| Although there are many ecological explanations for the prevalence of resource specialists, there is little phylogenetic evidence on their origins. In this thesis, I explore both the evolutionary patterns of resource use in a group of herbivorous insects, and the processes of diversification in herbivorous insects as related to their resource use. First, to study the evolutionary patterns of resource use, I provide a molecular phylogenetic study of the nineteen species in the bark beetle genus Dendroctonus (Coleoptera: Scolytidae). Given substantial variation in diet breadth, I ask two questions concerning the evolution of resource use in this group: (1) How conservative is the evolution of host use in these insects? and (2) Do specialists tend to arise from more generalized ancestors?;The phylogeny of Dendroctonus based on mitochondrial DNA sequences revealed that affiliations with Pinus and Picea hosts was conservative among beetle species, but there was no significant correspondence between the phylogeny of the beetles and the phylogeny of their hosts. Degree of specialization, as measured in the proportion of hosts used, was bimodally distributed with "generalist" species utilizing ;Using the phylogenetic history of the bark beetle genus Dendroctonus, I also explored the role of resources (i.e., hostplants) in the diversification of populations. Specifically, I compared the amount of gene flow within the generalist D. ponderosae to that within its sister species, D. jeffreyi. Estimates of gene flow from allozyme and mitochondrial DNA polymorphism revealed slightly more subdivision between generalist populations on various host species than among specialist populations on the same host species. However, a much stronger pattern of differentiation was observed among the specialist populations in the two different locales. Furthermore, I uncovered substantially less mtDNA polymorphism in the specialist than in the generalist. These two observations suggest that a patchier distributions of resources, and subsequent smaller population sizes, may lead to greater opportunities for population subdivision. |
