Abstract:
Parentia is a genus of the long-legged flies (Dolichopodidae: Diptera). This relatively unstudied genus is believed to include members endemic to New Zealand, Australia, and New Caledonia. One of the aims of this research was to produce the first molecular phylogeny of Parentia, using Maximum Likelihood and Bayesian approaches, for a comparison against the existing morphological phylogeny. The intent was to do this by extracting DNA from identified museum specimens, as well as freshly collected specimens. Another aim of this research was to develop an existing method of non-destructive DNA extraction for museum specimens of Diptera, to extract DNA for the phylogenetic analysis. This was done by varying extraction conditions for test specimens of Diptera (genus: Coelopella). The final aim of this research was to compare the original morphological analysis of the New Zealand Parentia, originally carried out using traditional metrics techniques, with data gathered using a geometric morphometric approach. The morphometric data were used to test physical characters (genital and Male Secondary Sexual Characters), used in the original phylogeny, for their usefulness in classification and taxonomy. The New Zealand Parentia also exhibit a complex biogeographic pattern, with some species overlapping in their geographic ranges, while others are isolated. It was decided to test whether there was a correlation between variation in reproductive and secondary sexual characters, and biogeographic overlap within the New Zealand Parentia. It was found that it was possible to develop a method of non-destructive DNA extraction for use on small museum specimens of Diptera, which recovered sufficient amounts of DNA (range 50-800 ng/μl) for subsequent analysis, and did not destroy the morphological integrity of the specimens. The phylogenetic analysis revealed significant incongruence between the morphological and the molecular trees (P<0.0001). Finally, the geometric morphometric analysis revealed that only one genital character could be confidently claimed to be reliable for classification. Three other characters did show separation between species, but the differences were not significant. The biogeographic comparison with the morphometric data revealed no correlation between variation in sexual characters and geographic overlap. Thus, interesting new details have been discovered about this relatively unknown genus.