Intraspecific variation in aspen phytochemistry: effects on performance of gypsy moths and forest tent caterpillars

Individual quaking aspen trees vary greatly in foliar chemistry and susceptibility to defoliation by gypsy moths and forest tent caterpillars. To relate performance of these insects to differences in foliar chemistry, we reared larvac from egg hatch to pupation on leaves from different aspen trees and analyzed leaf samples for water, nitrogen, total nonstructural carbohydrates, phenolic glycosides, and condensed tannins. Larval performance varied markedly among trees. Pupal weights of both species were strongly and inversely related to phenolic glycoside concentrations. In addition, gypsy moth performance was positively related to condensed tannin concentrations, whereas forest tent caterpillar pupal weights were positively associated with leaf nitrogen concentrations. A subsequent study with larvae fed aspen leaves supplemented with the phenolic glycoside tremulacin confirmed that the compound reduces larval performance. Larvae exhibited increased stadium durations and decreased relative growth rates and food conversion efficiencies as dietary levels of tremulacin increased. Differences in performance were more pronounced for gypsy moths than for forest tent caterpillars. These results suggest that intraspecific variation in defensive chemistry may strongly mediate interactions between aspen, gypsy moths and forest tent caterpillars in the Great Lakes region, and may account for differential defoliation of aspen by these two insect species.     

Differential toxicity of a phenolic glycoside from quaking aspen to Papilio glaucus butterfly subspecies,hybrids and backcrosses

Summary Papilio glaucus subspecies, hybrids and backcrosses exhibit greatly different abilities to use quaking aspen (Populus tremuloides) and other members of the Salicaceae as host plants. This study was conducted to test the hypotheses that phenolic glycosides account for the differences in larval performance, and that differential performance is correlated with differential larval esterase activities. To test the hypotheses we conducted first instar survival trials and fourth (penultimate) instar feeding trials with tremulacin, a phenolic glycoside. We also conducted assays of -glucosidase, esterase, and glutathione transferase activities, using midgut enzyme preparations from fifth instars. First instar survival on the tremulacin treated diet generally improved with a higher proportion of Papilio glaucus canadensis genes in the genotype, although survival in one backcross treatment was surprisingly low. Penultimate instars of P.g. glaucus and P.g. australis fed tremulacin treated black cherry leaves experienced a severe reduction in growth rate relative to larvae fed control leaves. This seriously suppressed growth was partially due to reduced consumption rates and reduced conversion efficiencies, however, approximate digestibility was not affected. In contrast, P. g. canadensis and hybrids showed no differences in growth rates between tremulacin treated and control leaves. Reciprocal backcrosses of hybrids with P. g. glaucus resulted in slightly suppressed growth on treated versus control leaves. The results suggest that after a certain threshold, increased proportions of P. g. glaucus genes resulted in poorer growth performance with tremulacin in the diet. Soluble esterase activities generally increased with the proportion of Papilio glaucus canadensis genes in the genotype, and paralleled overall trends in larval survival and feeding performance. We conclude that phenolic glycosides such as tremulacin are responsible for differential performance of Papilio glaucus subspecies, hybrids and backcrosses fed plants in the Salicaceae, and that detoxification of phenolic glycosides by midgut esterase explains why some Papilio glaucus genotypes can effectively utilize these plants.     

Foliar quality influences tree-herbivore-parasitoid interactions: effects of elevated CO<Subscript>2</Subscript>, O<Subscript>3</Subscript>, and plant genotype

