Predicted short-term radial-growth changes of trees based on past climate on Vancouver Island,British Columbia

Tree-ring radial expansion estimator (TREE) is an integrated radial growth model that allows users to define short-term climate change scenarios to anticipate the impact upon mature trees found growing at high elevation on Vancouver Island, British Columbia. Five individualistic models were built to represent the radial growth behaviour of mountain hemlock (Tsuga mertensiana (Bong.) Carr), yellow-cedar (Chamaecyparis nootkatensis (D. Don) Spach), western red-cedar (Thuja plicata Donn), Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco), and western hemlock (Tsuga heterophylla (Raf.) Sarg.) trees. The models were developed on climate-radial growth relationships incorporating Nanaimo climate station data, and were able to explain from 55 to 68 per cent of the variance in radial growth. The models can be run with modifications to yearly precipitation and temperature variables, giving the user the ability to investigate the radial-growth impacts of a wide range of possible climate change scenarios. Results from eight such scenarios show that species growing within their ecological limits illustrate a limited change in radial growth to forecasted climate, while species growing at an ecotonal boundary are usually very sensitive to a specific climate variables (e.g., July temperature). A forecasted alteration to this key variable will then radically alter the radial-growth rate of the species.     

Canopy Carbon Gain and Water Use: Analysis of Old-growth Conifers in the Pacific Northwest

This report summarizes our current knowledge of leaf-level physiological processes that regulate carbon gain and water loss of the dominant tree species in an old-growth forest at the Wind River Canopy Crane Research Facility. Analysis includes measurements of photosynthesis, respiration, stomatal conductance, water potential, stable carbon isotope values, and biogenic hydrocarbon emissions from Douglas-fir (Pseudotsuga menziesii), western hemlock (Tsuga heterophylla), and western red cedar (Thuja plicata). Leaf-level information is used to scale fluxes up to the canopy to estimate gross primary production using a physiology-based process model. Both light-saturated and in situ photosynthesis exhibit pronounced vertical gradients through the canopy, but are consistently highest in Douglas-fir, intermediate in western hemlock, and lowest in western red cedar. Net photosynthesis and stomatal conductance are strongly dependent on vapor-pressure deficit in Douglas-fir, and decline through the course of a seasonal drought. Foliar respiration is similar for Douglas-fir and western hemlock, and lowest for western red cedar. Water-use efficiency varied with species and tree height, as indexed using stable carbon isotopes values for foliage. Leaf water potential is most negative for Douglas-fir and similar for western hemlock and western red cedar. Terpene fluxes from foliage equal approximately 1% of the net carbon loss from the forest. Modeled estimates based on physiological measurements show gross primary productivity (GPP) to be about 22 Mg C m^–2 y^–1. Physiological studies will be necessary to further refine estimates of stand-level carbon balance and to make long-term predictions of changes in carbon balance due to changes in forest structure, species composition, and climate.     

Predicted short-term radial-growth changes of trees based on past climate on Vancouver Island, British Columbia

Tree-ring radial expansion estimator (TREE) is an integrated radial growth model that allows users to define short-term climate change scenarios to anticipate the impact upon mature trees found growing at high elevation on Vancouver Island, British Columbia. Five individualistic models were built to represent the radial growth behaviour of mountain hemlock (Tsuga mertensiana (Bong.) Carr), yellow-cedar (Chamaecyparis nootkatensis (D. Don) Spach), western red-cedar (Thuja plicata Donn), Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco), and western hemlock (Tsuga heterophylla (Raf.) Sarg.) trees. The models were developed on climate-radial growth relationships incorporating Nanaimo climate station data, and were able to explain from 55 to 68 per cent of the variance in radial growth. The models can be run with modifications to yearly precipitation and temperature variables, giving the user the ability to investigate the radial-growth impacts of a wide range of possible climate change scenarios. Results from eight such scenarios show that species growing within their ecological limits illustrate a limited change in radial growth to forecasted climate, while species growing at an ecotonal boundary are usually very sensitive to a specific climate variables (e.g., July temperature). A forecasted alteration to this key variable will then radically alter the radial-growth rate of the species.     

