No abstract

Abstract. A longstanding paradigm in ecology is that there are positive associations between herbivore diversity, specialization, and plant species diversity, with a focus on taxonomic diversity. However, phytochemical diversity is also an informative metric, as insect herbivores interact with host plants not as taxonomic entities, but as sources of nutrients, primary metabolites, and mixtures of attractant and repellant chemicals. The present research examines herbivore responses to phytochemical diversity measured as volatile similarity in the tropical genus Piper. We quantified associations between naturally occurring volatile variation and herbivory by specialist and generalist insects. Intraspecific similarity of volatile compounds across individuals was associated with greater overall herbivory. A structural equation model supported the hypothesis that plot level volatile similarity caused greater herbivory by generalists, but not specialists, which led to increased understory plant richness. These results demonstrate that using volatiles as a functional diversity metric is informative for understanding tropical forest diversity and indicate that generalist herbivores contribute to the maintenance of diversity.

Studies of herbivores and secondary consumer communities rarely incorporate a comprehensive characterization of primary producer trait variation, thus limiting our understanding of how plants mediate community assembly of consumers. We took advantage of recent technological developments for efficient generation of phytochemical, microbial and genomic data to characterize individual alfalfa plants (Medicago sativa; Fabaceae) growing in an old‐field, semi‐naturalized state for 770 traits (including 753 chemical features). Using random forest modelling, we investigated the effect of variation in these traits on arthropod and fungal assemblages while accounting for plant genetic structure. We found that traits indicative of plant vigour, including size, percentage of flowering stems and leaf area, were positively associated with arthropod richness and abundance. Most phytochemicals were, by comparison, poor predictors, although phytochemical diversity and several individual phenolic compounds were important. Plants with a higher proportion of flowering stems were hotspots of intertrophic interactions with higher species richness of secondary consumers. The effects of many traits on plant‐associated assemblages were best modelled as nonlinear functions, often incorporating threshold effects. Foliar fungal richness was not well predicted by our models, suggesting we have much to learn regarding the role of plant traits on phyllosphere fungi at small spatial scales. Our results support the need for characterization of multiple axes of plant phenotypes in studies of plant‐arthropod‐microbe communities and demonstrate the value of modern analytical techniques for understanding the nonlinear ways in which plant traits mediate the structure of associated biotic communities. A http://onlinelibrary.wiley.com/doi/10.1111/1365-2435.13060/suppinfo is available for this article.

This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Please cite this article as

Phytochemical traits are a key component of plant defense theory. Chemical ecology has been biased towards studying effects of individual metabolites even though effective plant defenses are comprised of diverse mixtures of metabolites. We tested the phytochemical landscape hypothesis, positing that trophic interactions are contingent upon their spatial location across a phytochemically diverse landscape. Specifically, intraspecific phytochemical changes associated with vertical strata in forests were hypothesised to affect herbivore communities of the neotropical shrub Piper kelleyi Tepe (Piperaceae). Using a field experiment, we found that phytochemical diversity increased with canopy height, and higher levels of phytochemical diversity located near the canopy were characterised by tradeoffs between photoactive and non‐photoactive biosynthetic pathways. For understory plants closer to the ground, phytochemical diversity increased as direct light transmittance decreased, and these plants were characterised by up to 37% reductions in herbivory. Our results suggest that intraspecific phytochemical diversity structures herbivore communities across the landscape, affecting total herbivory.

Plant–insect interactions are ubiquitous, and have been studied intensely because of their relevance to damage and pollination in agricultural plants, and to the ecology and evolution of biodiversity. Variation within species can affect the outcome of these interactions. Specific genes and chemicals that mediate these interactions have been identified, but genome‐ or metabolome‐scale studies might be necessary to better understand the ecological and evolutionary consequences of intraspecific variation for plant–insect interactions. Here, we present such a study. Specifically, we assess the consequences of genome‐wide genetic variation in the model plant Medicago truncatula for Lycaeides melissa caterpillar growth and survival (larval performance). Using a rearing experiment and a whole‐genome SNP data set (>5 million SNPs), we found that polygenic variation in M. truncatula explains 9%–41% of the observed variation in caterpillar growth and survival. Genetic correlations among caterpillar performance and other plant traits, including structural defences and some anonymous chemical features, suggest that multiple M. truncatula alleles have pleiotropic effects on plant traits and caterpillar performance (or that substantial linkage disequilibrium exists among distinct loci affecting subsets of these traits). A moderate proportion of the genetic effect of M. truncatula alleles on L. melissa performance can be explained by the effect of these alleles on the plant traits we measured, especially leaf toughness. Taken together, our results show that intraspecific genetic variation in M. truncatula has a substantial effect on the successful development of L. melissa caterpillars (i.e., on a plant–insect interaction), and further point toward traits potentially mediating this genetic effect.

Specialized plant-insect interactions are a defining feature of life on earth, yet we are only beginning to understand the factors that set limits on host ranges in herbivorous insects. To better understand the recent adoption of alfalfa as a host plant by the Melissa blue butterfly, we quantified arthropod assemblages and plant metabolites across a wide geographic region while controlling for climate and dispersal inferred from population genomic variation. The presence of the butterfly is successfully predicted by direct and indirect effects of plant traits and interactions with other species. Results are consistent with the predictions of a theoretical model of parasite host range in which specialization is an epiphenomenon of the many barriers to be overcome rather than a consequence of trade-offs in developmental physiology.

No abstract