This study examined the effects of carbon dioxide (CO₂)-, ozone (O₃)-, and genotype-mediated changes in quaking aspen (Populus tremuloides) chemistry on performance of the forest tent caterpillar (Malacosoma disstria) and its dipteran parasitoid (Compsilura concinnata) at the Aspen Free-Air CO₂ Enrichment (FACE) site. Parasitized and non-parasitized forest tent caterpillars were reared on two aspen genotypes under elevated levels of CO₂ and O₃, alone and in combination. Foliage was collected for determination of the chemical composition of leaves fed upon by forest tent caterpillars during the period of endoparasitoid larval development. Elevated CO₂ decreased nitrogen levels but had no effect on concentrations of carbon-based compounds. In contrast, elevated O₃ decreased nitrogen and phenolic glycoside levels, but increased concentrations of starch and condensed tannins. Foliar chemistry also differed between aspen genotypes. CO₂, O₃, genotype, and their interactions altered forest tent caterpillar performance, and differentially so between sexes. In general, enriched CO₂ had little effect on forest tent caterpillar performance under ambient O₃, but reduced performance (for insects on one aspen genotype) under elevated O₃. Conversely, elevated O₃ improved forest tent caterpillar performance under ambient, but not elevated, CO₂. Parasitoid larval survivorship decreased under elevated O₃, depending upon levels of CO₂ and aspen genotype. Additionally, larval performance and masses of mature female parasitoids differed between aspen genotypes. These results suggest that host-parasitoid interactions in forest systems may be altered by atmospheric conditions anticipated for the future, and that the degree of change may be influenced by plant genotype.     

Effect of nitrogen fertilization upon the secondary chemistry and nutritional value of quaking aspen (Populus tremuloides Michx.) leaves for the large aspen tortrix (Choristoneura conflictana (Walker))

Summary We investigated the effects of nitrogen fertilization upon the concentrations of nitrogen, condensed tannin and phenolic glycosides of young quaking aspen (Populus tremuloides) leaves and the quality of these leaves as food for larvae of the large aspen tortrix (Choristoneura conflictana), a Lepidopteran that periodically defoliates quaking aspen growing in North America. Nitrogen fertilization resulted in decreased concentrations of condensed tannin and phenolic glycosides in aspen leaves and an increase in their nitrogen concentration and value as food for the large aspen tortrix. These results indicate that plant carbon/nutrient balance influences the quality of aspen leaves as food for the large aspen tortrix in two ways, by increasing the concentrations of positive factors (e.g. nitrogen) and decreasing the concentrations of negative factors (eg. carbon-based secondary metabolites) in leaves. Addition of purified aspen leaf condensed tannin and a methanol extract of young aspen leaves that contained condensed tannin and phenolic glycosides to artificial diets at high and low levels of dietary nitrogen supported this hypothesis. Increasing dietary nitrogen increased larval growth whereas increasing the concentrations of condensed tannin and phenolic glycosides decreased growth. Additionally, the methanol extract prevented pupation. These results indicate that future studies of woody plant/insect defoliator interactions must consider plant carbon/nutrient balance as a potentially important control over the nutritional value of foliage for insect herbivores.     

Generalist predation on forest tent caterpillar varies with forest stand composition: an experimental study across multiple life stages

1. Generalist enemies can regulate low‐density forest insect populations, and are widely considered to cause greater mortality in more diverse habitats. Forest tent caterpillars (Malacosoma disstria Hübner; FTC) are a major defoliator of aspen (Populus tremuloides Micheaux) in the boreal forest, a region with a mosaic of forest stand types. This heterogeneity may influence FTC outbreaks if generalist predation or parasitism differs among stands of different tree composition. 2. Using exclusion experiments we estimate predation and parasitism of FTC across multiple life‐history stages in low‐density populations occupying both aspen (low diversity) and mixedwood stands (high diversity). 3. Arthropod and avian generalist predators were responsible for most natural enemy‐caused mortality of immature FTC, but their relative impacts varied among FTC life‐history stages. Contrary to expectation, predation on late instar larvae and pupae was higher in the less diverse aspen stands and early instar mortality did not differ. 4. By considering multiple life‐history stages, our results provide a more comprehensive view of natural enemy‐caused morality of immature FTC. Because generalist predation on FTC was higher in aspen than in mixedwood stands, we suggest that FTC populations may be slower to reach outbreak levels in aspen stands.     

