Diet-morphology correlations in the radiation of South American geophagine cichlids (Perciformes: Cichlidae: Cichlinae).

Genera within the South American cichlid tribe Geophagini display specialized feeding and reproductive strategies, with some taxa specialized for both substrate-sifting and mouth brooding. Several lineages within the clade also possess an epibranchial lobe (EBL), a unique pharyngeal structure that has been proposed to have a function in feeding and/or mouth brooding. A recently published genus-level phylogeny of Neotropical cichlids was used as the evolutionary framework for investigating the evolution of morphological features presumably correlated with diet and mouth brooding in the tribe Geophagini. We tested for possible associations between the geophagine epibranchial lobe and benthic feeding and mouth brooding. We also addressed whether the EBL may be associated with unique patterns of diversification in certain geophagine clades. Tests of binary character correlations revealed the EBL was significantly associated with mouth brooding. We also tested for a relationship between diet and morphology. We analyzed stomach contents and morphometric variation among 21 species, with data for two additional species obtained from the literature. Principal Components Analysis revealed axes of morphological variation significantly correlated with piscivory and benthivory, and both morphology and diet were significantly associated with phylogeny. These results suggest that the EBL could be an adaptation for either feeding or mouth brooding. The EBL, however, was not associated with species richness or accelerated rates of phyletic diversification.

Multi-locus phylogeny of dolphins in the subfamily Lissodelphininae: character synergy improves phylogenetic resolution.

BACKGROUND: Dolphins of the genus Lagenorhynchus are anti-tropically distributed in temperate to cool waters. Phylogenetic analyses of cytochrome b sequences have suggested that the genus is polyphyletic; however, many relationships were poorly resolved. In this study, we present a combined-analysis phylogenetic hypothesis for Lagenorhynchus and members of the subfamily Lissodelphininae, which is derived from two nuclear and two mitochondrial data sets and the addition of 34 individuals representing 9 species. In addition, we characterize with parsimony and Bayesian analyses the phylogenetic utility and interaction of characters with statistical measures, including the utility of highly consistent (non-homoplasious) characters as a conservative measure of phylogenetic robustness. We also explore the effects of removing sources of character conflict on phylogenetic resolution. RESULTS: Overall, our study provides strong support for the monophyly of the subfamily Lissodelphininae and the polyphyly of the genus Lagenorhynchus. In addition, the simultaneous parsimony analysis resolved and/or improved resolution for 12 nodes including: (1) L. albirostris, L. acutus; (2) L. obscurus and L. obliquidens; and (3) L. cruciger and L. australis. In addition, the Bayesian analysis supported the monophyly of the Cephalorhynchus, and resolved ambiguities regarding the relationship of L. australis/L. cruciger to other members of the genus Lagenorhynchus. The frequency of highly consistent characters varied among data partitions, but the rate of evolution was consistent within data partitions. Although the control region was the greatest source of character conflict, removal of this data partition impeded phylogenetic resolution. CONCLUSION: The simultaneous analysis approach produced a more robust phylogenetic hypothesis for Lagenorhynchus than previous studies, thus supporting a phylogenetic approach employing multiple data partitions that vary in overall rate of evolution. Even in cases where there was apparent conflict among characters, our data suggest a synergistic interaction in the simultaneous analysis, and speak against a priori exclusion of data because of potential conflicts, primarily because phylogenetic results can be less robust. For example, the removal of the control region, the putative source of character conflict, produced spurious results with inconsistencies among and within topologies from parsimony and Bayesian analyses.

Molecular phylogeny and evidence for an adaptive radiation of geophagine cichlids from South America (Perciformes: Labroidei).

Nucleotide sequences from the mitochondrial ND4 gene and the nuclear RAG2 gene were used to derive the most extensive molecular phylogeny to date for the Neotropical cichlid subfamily Geophaginae. Previous hypotheses of relationships were tested in light of these new data and a synthesis of all existing molecular information was provided. Novel phylogenetic findings included support for : (1) a 'Big Clade' containing the genera Geophagus sensu lato, Gymnogeophagus, Mikrogeophagus, Biotodoma, Crenicara, and Dicrossus; (2) a clade including the genera Satanoperca, Apistogramma, Apistogrammoides, and Taeniacara; and (3) corroboration for Kullander's clade Acarichthyini. ND4 demonstrated saturation effects at the third code position and lineage-specific rate heterogeneity, both of which influenced phylogeny reconstruction when only equal weighted parsimony was employed. Both branch lengths and internal branch tests revealed extremely short basal nodes that add support to the idea that geophagine cichlids have experienced an adaptive radiation sensu Schluter that involved ecomorphological specializations and life history diversification.

Higher-level systematics of rodents and divergence time estimates based on two congruent nuclear genes.

Phylogenetic analysis of over 4600 aligned nucleotide sequences from two nuclear genes, growth hormone receptor and BRCA1, provided congruent phylogenies depicting relationships among the major lineages of rodents. Separate and combined analyses resulted in five major conclusions: (1) strong support for a monophyletic Myodonta (containing the superfamilies Muroidea + Dipodoidea), with subfamily Gerbillinae being more closely related to Murinae than is Sigmodontinae; (2) a sister-group relationship between the family Castoridae and the superfamily Geomyoidea; (3) monophyly of Ctenohystrica (containing the suborders Sciuravida and Hystricognatha); (4) a near polytomy among Myodonta (suborder Myomorpha), Pedetes (family Pedetidae, suborder Anomaluromorpha), Castoridae (suborder Sciuromorpha) + Geomyoidea (suborder Myomorpha), and Ctenohystrica; and (5) basal position of a monophyletic group containing Graphiurus (family Gliridae, suborder Myomorpha) + two members of the Sciuromorpha (Sciuridae + Aplodontidae). Divergence dates among rodents and primates were also estimated using the combined data. Applying a global molecular clock and a primate calibration point, divergence dates among rodents exceeded fossil-based dates but were generally compatible with other molecule-based dates estimated under similar conditions. However, when a relaxed molecular clock was applied, estimated divergence dates were highly compatible with the fossil record.