DNA taxonomy: How many DNA sequences are needed for solving a taxonomic problem? The case of two parapatric species of louse flies (Diptera: Hippoboscidae: Ornithomya Latreille, 1802)

Abstract

Using molecular and morphological data, we investigate the validity of two hippoboscid species, viz. Ornithomya fringillina (Curtis) and O. chloropus (Bergroth) that are parapatric in northern Europe and had previously been synonymized. We study four morphological characters that had been proposed as diagnostic for species separation, but only one, relatively weak character (a gena marking), has discontinuous variation across the putative species. In order to collect more evidence for deciding on the species status, we sequenced an approximately 810 bp long region of COI for 13 specimens from sympatric populations. The signal from the sequences suggests that O. fringillina and O. chloropus are different species because the interspecifi c genetic distances between the taxa are twenty times larger than the intraspecifi c variability of O. fringillina. We argue that even a small number of sequences can yield signifi cant information on taxonomic issues as long as the specimens are predominantly collected for (1) those species/populations whose status is diffi cult to resolve based on morphological information and (2) those specimens that come from sympatric populations of the “problematic” species. (3) We also suggest that the status of a rare species can be adequately addressed with very few sequences as long as the intraspecifi c variability of more common, close relatives have been adequately assessed.