Tri-trophic effects of plant defenses: chickadees consume caterpillars based on host leaf chemistry

Few studies have addressed how plant chemical defenses that directly affect herbivores in turn affect consumption patterns of vertebrates at higher trophic levels. We studied how variable foliar chemistry of trembling aspen ( Populus tremuloides Michx.) affects the diet preferences of an avian insectivore feeding on an introduced herbivore, the gypsy moth ( Lymantria dispar L.).
Black-capped chickadees ( Poecile atricapilla ) were offered paired choices of gypsy moth caterpillars feeding on one of three genotypes of aspen that differed in chemical composition. Chickadees chose to eat caterpillars fed aspen foliage with low levels of both condensed tannins and phenolic glycosides, or caterpillars fed foliage with high levels of tannins and low levels of phenolic glycosides, over caterpillars fed foliage with low levels of condensed tannins and high levels of phenolic glycosides. In addition, diet choices of the birds were affected by their previous experience. These findings are consistent with the "extended phenotype" concept, in that genetically-based chemical traits in an ecologically dominant plant influence the feeding behavior of third trophic level organisms, whose efficacy as regulators of herbivore populations may in turn be modified.     

Social Cues and Following Behavior in the Forest Tent Caterpillar

Colonies of the social caterpillar Malacosoma disstria Hubner (Lepidoptera: Lasiocampidae) travel in groups following silk trails marked with pher-omone. This study examined first, the cues involved in following behavior and second, the responses to these cues at different larval stadia. Both second and fourth instar larvae discriminated between fresh and older trails, and travelled faster in the presence of trails. In addition to trail following, young caterpillars exhibited leader following, which might be particularly important in exploring unmarked territory. Indeed, second instar caterpillars were more likely to travel together when trails were absent. Fourth instar larvae exhibited greater independent locomotion in the absence of trails than did younger larvae. These findings help explain patterns of social behavior observed in forest tent caterpillar colonies in the field.     

Clonal variation in foliar chemistry of aspen: effects on gypsy moths and forest tent caterpillars

Quaking aspen (Populus tremuloides) exhibits striking intraspecific variation in concentrations of phenolic glycosides, compounds that play important roles in mediating interactions with herbivorous insects. This research was conducted to assess the contribution of genetic variation to overall phenotypic variation in aspen chemistry and interactions with gypsy moths (Lymantria dispar) and forest tent caterpillars (Malacosoma disstria). Thirteen aspen clones were propagated from field-collected root material. Insect performance assays, measuring survival, development, growth, and food utilization indices, were conducted with second and/or fourth instars. Leaf samples were assayed for water, nitrogen, total nonstructural carbohydrates, condensed tannins, and phenolic glycosides. Results showed substantial among-clone variation in the performance of both insect species. Chemical analyses revealed significant among-clone variation in all foliar constituents and that variation in allelochemical contents differed more than variation in primary metabolites. Regression analyses indicated that phenolic glycosides were the dominant factor responsible for among-clone variation in insect performance. We also found significant genetic trade-offs between growth and defense among aspen clones. Our results suggest that genetic factors are likely responsible for much of the tremendous phenotypic variation in secondary chemistry exhibited by aspen, and that the genetic structure of aspen populations may play important roles in the evolution of interactions with phytophagous insects. Received: 14 May 1996 / Accepted: 29 January 1997     

Enriched atmospheric CO2 and defoliation: effects on tree chemistry and insect performance

We examined the effects of CO₂ and defoliation on tree chemistry and performance of the forest tent caterpillar, Malacosoma disstria. Quaking aspen (Populus tremuloides) and sugar maple (Acer saccharum) trees were grown in open-top chambers under ambient or elevated concentrations of CO₂. During the second year of growth, half of the trees were exposed to free-feeding forest tent caterpillars, while the remaining trees served as nondefoliated controls. Foliage was collected weekly for phytochemical analysis. Insect performance was evaluated on foliage from each of the treatments. At the sampling date coincident with insect bioassays, levels of foliar nitrogen and starch were lower and higher, respectively, in high CO₂ foliage, and this trend persisted throughout the study. CO₂-mediated increases in secondary compounds were observed for condensed tannins in aspen and gallotannins in maple. Defoliation reduced levels of water and nitrogen in aspen but had no effect on primary metabolites in maple. Similarly, defoliation induced accumulations of secondary compounds in aspen but not in maple. Larvae fed foliage from the enriched CO₂ or defoliated treatments exhibited reduced growth and food processing efficiencies, relative to larvae on ambient CO₂ or nondefoliated diets, but the patterns were host species-specific. Overall, CO₂ and defoliation appeared to exert independent effects on foliar chemistry and forest tent caterpillar performance.     

