| Abstract | Bruce (2005) presents provocative arguments in favor of an ancestral biphasic life-history mode for plethodontid salamanders, in particular for the genera formerly assigned to subfamily Desmognathinae Wake 1966: Desmognathus + Phaeognathus, the supergenus Desmognathus (sensu Chippindale et al. 2004). Bruce disputes the inferences of Chippindale et al. (2004) that a reversal from direct development to an aquatic larval stage occurred in Desmognathus (see also Bonett et al. 2005; Mueller et al. 2004). For decades, Bruce has played a leading role in studies of plethodontid life-history evolution (e.g., Bruce et al. 2000), and his disagreement with our conclusions is a cause for concern that must thoroughly be addressed. The possibility that direct development is the ancestral state for supergenus Desmognathus (Desmognathus hereafter) has been discussed by several authors (e.g., Collazo and Marks 1994; Marks 2000; Titus and Larson 1996). The new and largely concordant phylogenetic hypotheses of Chippindale et al. (2004), Mueller et al. (2004), and Macey (2005) provide an opportunity to trace life-history evolution in plethodontids within a rigorous phylogenetic framework. Here, we address Bruce's arguments and show that the preponderance of evidence supports re-evolution of the aquatic larval stage in Desmognathus. Bruce offers six key arguments in favor of biphasic development as the primitive condition for Desmognathus. We will address these in the order in which they appear in his paper. We think that the issues can be distilled to four questions: (1) Is the family-level placement of plethodontids correct? (2) Is the phylogenetic position of supergenus Desmognathus within Plethodontidae correct? (3) Is the ancestral reconstruction for the evolution of direct development correct? (4) How strong is the evidence that competition in terrestrial environments drove the re-evolution of the larval stage?
|
