Ecological and behavioral aspects of the interaction between host plants and the checkerspot butterfly, Euphydryas edith

This dissertation explores some of the ecological and evolutionary interactions of one particular herbivorous insect, the butterfly Euphydryas editha, with a variety of potential host plants. The diet of any particular butterfly population which one observes (i.e. which plant species actually get egg masses on them in the field), can be viewed as a single event peak arising from: (1) the decisions of individual females which lead up to population-level behaviors of the butterfly and (2) resistances of individual plants to initial attack and responses of individuals once attacked, which lead up to population-level susceptibility of each potential host species. I explored how a female butterfly finds host plants, and whether these host-finding mechanisms may change, both during evolution and during the life of a single insect. I then explored how attack by the insects affects host fitness within a single plant population. I discovered circumstances in which resistance might not evolve, even when a plant species is under continual attack.;Specifically, I show that E. editha search for host plants using fixed preferences for visual cues such as leaf shape and plant height. These cues do not change during the life of the insect, but may vary substantially between conspecific butterfly populations. Once the plant has been attacked, herbivory had substantial impact on plant fitness, but only at low and moderate plant densities. At high density, herbivory did not significantly alter plant fitness. The fitness effects at low and moderate densities lead to the expectation that the insect may cause evolution of plant resistance. However, even if resistance were to evolve in a particular plant population that was being attacked, the lack of ability to learn in E. editha would impede its ability to shift on to alternate host plant species at the same site. Additionally, this fixed behavior pattern would constrain the colonization of novel host plants in other potential habitat patches, should a female fly astray from her local population. Thus, although a "co-evolutionary arms race" in which each new plant defense gene is but one step ahead of the development of an herbivore counter-strategy may exists in some systems, this study showed that both E. editha and its associated host-plants are constrained in their ability to evolve defense/exploitation strategies.