Complex effects of herbivory: Indirect effects, higher-order interactions, and interaction with elevated carbon dioxide level.

Abstract

There is little research addressing interactions between herbivory and other important ecological factors and processes. The small body of work with this approach is mostly qualitative. Only a quantitative approach enables one to determine if this interaction is a simple additive one or whether there are higher-order interactions. I studied a three species system consisting of a specialist herbivore (Colorado potato beetle, Leptinotarsa decemlineata), its host plant (potato, Solanum tuberosum), and a non-host plant (soy, Glycine max). I found no higher-order interactions between herbivory and competition. These direct effects were additive, thus allowing accurate prediction of the joint outcome of herbivory and competition from the outcomes of herbivory and competition in two species subsystems. Changes in plant architecture by an herbivore would likely result in high-order interactions between herbivory and competition, but I was unable to modify architecture using plant growth regulators so this hypothesis could not be tested. The indirect facilitation of a non-host plant by an herbivore is the basis for biological control of weeds. I discuss inundative biological control of weeds using insect herbivores. Despite its limitations, I conclude that it has been unduly ignored and has biological control of weeds using insect herbivores. Despite its limitations, I conclude that it has been unduly ignored and has potential to be an important component of the assemblage of available weed control methods. I also examined how herbivory modifies plant response to CO$\sb2$ level. I measured the response of tomato (Lycopersicon esculentum) to manual defoliation (simulated herbivory) at ambient and doubled CO$\sb2$ levels. Defoliation did not affect total plant biomass but changed biomass allocation, reducing total fruit biomass while increasing aboveground vegetative biomass. However, I found no significant interaction between defoliation and response to elevated CO$\sb2.$. If these results are generally true, then experiments involving subsets of species and possible interactions can be used to understand how larger complex communities function.