Effects of clipping and N fertilization on insect herbivory and infestation by pathogenic fungi on bilberry

Interactions among herbivores or between herbivores and other plant natural enemies, such as fungal pathogens, range from competition to facilitation. Moreover, the outcome of these interactions depends on the ecological context where they occur. In this study we examined the effects of clipping, as a surrogate of herbivory by ungulates, on the damage caused by two types of natural enemies (herbivorous insects and foliar fungal pathogens) on bilberry, Vaccinium myrtillus, in combination with nitrogen (N) fertilization representing current N atmospheric deposition. To examine whether the responses of both these natural enemies were mediated by changes in the plant, we estimated the effects of the treatments on bilberry growth and branching and on chlorophyll content as proxy of N content in leaves. Clipping increased the proportion of leaves damaged by herbivorous insects regardless of whether it was combined with N fertilization or not in 2008. In 2007 and 2009 repeated damage to the shrub also facilitated insect herbivory but only under N applications. Regarding fungal infestation incidence, clipping decreased the proportion of infected leaves in all the years considered but only in fertilized plots. Our results suggest that vertebrate herbivores facilitate insect herbivory and reduce fungal infestation but that these effects are dependent on nutritional conditions. Moreover, we found a negative residual correlation between insect herbivory and fungal infestation on bilberry leaves. Therefore, interactions between insect herbivores and fungal pathogens could be implicated in the final outcome of interactions between browsing ungulates and both bilberry natural enemies.     

Interactive effects of clipping and nutrient availability on the compensatory growth of a grass species

Resource availability is an important factor affecting the capacity of compensatory growth after grazing. We performed a greenhouse experiment with Poa bulbosa, a small perennial grass of the Mediterranean and Central Asian grasslands, to test the importance of nutrient availability for compensatory growth after clipping. We also compared the results with predictions of the limited resource model (LRM). Plants were grown at low and high fertilization levels and subjected to a clipping treatment. Contrary to the LMR, we found that in Poa plants compensatory growth occurred under the high fertilization level, while it did not occur under the low level. The LMR predicts a higher tolerance for grazing in the stressful environment. Our plants showed a significant decrease in their relative growth rates (RGR) after clipping. Although the plants allocated a 32–188% greater fraction of the mass to lamina growth after clipping, this greater allocation to the leaves did not fully compensate for the initial reduction in leaf area ratio (LAR). A sensitivity analysis showed for the clipped plants under the high fertilization treatment, that changes in leaf allocation (f _lam) enabled the plants to compensate for a part of the potential loss caused by defoliation. Probably, the increased biomass allocation comes largely from the bulbs. We conclude that the inconsistency of the LRM with our results originates in the lack of compensatory mechanisms in the model. To better understand how environmental conditions affect tolerance to herbivory, the effects of compensatory growth should be taken into account.     

Ecological Setting of the Wind River Old-growth Forest

The Wind River old-growth forest, in the southern Cascade Range of Washington State, is a cool (average annual temperature, 8.7°C), moist (average annual precipitation, 2223 mm), 500-year-old Douglas-fir–western hemlock forest of moderate to low productivity at 371-m elevation on a less than 10% slope. There is a seasonal snowpack (November–March), and rain-on-snow and freezing-rain events are common in winter. Local geology is characterized by volcanic rocks and deposits of Micocene/Oligocene Micocene-Oligocene (mixed) Micocene and Quaternary age, as well as intrusive rocks of Miocene age. Soils are medial, mesic, Entic Vitrands that are deep (2–3 m), well drained, loams and silt loams, generally stone free, and derived from volcanic tephra. The vegetation is transitional, between the Western Hemlock Zone and the Pacific Silver Fir Zone, and the understory is dominated by vine maple, salal, and Oregon grape. Stand structural parameters have been measured on a 4-ha plot. There are eight species of conifers, with a stand density of 427 trees ha^–1 and basal area of 82.9 m² ha^–1. Dominant conifers include Douglas-fir (35 trees ha^–1), western hemlock (224 trees ha^–1), Pacific yew (86 trees ha^–1), western red cedar (30 trees ha^–1), and Pacific silver fir (47 trees ha^–1). The average height of Douglas-fir is 52.0 m (tallest tree, 64.6 m), whereas western hemlock averages 19.0 m (tallest tree, 55.7 m). The regional disturbance regime is dominated by high-severity to moderate-severity fire, from which this forest is thought to have originated. There is no evidence that fire has occurred in the forest after establishment. Primary agents of stand disturbance, which act at the individual to small groups of trees scale, are wind, snow loads, and drought, in combination and interacting with root-rot and butt-rot fungi, heart-rot fungi, dwarf mistletoe, and bark beetles. The forest composition is slowly shifting from dominance by Douglas-fir, a shade-intolerant species, to western hemlock, western red cedar, Pacific yew, and Pacific silver fir, all shade-tolerant species. The Wind River old-growth forest fits the regional definition of Douglas-fir old growth on western hemlock sites.     