Effects of variable phytochemistry and budbreak phenology on defoliation of aspen during a forest tent caterpillar outbreak

1 The present study assessed the relationship between clonally variable rates of defoliation in trembling aspen (Populus tremuloides Michx.) and two potential resistance traits: defensive chemistry and leaf phenology. 2 In 2001, coincident with a major outbreak of the forest tent caterpillar (Malacosoma disstria Hubner) in the northcentral U.S.A., we monitored defoliation rates, phytochemical composition, and foliar development in 30 clones of trembling aspen. Leaf chemistry was also assessed in re‐flushed leaves and 2 years post‐outbreak. 3 Early in the season, differences in defoliation among clones were substantial but, by mid‐June, all clones were completely defoliated. Leaf nitrogen, condensed tannins, and phenolic glycosides varied among clones but did not relate to defoliation levels. Budbreak phenology differed by 3 weeks among clones and clones that broke bud early or late relative to forest tent caterpillar eclosion experienced reduced rates of defoliation. 4 Defoliation led to increased tannins and slight decreases in phenolic glycoside concentrations in damaged leaf remnants, but to moderately decreased tannins and a six‐fold increase in phenolic glycosides in reflushed leaves. This shift in chemical composition may significantly affect late season herbivores. 5 These results suggest that aspen chemical resistance mechanisms are ineffective during intense episodic eruptions of outbreak folivores such as the forest tent caterpillar. Variable budbreak phenology may lead to differential susceptibility during less intense outbreak years and, at peak forest tent caterpillar population densities, mechanisms affording tolerance are probably more important than chemical defences.     

The effect of food limitation on immunity factors and disease resistance in the western tent caterpillar

Epizootics of nucleopolyhedrovirus characterize declines of cyclic populations of western tent caterpillars, Malacosoma pluviale californicum. In field populations, infection can be apparently lacking in one generation and high in the next. This may suggest an increase in the susceptibility to infection of larvae at peak density or the triggering of a vertically transmitted virus. Here, we test the hypothesis that reduced food availability, as may occur during population outbreaks of tent caterpillars, influences the immunocompetence of larvae and increases their susceptibility to viral infection. We compared immunity factors, hemolymph phenoloxidase and hemocyte numbers, and the susceptibility to nucleopolyhedroviral infection of fifth instar larvae that were fully or partially fed as fourth instars. To determine if maternal or transgenerational influences occurred, we also determined the susceptibility of the offspring of the treated parents to viral infection. Food limitation significantly reduced larval survival, development rate, larval and pupal mass, moth fecundity and levels of hemolymph phenoloxidase, but not the numbers of hemocytes. Neither the food-reduced larvae nor their offspring were more susceptible to viral infection and were possibly even less susceptible at intermediate viral doses. Food reduction did not activate latent or covert viral infection of larvae as might be expected as a response to stress. We conclude that reducing the food intake of fourth instar larvae to an extent that had measurable and realistic impacts on their life history characteristics was not translated into increased susceptibility to viral infection.     

Effects of elevated carbon dioxide and ozone on the phytochemistry of aspen and performance of an herbivore

The purpose of this study was to assess the independent and interactive effects of CO(2), O(3), and plant genotype on the foliar quality of a deciduous tree and the performance of a herbivorous insect. Two trembling aspen (Populus tremuloides Michaux) genotypes differing in response to CO(2) and O(3) were grown at the Aspen FACE (Free Air CO(2) Enrichment) site located in northern Wisconsin, USA. Trees were exposed to one of four atmospheric treatments: ambient air (control), elevated carbon dioxide (+CO(2); 560 microl/l), elevated ozone (+O(3); ambient x1.5), and elevated CO(2)+O(3). We measured the effects of CO(2) and O(3) on aspen phytochemistry and on performance of forest tent caterpillar (Malacosoma disstria Hübner) larvae. CO(2) and O(3) treatments influenced foliar quality for both genotypes, with the most notable effects being that elevated CO(2) reduced nitrogen and increased tremulacin levels, whereas elevated O(3) increased early season nitrogen and reduced tremulacin levels, relative to controls. With respect to insects, the +CO(2) treatment had little or no effect on larval performance. Larval performance improved in the +O(3) treatment, but this response was negated by the addition of elevated CO(2) (i.e., +CO(2)+O(3) treatment). We conclude that tent caterpillars will have the greatest impact on aspen under current CO(2) and high O(3) levels, due to increases in insect performance and decreases in tree growth, whereas tent caterpillars will have the least impact on aspen under high CO(2) and low O(3) levels, due to moderate changes in insect performance and increases in tree growth.     

