Variation in plant‐mediated intra‐ and interspecific interactions among insect herbivores: effects of host genotype

Abstract

AbstractMany studies show that indirect interactions between insect herbivores via shared host plants are common and often mediated by plant‐induced responses to damage. However, we lack some types of data that will be needed to model how plant‐mediated interactions on individual plants contribute to the population dynamics of interacting herbivores. Specifically, there are few simultaneous characterizations of both the intra‐ and interspecific effects that are mediated by the host plant, as well as host plant constitutive resistance. Additionally, as herbivores are likely to move among plants that differ in quality, we must consider how this set of intra‐ and interspecific effects differs among plant genotypes—that is, how plant‐mediated effects genetically vary or covary. We examined how the set of intra‐ and interspecific indirect effects involving the insect folivores Leptinotarsa juncta and Manduca sexta varies across different genotypes of a shared host plant, Solanum carolinense. We damaged 12 plant genotypes using both herbivore species, then measured effects on the growth of both con‐ and heterospecifics, as well as constitutive resistance to each herbivore. We then tested for genetic variation and covariation in plant‐mediated effects and constitutive resistance among plant genotypes. We found that on average, there were significant negative intraspecific plant‐mediated effects on the growth rate of both herbivores, as well as asymmetric negative interspecific effects of M. sexta on L. juncta. Both intra‐ and interspecific effects varied across plant genotypes. For example, the interspecific effect of M. sexta on L. juncta ranged from significantly negative to significantly positive. Additionally, there were strong correlations among the individual effects mediated by S. carolinense, particularly between constitutive resistance and both intra‐ and interspecific effects. We find that these genetic correlations might limit the types and strength of interactions that take place across multiple genotypes of the same plant species. Our results suggest that future models of plant‐mediated interactions between herbivores should account for patterns of genetic variation and covariation when scaling from individual interactions to population‐level processes.