Morphology and Placental Mammal Phylogeny

Abstract

In a recent article placental mammal phylogeny (Springer et al, 2007), we discussed evidence for correlated character evolution among morphological characters. We also performed pseudoextinction analy ses that assessed whether placental orders remained in the expected superordinal group (Afrotheria, Xenarthra, Euarchontoglires, Laurasiatheria) when molecular and soft-tissue data were coded as missing and only oste ological data from Asher et al. (2003) remained for the pseudoextinct taxa. Finally, we examined congruence among 21 molecular data partitions and Asher et al/s (2003) morphological data. Our results demonstrated that most placental orders moved a different superor dinal group when treated as pseudoextinct and also that Asher et al/s (2003) morphological data consistently emerged as the most incongruent data partition. Based these results, we questioned the ability current morphological data sets and phylogenetic methods reconstruct higher level relationships among placental mammals. In their response our paper, Asher et al. (2008) raise several objections including (1) continued debate over our preferred 4-clade topology (p. 311) that renders our conclusions morphology premature (p. 311); (2) basing our conclusions on a single morpho logical data set (p. 312); (3) our use pseudoextinction techniques to make broad generalizations about the quality data for mammal phylogeny reconstruction (p. 313); and (4) our conditional acknowledgement the primacy of morphological data infer phylogeny fossil taxa (p. 313). Asher et al. (2008) make a number useful points, but as discussed below these do not diminish the main conclusions our earlier paper.