| Abstract: | The knowledge of evolutionary relationships among oscine songbirds has been largely improved in recent years by molecular phylogenetic studies. However, current knowledge is still largely based on sequence data from a limited number of loci. In this study, we re-evaluate relationships among basal lineages within the 'core Corvoidea' and Passerida radiations, by adding additional loci to previously published data. The trees obtained from the individual genes suggest incongruent topologies. Especially the positions of Callaeatidae (wattlebirds), Cnemophilidae (satinbirds) and Melanocharitidae (longbills and berrypeckers) vary among the trees, but RAG-1 is the only gene that unambiguously suggested a 'core Corvoidea' affinity for these taxa. Analyses of various combined data sets show that the phylogenetic positions for Callaeatidae, Cnemophilidae and Melanocharitidae largely depend on which genes that have been combined. As the RAG-1 gene has contributed to a majority of the phylogenetic information in previous studies, it has deeply influenced previous molecular affinities of these taxa. Based on the current data, we found a reasonable support for a Passerida affinity of Callaeatidae and Cnemophilidae, contrary to previous molecular studies. The position of Melanocharitidae is more unstable but a basal position among Passerida is congruent with a deletion observed in the glyceraldehyde-3-phosphodehydrogenase (GAPDH) loci. Molecular clock estimations conducted on the combined data sets were generally found to be similar, but for some divergences significant differences were found. These results illustrate the potential problem of phylogenies predominantly based on characters from one or a few loci, and exemplify the importance of well-supported phylogenies before reasonable time estimates of passerine divergences could be achieved. |
