Plant use patterns of a herbivorous insect: Theory, measurement and application

Within some insect populations the abundance of individuals persists at extraordinarily high densities, while other populations are consistently sparse. To understand what drives these differences ecologists have often focused on extrinsic factors such as predation and environment. My research explores how intrinsic factors, such as oviposition preference and offspring performance, are related to population abundance of the diamondback moth, Plutella xylostella (Plutellidae).; First, I examined whether preference for particular food plants is modified by the presence of alternative plant species. I measured oviposition preferences of P. xylostella using choice and no-choice assays. Moths given a choice exhibited more distinct preferences than those exposed to only a single plant. Using different preference assays can affect interpretation of preference patterns and diet breadth.; Second, I tested whether population growth and plant use traits were correlated with field density of ten P. xylostella populations. Most performance traits did not correlate with population abundance, but ovipositional diet breadth was slightly broader among populations that persist at high densities.; Third, ovipositing P. xylostella often prefer poor quality plants. Using a simple demographic model, I demonstrated that mortality during adult movement may favor the evolution of weak preferences.; Finally, I tested whether plant use traits can affect insect abundance in the field. I repeatedly sampled cabbage plots with and without candytuft (Iberis umbellata) borders. I found many predators on candytuft, but few on cabbages. Further, there were fewer butterfly (Pieris rapae) eggs on cabbages near borders, apparently because candytuft disrupted butterfly oviposition behavior. Overall, plant use traits did affect population abundance on a local scale.; Generally, plant use traits were weakly correlated with population density. Although much research has examined the role of extrinsic factors in population abundance, I found that extrinsic factors, such as the presence of additional food plants, mortality associated with movement, and the presence of vegetative borders, can also influence plant use traits. My findings indicate that to make evolutionary sense of plant use traits, ecologists must consider these traits in the context of extrinsic factors.