The influence of phenolic compounds on the suitability of three Salix species as hosts for the willow beetle Phratora vulgatissima

The phenolic glycosides salicin and salicortin were found to influence larval growth and development rates and adult feeding preference of Phratora vulgatissima in laboratory feeding studies. Salicortin was more toxic to larvae than salicin, and none of the third instar larvae fed on Salix viminalis leaves amended with 1.52% (fresh mass) salicortin pupated. Condensed tannins (proanthocyanadins) did not affect larval performance. It was concluded that Salix burjatica resistance to willow beetle is due to the high levels of salicortin which occur in leaves of this species.     

Increased mortality of gypsy moth Lymantria dispar (L.) (Lepidoptera: Lymantriidae) exposed to gypsy moth nuclear polyhedrosis virus in combination with the phenolic gycoside salicin

Second instar gypsy moth, Lymantria dispar (L.) (Lepidoptera: Lymantriidae), larvae suffered significantly greater mortality from aerially applied gypsy moth nuclear polyhedrosis virus (Gypchek) when the virus was consumed on quaking aspen, Populus tremuloides Michx., versus red oak, Quercus spp. L., foliage. Laboratory assays in which various doses of Gypchek and salicin (a phenolic glycoside present in aspen foliage) were tested in combination demonstrated that salicin significantly increased total larval mortality and lowered the LD50 estimates (dose of Gypchek that resulted in 50% population mortality) for the virus, although not significantly. While salicin did not impact larval survival in the absence of Gypcek, it did act to significantly deter feeding when it was present in high concentrations (up to 5.0%) within the treatment formulations. The enhanced activity of Gypchek in the presence of salicin is similar to prior reports of enhanced activity of the bacterial pathogen Bacillus thuringiensis when consumed concurrently with phenolic glycosides commonly present in aspen foliage. The enhancement of viral activity is in contrast to the inhibitory effects on the virus reported for another common group of phenolic compounds, tannins.     

Nuclear polyhedrosis virus as a biological control agent for Malacosoma americanum (Lepidoptera: Lasiocampidae)

In addition to damaging trees, the eastern tent caterpillar is implicated in early fetal loss and late‐term abortion in horses. In a field study, we evaluated the potential biological control of the caterpillar using eastern tent caterpillar nuclear polyhedrosis virus (ETNPV), a naturally occurring virus that is nearly species‐specific. Egg masses were hatched and second instar larvae were fed virus‐inoculated foliage to propagate the virus in vivo. Then, a viral pesticide was formulated at concentrations of 10⁴, 10⁶ and 10⁸ polyhedral inclusion bodies per ml. The pesticide was applied to foliage on which second, third and fourth instar caterpillars were feeding. When the majority of surviving larvae reached the sixth instar, colonies were collected and the surviving caterpillars counted. Mean numbers of surviving caterpillars per treatment were compared via 95% bootstrap confidence intervals. The data indicate second instar caterpillars were highly susceptible to the virus, but only at the highest concentration tested. Third instar caterpillars were also somewhat susceptible to high virus concentrations, while fourth instar caterpillars were fairly resistant. Our data provide the strongest evidence to date that ETNPV can be propagated, harvested and refined for formulation as a biological control agent for eastern tent caterpillar. Its use on this insect may be merited in circumstances where landowners and managers need to protect trees and horses.     