Phytochemical variation within a single plant species influences foraging behavior of deer

To determine how black‐tailed deer Odocoileus hemionus columbianus respond to phytochemical cues while browsing in heterogeneous phytochemical environments, we offered captive and free‐range deer cloned rooted cuttings and seedlings of western redcedar Thuja plicata selected for varying monoterpene content. Black‐tailed deer were thus allowed to browse among a controlled array of phytochemical cues in a series of experiments designed to evaluate foraging behavior at fine (within plot) and coarse (plot selection) scales. Within‐plot diet selection experiments demonstrated that browse preference for individual western redcedar plants was a function of foliar monoterpene concentration. Individual plant palatability combined with momentary maximization foraging strategy promoted survival of heavily defended plants. Among‐plot foraging experiments demonstrated that coarse‐scale foraging preferences were strongly influenced by distributions of high monoterpene‐containing western redcedar in available plots. Olfaction may play a significant role in both fine and coarse‐scale browse behaviors of deer as they employ a risk‐averse foraging strategy.     

Defense mechanisms of conifers : relationship of monoterpene cyclase activity to anatomical specialization and oleoresin monoterpene content

Cell-free extracts from Pinus ponderosa Lawson (ponderosa pine) and Pinus sylvestris L. (Scotch pine) wood exhibited high levels of monoterpene synthase (cyclase) activity, whereas bark extracts of these species contained no detectable activity, and they inhibited cyclase activity when added to extracts from wood, unless polyvinylpyrrolidone was included in the preparation. The molecular mass of the polyvinylpyrrolidone added was of little consequence; however, polyvinylpolypyrrolidone (a cross-linked insoluble form of the polymer) was ineffective in protecting enzyme activity. Based on these observations, methods were developed for the efficient extraction and assay of monoterpene cyclase activity from conifer stem (wood and bark) tissue. The level of monoterpene cyclase activity for a given conifer species was shown to correlate closely with the monoterpene content of the oleoresin and with the degree of anatomical complexity of the specialized resin-secreting structures. Cyclase activity and monoterpene content were lowest in the stems of species containing only isolated resin cells, such as western red cedar (Thuja plicata D. Don). Increasing levels of cyclase activity and oleoresin monoterpenes were observed in advancing from species with multicellular resin blisters (true firs [Abies]) to those with organized resin passages, such as western larch (Larix occidentalis Nutt.), Colorado blue spruce (Picea pungens Engelm.) and Douglas-fir (Pseudotsuga menziesii [Mirb.] Franco). The highest levels of cyclase activity and oleoresin monoterpenes were noted in Pinus species that contain the most highly developed resin duct systems. The relationship between biosynthetic capacity, as measured by cyclase activity, monoterpene content, and the degree of organization of the secretory structures for a given species, may reflect the total number of specialized resin-producing cells per unit mass of stem tissue.     

Simulated ungulate herbivory affects differently two herbivorous arthropod guilds in bilberry

Dense ungulate populations, both domestic and wild, accompanied by high grazing intensities, are known to have effects on the abundance and diversity of the community of arthropods hosted by plants. However, research combining ungulate herbivory with nutrient addition effects is scarce. By means of a nitrogen (N) addition and clipping experiment, we investigated the effects of both factors on the diversity, abundance and trophic structure of the arthropod community hosted by the bilberry, Vaccinium myrtillus, focusing on the response of herbivorous arthropods by considering leaf-chewers and sap-feeders separately. In addition, we assessed altitudinal variation on the arthropod community and also specifically on herbivores. Neither of the variables estimated for the arthropod community (total abundance, richness and biomass) were affected by simulated ungulate herbivory. However, clipping reduced the abundance of sap-feeders although had no significant effect on leaf-chewer insects. We therefore suggest that ungulates can compete with sap-feeders by reducing suitable food resources, or via changing plant nutritional state. Moreover, the lack of response on leaf-chewers could correspond with the absence of chemical defence in bilberry after repeated damage. Despite this effect on sap-feeder abundance, clipping did not cause a cascade effect on the trophic structure of the arthropod community associated with bilberry, at least in the short term. In turn, N fertilization at levels close to atmospheric N deposition did not cause any significant effect on the invertebrate community. Finally, the number of arthropod morphospecies and biomass decreased with altitude.     

