THE EFFECTS OF WINTER LENGTH ON THE GENETICS OF APPLE AND HAWTHORN RACES OF RHAGOLETIS POMONELLA (DIPTERA: TEPHRITIDAE).

Host plant-associated fitness trade-offs are central to models of sympatric speciation proposed for certain phytophagous insects. But empirical evidence for such trade-offs is scant, which has called into question the likelihood of nonallopatric speciation. Here, we report on the second in a series of studies testing for host-related selection on pupal life-history characteristics of apple- (Malus pumila L.) and hawthorn- (Crataegus mollis L. spp.) infesting races of the Tephritid fruit fly, Rhagoletis pomonella (Walsh). In particular, we examine the effects of winter length on the genetics of these flies. We have previously found that the earlier fruiting phenology of apple trees exposes apple-fly pupae to longer periods of warm weather preceding winter than hawthorn-fly pupae. Because R. pomonella has a facultative diapause, we hypothesized that this selects for pupae with more recalcitrant pupal diapauses (or slower metabolic/development rates) in the apple-fly race. A study in which we experimentally manipulated the length of the prewintering period for hawthorn-origin pupae supported this prediction. If the period preceding winter is important for apple- and hawthorn-fly pupae, then so too should be the length (duration) of winter; the rationale for this prediction is that "fast developing" pupae that break diapause too early will deplete their energy reserves and disproportionately die during long winters. To test this possibility, we chilled apple- and hawthorn-origin pupae collected from a field site near Grant, Michigan, in a refrigerator at 4°C for time periods ranging from one week to two years. Our a priori expectation was that longer periods of cold storage would select against allozyme markers that were associated with faster rates of development in our earlier study. Since these electromorphs are typically found at higher frequencies in hawthorn flies, extending the overwintering period should favor "apple-fly alleles" in both races. The results from this "overwinter" experiment supported the diapause hypothesis. The anticipated genetic response was observed in both apple and hawthorn races, as allele frequencies became significantly more "apple-fly-like" in eclosing adults surviving longer chilling periods. This indicates that it is the combination of environmental conditions before and during winter that selects on the host races. Many tests for trade-offs fail to adequately consider the interplay between insect development, host plant phenology, and local climatic conditions. Our findings suggest that such oversight may help to explain the paucity of reported fitness trade-offs.