Effects of host switching on gypsy moth (Lymantria dispar (L.)) under field conditions

Effects of various single and two species diets on the performance of gypsy moth (Lymantria dispar (L.)) were studied when this insect was reared from hatch to population on intact host trees in the field. The tree species used for this study were red oak (Quercus rubra L.), white oak (Q. alba L.), bigtooth aspen (Populus grandidentata Michaux), and trembling aspen (P. tremuloides Michaux). These are commonly available host trees in the Lake States region. The study spanned two years and was performed at two different field sites in central Michigan. Conclusions drawn from this study include: (1) Large differences in gypsy moth growth and survival can occur even among diet sequences composed of favorable host species. (2) Larvae that spent their first two weeks feeding on red oak performed better during this time period than larvae on all other host species in terms of mean weight, mean relative growth rate (RGR), and mean level of larval development, while larvae on a first host of bigtooth aspen were ranked lowest in terms of mean weight, RGR, and level of larval development. (3) Combination diets do not seem to be inherently better or worse than diets composed of only a single species; rather, insect performance was affected by the types of host species eaten and the time during larval development that these host species were consumed instead of whether larvae ate single species diets or mixed species diets. (4) In diets composed of two host species, measures of gypsy moth performance are affected to different extents in the latter part of the season by the two different hosts; larval weights and development rates show continued effects of the first host fed upon while RGRs, mortality, and pupal weights are affected strongly by the second host type eaten. (5) Of the diets investigated in this study, early feeding on red oak followed by later feeding on an aspen, particularly trembling aspen, is most beneficial to insects in terms of attaining high levels of performance throughout their lives.     

Leaf age effects of elevated CO2-grown white oak leaves on spring-feeding lepidopterans

Folivorous insect responses to elevated CO₂-grown tree species may be complicated by phytochemical changes as leaves age. For example, young expanding leaves in tree species may be less affected by enriched CO₂-alterations in leaf phytochemistry than older mature leaves due to shorter exposure times to elevated CO₂ atmospheres. This, in turn, could result in different effects on early vs. late instar larvae of herbivorous insects. To address this, seedlings of white oak (Quercus alba L.), grown in open-top chambers under ambient and elevated CO₂, were fed to two important early spring feeding herbivores; gypsy moth (Lymantria dispar L.), and forest tent caterpillar (Malacosoma disstria Hübner). Young, expanding leaves were presented to early instar larvae, and older fully expanded or mature leaves to late instar larvae. Young leaves had significantly lower leaf nitrogen content and significantly higher total nonstructural carbohydrate:nitrogen ratio as plant CO₂ concentration rose, while nonstructural carbohydrates and total carbon-based phenolics were unaffected by plant CO₂ treatment. These phytochemical changes contributed to a significant reduction in the growth rate of early instar gypsy moth larvae, while growth rates of forest tent caterpillar were unaffected. The differences in insect responses were attributed to an increase in the nitrogen utilization efficiency (NUE) of early instar forest tent caterpillar larvae feeding on elevated CO₂-grown leaves, while early instar gypsy moth larval NUE remained unchanged among the treatments. Later instar larvae of both insect species experienced larger reductions in foliage quality on elevated CO₂-grown leaves than earlier instars, as the carbohydrate:nitrogen ratio of leaves substantially increased. Despite this, neither insect species exhibited changes in growth or consumption rates between CO₂ treatments in the later instar. An increase in NUE was apparently responsible for offsetting reduced foliar nitrogen for the late instar larvae of both species.     

Genotype and environment determine allocation to and costs of resistance in quaking aspen

Although genetic variability and resource availability both influence plant chemical composition, little is known about how these factors interact to modulate costs of resistance, expressed as negative correlations between growth and defense. We evaluated genotype × environment effects on foliar chemistry and growth of quaking aspen (Populus tremuloides) by growing multiple aspen genotypes under variable conditions of light and soil nutrient availability in a common garden. Foliage was analyzed for levels of nitrogen, phenolic glycosides and condensed tannins. Bioassays of leaf quality were conducted with fourth-stadium gypsy moth (Lymantria dispar) larvae. Results revealed strong effects of plant genotype, light availability and nutrient availability; the importance of each factor depended upon compound type. For example, tannin concentrations differed little among genotypes and across nutrient regimes under low light conditions, but markedly so under high light conditions. Phenolic glycoside concentrations, in contrast, were largely determined by genotype. Variation in phenolic glycoside concentrations among genotypes was the most important factor affecting gypsy moth performance. Gypsy moth biomass and development time were negatively and positively correlated, respectively, with phenolic glycoside levels. Allocation to phenolic glycosides appeared to be costly in terms of growth, but only under resource-limiting conditions. Context-dependent trade-offs help to explain why costs of allocation to resistance are often difficult to demonstrate.     