Plants can benefit from herbivory: stimulatory effects of sheep saliva on growth of Leymus chinensis

Background

Plants and herbivores can evolve beneficial interactions. Growth factors found in animal saliva are probably key factors underlying plant compensatory responses to herbivory. However, there is still a lack of knowledge about how animal saliva interacts with herbivory intensities and how saliva can mobilize photosynthate reserves in damaged plants.

Methodology/Principal Findings

The study examined compensatory responses to herbivory and sheep saliva addition for the grass species Leymus chinensis in three experiments over three years. The first two experiments were conducted in a factorial design with clipping (four levels in 2006 and five in 2007) and two saliva treatment levels. The third experiment examined the mobilization and allocation of stored carbohydrates following clipping and saliva addition treatments. Animal saliva significantly increased tiller number, number of buds, and biomass, however, there was no effect on height. Furthermore, saliva effects were dependent on herbivory intensities, associated with meristem distribution within perennial grass. Animal saliva was found to accelerate hydrolyzation of fructans and accumulation of glucose and fructose.

Conclusions/Significance

The results demonstrated a link between saliva and the mobilization of carbohydrates following herbivory, which is an important advance in our understanding of the evolution of plant responses to herbivory. Herbivory intensity dependence of the effects of saliva stresses the significance of optimal grazing management.     

Spatial patterns and competition of tree species in a Douglas-fir chronosequence on Vancouver Island

While the successional dynamics and large-scale structure of Douglas-fir forest in the Pacific Northwest region is well studied, the fine-scale spatial characteristics at the stand level are still poorly understood. Here we investigated the fine-scale spatial structure of forest on Vancouver Island, in order to understand how the three dominant species, Douglas-fir, western hemlock, and western redcedar, coexist and partition space along a chronosequence comprised of immature, mature, and old-growth stands. We quantified the changes in spatial distribution and association of the species along the chronosequence using the scale-dependent point pattern analyses pair-correlation function g(r) and Ripley's L-function. Evidence on intra- and inter-specific competition was also inferred from correlations between nearest-neighbor distances and tree size. Our results show that 1) the aggregation of Douglas-fir in old-growth was primarily caused by variation in local site characteristics, 2) only surviving hemlock were more regular than their pre-mortality patterns, a result consistent with strong intra-specific competition, 3) inter-specific competition declined rapidly with stand age due to spatial resource partitioning, and (4) tree death was spatially randomly distributed among larger overstory trees. The study highlights the importance of spatial heterogeneity for the long-term coexistence of shade-intolerant pioneer Douglas-fir and shade-tolerant western hemlock and western redcedar.     

Phylogenetic structure of ectomycorrhizal fungal communities of western hemlock changes with forest age and stand type

On Vancouver Island, British Columbia, fertilization with nitrogen (N) and phosphorus (P) following clearcutting increases growth of western hemlock. To explore whether fertilization also resulted in ectomycorrhizal fungal communities that were more or less similar to neighboring unlogged stands, we sampled roots from western hemlock from three replicate plots from each of five different, well-characterized, forest stand types that differed in site type, and in logging and fertilization history. We harvested four samples of 100 ectomycorrhizal root tips from each plot, a total of 60 samples per stand type. From each sample, we analyzed fungal ribosomal internal transcribed spacers and 28S DNA, sequencing 15–29 clones per sample and 60–116 clones per plot. We detected 147 fungal operational taxonomic units among a total of 1435 sequences. Craterellus tubaeformis was frequently present and resulted in a pattern of phylogenetic overdispersion in the fungal communities. Fungal species composition was strongly correlated with foliar nitrogen concentration. However, other site quality factors were also important because the fertilized regenerating hemlock and mature hemlock-amabilis fir forests had similar foliar nitrogen content but little overlap in fungal species. Compared with unfertilized regenerating forests, fungal communities in N?+?P-fertilized regenerating forests had significantly more species overlap with old growth forests. However, the fungal communities of all regenerating forest were similar to one another and all differed significantly from older forests. By correlating fungal clades with habitats, this research improves understanding of how forest management can contribute to maintaining diverse ectomycorrhizal fungal communities across a landscape.     

