The effects of climate, host plant phenology and host fidelity on the genetics of apple and hawthorn infesting races ofRhagoletis pomonella

True fruit flies belonging to theRhagoletis pomonella (Walsh) sibling species complex have been proposed to speciate sympatrically by shifting and adapting to new host plants. Here, we report the results from a series of ecological and genetic experiments conducted at a study site near Grant, Michigan, U.S.A., aimed at clarifying the relationship between host specialization and reproductive isolation for these flies. Our findings indicate that apple (Malus pumila) and hawthorn (Crataegus mollis) infesting populations ofR. pomonella are partially allochronically isolated. Differences in the timing of adult eclosion account for part of the allochronic divergence, as apple adults emerge approximately ten days earlier than hawthorn flies in the field. Genetic analyses across different life history stages of the fly show that adults do not randomly move between apple and hawthorn trees, but trend to attack the same species of plant that they infested as larvae. Estimates of interhost migration from the allozyme data suggest that from 2.8 to 10% of the apple population is of hawthorn origin and that over 20% of the hawthorn population is of apple origin. The length and quality of the growing season appear to affect the genetic composition of the host races, as allele frequencies in the hawthorn population are correlated with ambient temperature and rainfall during the spring of the preceding year. Finally, allele frequencies for six allozyme loci displaying host associated differentiation also show significant linear regressions with the timing of adult eclosion within both races. These regressions establish a link between allozyme loci displaying inter-host differentiation and a developmental trait (adult eclosion) responsible for partially isolating the races. The slopes of the regressions are paradoxical, however, as they suggest that apple adults should eclose later, not earlier, than hawthorn flies. We conclude by discussing potential resolutions to the eclosion time paradox.