PHYLOGENETIC-RELATIONSHIPS AMONG AUSTRALIAN ELAPID SNAKES - THE SOFT ANATOMICAL DATA RECONSIDERED

On the basis of an extensive set of visceral and scale characters, Wal lach (1985) proposed a detailed phylogenetic scheme for all the Austra lian elapids, down to species level. The shortest tree found in that a nalysis is here shown to contain 592 steps. However, a re-analysis of the same data using PAUP 3.1.1 reveals that there are 258 most parsimo nious trees, each with only 578 steps. The strict consensus of these t rees is much less resolved than Wallach's tree, and has a different to pology. For example, Echiopsis is most closely related to Suta fasciat a rather than to the Notechis lineage, and Demansia is more closely re lated to advanced elapids (such as the Notechis lineage) than to Oxyur anus and Pseudonaja. Many of the larger (suprageneric) groupings propo sed by Wallach are paraphyletic in the PAUP consensus tree. Almost all the groupings in this tree, however, can be collapsed with the additi on of a single extra step. There are more than 32 000 cladograms at 57 9 steps, one step longer than the 258 most parsimonious cladograms. A strict consensus tree of cladograms 578 and 579 steps long is almost c ompletely unresolved. The visceral and external morphological traits, therefore, are not as phylogenetically informative as previously propo sed, at least with respect to the Australian elapid radiation. These t ypes of characters might not be very phylogenetically informative at h igher (intergeneric) levels, although much more data are required to t est this hypothesis.