pollination failure was responsible for the reduced seed set in the single clones.     

Effects of clonal fragmentation on intraspecific competition of a stoloniferous floating plant

Disturbance is common and can fragment clones of plants. Clonal fragmentation may affect the density and growth of ramets so that it could alter intraspecific competition. To test this hypothesis, we grew one (low density), five (medium density) or nine (high density) parent ramets of the floating invasive plant Pistia stratiotes in buckets, and newly produced offspring ramets were either severed (with fragmentation) or remained connected to parent ramets (no fragmentation). Increasing density reduced biomass of the whole clone (i.e. parent ramet plus its offspring ramets), showing intense intraspecific competition. Fragmentation decreased biomass of offspring ramets, but increased biomass of parent ramets and the whole clone, suggesting significant resource translocation from parent to offspring ramets when clones were not fragmented. There was no interaction effect of density x fragmentation on biomass of the whole clone, and fragmentation did not affect competition intensity index. We conclude that clonal fragmentation does not alter intraspecific competition between clones of P. stratiotes, but increases biomass production of the whole clone. Thus, fragmentation may contribute to its interspecific competitive ability and invasiveness, and intentional fragmentation should not be recommended as a measure to stop the rapid growth of this invasive species.     

Density- and growth stage-dependent responses to defoliation in two rhizomatous grasses

Summary Responses to defoliation were studied in two tallgrass prairie perennials (Andropogon gerardii and Panicum virgatum) established from seed at three densities. P. virgatum was also grown from transplanted rhizomes of established clones. Plants of both species displayed a continuum of responses to defoliation, from large reductions in biomass, tillering and seed production to significant increases in one or more performance measures. In crowded populations, defoliation shifted plants into subordinate positions within the competitive hierarchy. Plants competing intraspecifically and those that were initially small suffered more from defoliation than either plants grown at low density or those that were larger than their neighbors. At the highest plant density, the effects of defoliation or initial plant size were overshadowed by the effects of crowding. When defoliated and grown at similar densities, P. virgatum and A. gerardii grown from seed showed large reductions in biomass, seed production, and new rhizome production, but established P. virgatum ramets grown from rhizomes showed increases in these performance measures. Thus, herbivory may be particularly detrimental to P. virgatum during juvenile stages before perennating organs have developed. Overcompensation of P. virgatum clones in response to defoliation only occurred if all ramets within the clone were defoliated. In clones containing both defoliated and undamaged ramets, there were no differences in their performance, suggesting that genets are capable of integrating the effects of differential defoliation among shoots. Defoliated P. virgatum clones allocated a smaller fraction of their total biomass to new rhizomes, indicating that the short-term regrowth response following defoliation may incur a longer-term cost associated with gradual reduction in biomass of the perennating organs and reduced genet success.