Direct and indirect effects of allelopathy in the soil legacy of an exotic plant invasion

Invasive species may leave behind legacies that persist even after removal, inhibiting subsequent restoration efforts. We examined the soil legacy of Cytisus scoparius, a nitrogen-fixing, putatively allelopathic shrub invading the western US. We tested the hypothesis that allelopathy plays a critical role in the depressive effect of Cytisus on the key native Douglas-fir, both directly on tree growth and indirectly via effects on its ectomycorrhizal fungi (EMF). In a greenhouse factorial experiment, we used activated carbon to inhibit Cytisus-produced allelochemicals and sucrose to reduce elevated nitrogen (N). We found that: (1) Cytisus-invaded soils depressed Douglas-fir growth compared to uninvaded forest soils. The effect of adding Cytisus litter was positive (possibly reflecting an N fertilization effect) only in the presence of activated carbon, providing evidence for a role of allelopathic compounds. Activated carbon did not increase growth in the absence of Cytisus litter. Finally, sucrose addition provided weak support for a nitrogen effect of Cytisus litter. (2) Seedlings grown in Cytisus soils had lower EMF abundance compared to those in uninvaded forest soils. In forest soil from one site, adding Cytisus litter also decreased EMF abundance. Douglas-fir growth increased significantly with EMF across sites and soils suggesting that changes in EMF were linked to tree growth. The fungal taxon Cenococcum geophilum was significantly depressed in Cytisus soils compared to forest soils, while Rhizopogon rogersii abundance was similar across soil types. These results together suggest an overall negative effect of Cytisus on the growth of a dominant native tree and its fungal symbionts. Our study suggests how the role of allelopathy in ecological restoration may play out on two time scales: through immediate, direct impacts on native plants as well as through long-term, persistent impacts mediated by the collapse or transformation of microbial communities.     

Growth,reproduction and defence in nettles: responses to herbivory modified by competition and fertilization

We studied the effects of environmental factors on the ability to compensate for simulated herbivory in an annual (Urtica urens) and two perennial (U. dioica subspp. dioica and sondenii) nettle species in order to test the compensatory continuum hypothesis. Further, we studied the expression of costs of structural defence in different environmental conditions. Herbivory interacted significantly with density and fertilization for height growth, branching and reproductive traits. The negative effects of simulated herbivory on relative height growth were less detrimental for plants grown in low density and high fertilization level, which supports the compensatory continuum hypothesis. However, with respect to other traits measured, our results do not support the idea of compensatory continuum. The negative effects of apical excision on inflorescence biomass were relatively more prominent in plants growing at low density and receiving more fertilization than on those growing in worse conditions. In addition, branching was reduced by apical excision regardless of resource levels. The lack of compensation for herbivory is explained by the role which the damaged or removed tissue plays in plant development and function. In the perennial U. dioica, defensive responses to herbivory, measured as changes in trichome density, were stronger than in the annual U. urens. In the southern subspecies dioica, trichome density increased in newly emerged leaves after leaf clipping. In the northern subspecies sondenii, trichome density increased after apical excision. The differences in the defensive responses between the two subspecies of U. dioica may be due to differential natural herbivory pressures on subspecies inhabiting different geographical regions. We observed negative phenotypic correlations between structural defence and other plant traits, which suggests the existence of costs of defence. In addition to differences among the species studied, the expression of costs of defence was dependent on the resource levels.     

Indirect effects of deer on insect pests and soybean plants

1. White-tailed deer (Odocoileus virginianus Zimmermann) and insect pests negatively affect soybean production; however, little is known about how these herbivores potentially interact to affect soybean yield. Previous studies have shown deer browse on non-crop plants affects insect density and insect-mediated leaf damage, which together reduce plant reproductive output. In soybeans, reproductive output is influenced by direct and indirect interactions of different herbivores.
2. Here, we quantified indirect interactions between two groups of herbivores (mammals and insects) and their effects on soybean growth and yield. We examined responses of insect pest communities along a gradient of deer herbivory (29% to 49% browsed stems) in soybean monocultures.
3. Structural equation models showed that deer browse had direct negative effects on soybean plant height and yield. Deer browse indirectly decreased insect-mediated leaf damage by reducing plant height. Deer browse also indirectly increased pest insect abundance through reductions in plant height. Similarly, deer herbivory had an indirect positive effect on leaf carbon: nitrogen ratios through changes in plant height, thereby decreasing leaf nutrition.
4. These results suggest that pest insect abundance may be greater on soybean plants in areas of higher deer browse, but deer browse may reduce insect herbivory through reduced leaf nutrition.

