SPECIATION IN THE GENUS CRINIA (ANURA: MYOBATRACHIDAE) IN SOUTHERN AUSTRALIA: A PHYLOGENETIC ANALYSIS OF ALLOZYME DATA SUPPORTING ENDEMIC SPECIATION IN SOUTHWESTERN AUSTRALIA.

Abstract

Where there are several closely related species with sympatric distributions occurring in an isolated habitat, the cause is generally ascribed to multiple invasion (Mayr, 1963 p. 504). White (1978p. 115) has criticized the indiscriminate application of this principle in the Australian biota in the following way: such hypotheses are produced solely to satisfy the a priori postulate that speciation is only possible where populations are completely separated by major topographic barriers. Interpretations on the origin of the frog fauna of southwestern Australia have suffered in this regard, in that the region is topographically featureless (Prider, 1966) and isolated by desert from the rest of Australia while the frog species there have largely sympatric distributions. Thus it was postulated that frogs from southeastern Australia crossed the arid Nullarbor plain during several cycles of increased rainfall caused by Pleistocene glaciation. Speciation occurred after populations were separated each time the Nullarbor plain again became arid. The direction of invasion was determined on the basis of the number of species found in each area so that invasion proceeded from centres of low species number to areas which now have many species. Among the birds of southern Australia, for example, invasion is postulated to have gone from west to east (Keast, 1961). In the case of frogs, the proposed invasions were from east to west, and most authors have assumed that each southwestern species is most closely related to a southeastern species (e.g., Main et al., 1958; Martin, 1972; Littlejohn, 1981) and that several southwestern species may be related to one southeastern species. White (1978 p. 113) has challenged this interpretation, suggesting that the number of species in southwestern Australia is the result of endemic speciation, as such an hypothesis makes fewer assumptions. According to this interpretation, each southwestern species is most closely related to some other southwestern species. These alternative biogeographic hypotheses imply different phylogenies and these can be tested by an independent phylogenetic analysis. In the present study the alternative evolutionary scenarios were tested using seven species from the genus Crinia (six of these species were formerly in Ranidella [Blake, 1973] but were returned to Crinia by Heyer et al., 1982). The characters used were allozymes as most of these species are morphologically indistinguishable. Crinia is of especial interest as the first hypothesis invoking multiple invasion for the origin of frog species in southern Australia was formulated for this genus, before being extended to include not only Heleioporus and Neobatrachus (Main et al., 1958) but three other genera as well (Main, 1968).