Host race or species? Allozyme characterization of the 'flowering dogwood fly', a member of the Rhagoletis pomonella complex

The term 'flowering dogwood fly' has been used in the literature for a poorly understood member of the Rhagoletis pomonella sibling species complex infesting the fruits of flowering dogwood (Cornus florida). Electrophoretic analysis of 17 allozyme loci in 21 populations reveals significant frequency differences between the flowering dogwood fly and its closest relative the apple maggot fly, R. pomonella, and between it and the somewhat more distant 'sparkleberry fly'. Frequency differences between the flowering dogwood fly and R. pomonella are as great as 0.817 in the north, but are less in the south, with a maximum difference at one site of only 0.328. No fixed allozyme differences distinguish the flowering dogwood fly anywhere; its only consistent, unique feature is the highest frequency of Aat-2⁵⁹ in the pomonella species group. Population structure of the flowering dogwood fly is moderate with F(ST) = 0.084 and fewer latitudinal clines than R. pomonella. The conclusion from the allozyme and life history data is that the flowering dogwood fly is a species, although some interspecific gene flow may be occurring. Additional issues discussed include how to estimate interspecific gene flow when genetic markers are under divergent selection, the appropriate species concept when there is gene flow, and the future of the flowering dogwood fly in the face of the dogwood anthracnose epidemic. The possible utility of a new species concept for phytophagous insects, using as a criterion the capacity of a host race to regenerate the ancestral population, is also discussed.