Plant-associated bacteria degrade defense chemicals and reduce their adverse effects on an insect defoliator

Phytophagous insects must contend with numerous secondary defense compounds that can adversely affect their growth and development. The gypsy moth (Lymantria dispar) is a polyphagous herbivore that encounters an extensive range of hosts and chemicals. We used this folivore and a primary component of aspen chemical defenses, namely, phenolic glycosides, to investigate if bacteria detoxify phytochemicals and benefit larvae. We conducted insect bioassays using bacteria enriched from environmental samples, analyses of the microbial community in the midguts of bioassay larvae, and in vitro phenolic glycoside metabolism assays. Inoculation with bacteria enhanced larval growth in the presence, but not absence, of phenolic glycosides in the artificial diet. This effect of bacteria on growth was observed only in larvae administered bacteria from aspen foliage. The resulting midgut community composition varied among the bacterial treatments. When phenolic glycosides were included in diet, the composition of midguts in larvae fed aspen bacteria was significantly altered. Phenolic glycosides increased population responses by bacteria that we found able to metabolize these compounds in liquid growth cultures. Several aspects of these results suggest that vectoring or pairwise symbiosis models are inadequate for understanding microbial mediation of plant–herbivore interactions in some systems. First, bacteria that most benefitted larvae were initially foliar residents, suggesting that toxin-degrading abilities of phyllosphere inhabitants indirectly benefit herbivores upon ingestion. Second, assays with single bacteria did not confer the benefits to larvae obtained with consortia, suggesting multi- and inter-microbial interactions are also involved. Our results show that bacteria mediate insect interactions with plant defenses but that these interactions are community specific and highly complex.     

Effects of CO2 and light on tree phytochemistry and insect performance

Direct and interactive effects of CO₂ and light on tree phytochemistry and insect fitness parameters were examined through experimental manipulations of plant growth conditions and performance of insect bioassays. Three species of deciduous trees (quaking aspen, Populus tremuloides; paper birch, Betula papyrifera; sugar maple, Acer saccharum) were grown under ambient (387±8 μL/L) and elevated (696±2 μL/L) levels of atmospheric CO₂, with low and high light availability (375 and 855 μmol×m^?2×s^?1 at solar noon). Effects on the population and individual performance of a generalist phytophagous insect, the white‐marked tussock moth (Orgyia leucostigma) were evaluated. Caterpillars were reared on experimental trees for the duration of the larval stage, and complementary short‐term (fourth instar) feeding trials were conducted with insects fed detached leaves.
Phytochemical analyses demonstrated strong effects of both CO₂ and light on all foliar nutritional variables (water, starch and nitrogen). For all species, enriched CO₂ decreased water content and increased starch content, especially under high light conditions. High CO₂ availability reduced levels of foliar nitrogen, but effects were species specific and most pronounced for high light aspen and birch. Analyses of secondary plant compounds revealed that levels of phenolic glycosides (salicortin and tremulacin) in aspen and condensed tannins in birch and maple were positively influenced by levels of both CO₂ and light. In contrast, levels of condensed tannins in aspen were primarily affected by light, whereas levels of ellagitannins and gallotannins in maple responded to light and CO₂, respectively.
The long‐term bioassays showed strong treatment effects on survival, development time, and pupal mass. In general, CO₂ effects were pronounced in high light and decreased along the gradient aspen birch maple. For larvae reared on high light aspen, enriched CO₂ resulted in 62% fewer survivors, with increased development time, and reduced pupal mass. For maple‐fed insects, elevated CO₂ levels had negative effects on survival and pupal mass in low light. For birch, the only negative CO₂ effects were observed in high light, where female larvae showed prolonged development. Fourth instar feeding trials demonstrated that low food conversion efficiency reduced insect performance. Elevated levels of CO₂ significantly reduced total consumption, especially by insects on high light aspen and low light maple.
This research demonstrates that effects of CO₂ on phytochemistry and insect performance can be strongly light‐dependent, and that plant responses to these two environmental variables differ among species. Overall, increased CO₂ availability appeared to increase the defensive capacity of early‐successional species primarily under high light conditions, and of late‐successional species under low light conditions. Due to the interactive effects of tree species, light, CO₂, and herbivory, community composition of forests may change in the future.