     

Path analysis of natural selection on plant chemistry: the xylem resin of ponderosa pine

We analyzed the pattern of correlations among fitness components, herbivory, and resin characteristics in a natural all-aged stand of ponderosa pine, to infer the strength and mechanism of natural selection on plant chemistry. Male and female cone production were monitored yearly for 15 years, and levels of herbivory for 9 years in 165 trees. Resin flow rate and monoterpene composition were determined for these same trees. Multiple regression of fitness components on resin characteristics showed significant associations consistent with directional selection for increased resin flow rates and increased proportions of α- and β-pinene, myrcene and terpinolene. However, negative correlations among monoterpene fractions of the resin constrained the overall selection. Selective herbivory by aphids approached statistical significance and monoterpenes showed some (non-significant) effect as deterrents against deer browse. Resin characteristics were not correlated with attack by cone insects or porcupines. However, the association between resin characteristics and fitness is significantly different from that predicted by the path coefficients involving herbivores. Therefore the hypothesis that these herbivores mediate selection on the resin is not supported by the observed pattern of correlations among resin characteristics, herbivory, growth and fecundity. In this population, most of the association between resin characteristics and fitness appears to be mediated by some other factor independent of attack by herbivore species present. Received: 18 March 1996/Accepted: 18 July 1996     

Effect of herbivory on the growth and photosynthesis of replanted Calligonum caput-medusae saplings in an infertile arid desert

Replanted Calligonum caput-medusae saplings in the Tarim River watershed face short-term and frequent herbivory by goats, which can result in either growth inhibition or stimulation. The effects of herbivory on shrub saplings are unclear. We simulated herbivory with clipping to test two hypotheses. We hypothesized that (1) moderate herbivory may positively affect replanted shrub saplings due to overcompensatory growth and compensatory photosynthesis and that (2) high amounts of defoliation may change water availability and impair photosynthesis and growth of saplings. We applied four defoliation treatments (0, 30, 50, and 70 %) to 2-year-old C. caput-medusae saplings to test the effects of herbivory. Moderately defoliated (~30 %) saplings grew faster and had higher photosynthetic performance than controls; however, defoliation of 50 % or more reduced growth due to undercompensatory photosynthesis and reduced water availability. Non-photochemical quenching by thermal dissipation provided photoprotection when absorbed light energy used in PSII photochemistry was inhibited, reducing excess excitation energy and allowing saplings with high amounts of defoliation to maintain adequate photosystem functioning. This suggests that moderate herbivory of replanted shrubs used as forage in arid ecological restoration projects is feasible, but that uncontrolled grazing should be forbidden.     

Patterns of growth compensation in eastern white pine (Pinus strobus L.): the influence of herbivory intensity and competitive environments

To investigate complex growth compensation patterns, white pine (Pinus strobus L.) seedlings were clipped to simulate different herbivory levels. Seedlings were growing with different understory competition levels (created through monthly weeding vs no brush control) under a range of overstory canopy closures. Compensation patterns varied for the different growth and size measures. After one growing season, seedlings did not fully compensate for lost biomass regardless of the competitive environments of the seedlings. Although relative height growth was stimulated by light intensity clipping (20-40% of last-year shoots removed), relative diameter growth, total biomass, and biomass growth of seedlings declined sharply with increasing clipping intensity. Likewise, all growth parameters declined with increasing interspecific competition. Results showed that seedlings in highly competitive environments showed smaller growth loss due to clipping than those in competition-free environments, presumably because seedlings experiencing high interspecific competition devoted more energy to maintaining apical dominance and a balanced shoot-root ratio. While competition from canopy trees altered compensatory patterns, competition from understory vegetation only altered the magnitude, but not the patterns, of compensatory growth. We suggest that compensatory growth follows a complex pattern that will vary with the parameters measured, competitive conditions, and clipping intensities. Our results support the assertion that overcompensation may be an adaptation to competitive ability, rather than a response to herbivory itself.