FLIGHTLESSNESS AND PHYLOGENY AMONGST ENDEMIC RAILS (AVES, RALLIDAE) OF THE NEW-ZEALAND REGION

The phylogenetic relationships of a number of flightless and volant ra ils have been investigated using mtDNA sequence data. The third domain of the small ribosomal subunit (12S) has been sequenced for 22 taxa, and part of the 5' end of the cytochrome-b gene has been sequenced for 12 taxa. Additional sequences were obtained from outgroup taxa, two s pecies of jacana, sarus crane, spur-winged plover and kagu. Extinct ra ils were investigated using DNA extracted from subfossil bones, and in cases where fresh material could not be obtained from other errant ta xa, feathers and museum skins were used as sources of DNA. Phylogeneti c trees produced from these data have topologies that are, in general, consistent with data from DNA-DNA hybridization studies and recent in terpretations based on morphology. Gallinula chloropus (moorhen) group s basally with Fulica (coots), Amaurornis (= Megacrex) ineptus falls w ithin the Gallirallus/Rallus group, and Gallinula (= Porphyrula) matin ica is basal to Porphyrio (swamphens) and should probably be placed in that genus. Subspecies of Porphyrio porphyrio are paraphyletic with r espect to Porphyrio mantelli (takahe). The Northern Hemisphere Rallus aquaticus is basal to the south-western Pacific Rallus (or Gallirallus ) group. The flightless Rallus philippensis dieffenbachii is close to Rallus modestus and distinct from the volant Rallus philippensis, and is evidently a seperate species. Porzana (crakes) appears to be more c losely associated with Porphyrio than Rallus. Deep relationships among the rails remain poorly resolved. Rhynochetos jubatus (kagu) is close r to the cranes than the rails in this analysis. Genetic distances bet ween flightless rails and their volant counterparts varied considerabl y with observed 12S sequence distances, ranging from 0.3 % (Porphyrio porphyrio melanotus and P. mantelli mantelli) to 7.6 % (Rallus modestu s and Rallus philippensis). This may be taken as an indication of the rapidity with which flightlessness can evolve, and of the persistence of flightless taxa. Genetic data supported the notion that flightless taxa were independently derived, sometimes from similar colonizing anc estors. The morphology of flightless rails is apparently frequently do minated by evolutionary parallelism although similarity of external ap pearance is not an indication of the extent of genetic divergence. In some cases taxa that are genetically close are morphologically distinc t from one another (e.g. Rallus (philippensis) is) dieffenbachii and R . modestus), whilst some morphologically similar taxa are evidently in dependently derived (e.g. Porphyio mantelli hochstetteri and P. m. man telli).