A FIELD TEST FOR HOST-PLANT DEPENDENT SELECTION ON LARVAE OF THE APPLE MAGGOT FLY, RHAGOLETIS POMONELLA.

Host-plant dependent fitness trade-offs refer to traits that enhance the performance of an insect on one plant species to its detriment on others. Such trade-offs are central to models of sympatric speciation via host shifts, but have proven difficult to empirically demonstrate. Here, we test for host-plant dependent selection on larvae of apple (Malus pumila L.)- and hawthorn (Crataegus mollis L. spp.)-infesting races of Rhagoletis pomonella (Walsh). Samples of larvae were reared in the field and under protective conditions in a garage. Our rationale was that the garage should slow rates of fruit rot relative to the field, relaxing selection pressures associated with declining fruit quality. Four findings emerged from the study. (1) Larvae suffered higher mortality in fruits in the field than the garage. (2) The increase in mortality was greater for larvae in haws. (3) Larvae possessing the alleles Me 100, Acon-2 95, and Mpi 37, three allozymes displaying host-related differentiation in R. pomonella that map to linkage group II in the fly, left fruits earlier than other genotypes. (4) Allele frequencies for Me 100, Acon-2 95, and Mpi 37 were significantly higher in both apple and haw larvae surviving the field versus the garage treatment. Our results suggested that field conditions favored larvae that rapidly developed and left rotting fruits. Since these individuals tended to possess the alleles Me 100, Acon-2 95, and Mpi 37, frequencies of these allozymes were higher in the field. Selection on larvae was directional for Me 100, Acon-2 95, and Mpi 37 (or linked genes) in both host races. We previously showed that these same alleles can be disfavored in the pupal stage, especially in the apple race, where they correlate with premature diapause termination. Fitness trade-offs in Rhagoletis may therefore be due as much to differences in the relative strengths of directional selection pressures acting on different life stages as to disruptive selection affecting any one particular stage. The necessity to consider details of the entire life-cycle highlights one of the many challenges posed to documenting fitness trade-offs for phytophagous insects.