Phylogenetic analysis of the direct-developing frogs (Anura, Terrarana, Craugastoridae)

• Autor(a) principal: Ospina-Sarria, Jhon Jairo Ospina
• Data de Publicação: 2018
• Tipo de documento: Tese
• Idioma: eng
• Título da fonte: Biblioteca Digital de Teses e Dissertações da USP
• Texto Completo: http://www.teses.usp.br/teses/disponiveis/41/41133/tde-01032019-140429/
• topic: Análises de evidência total; Morfologia; Morphology; Terrarana; Terrarana; Total evidence analysis

A major recent trend in systematics is the re-integration of morphological data into total evidence analysis. To date, few studies have explored the effects of incorporating morphological and molecular data into total evidence analyses; however, those that have done so have found that even a comparatively small phenomic dataset can have disproportionately large impacts on results. Given the demonstrated importance of morphological characters in testing the phylogenetic relationships, herein I provide a case in point for the exploration of the effects of incorporating morphological evidence as an independent source of phylogenetic evidence, and an important framework to test hypotheses supported by molecular evidence, in the superfamily Brachycephaloidea. In this study, 338 amphibian species are included, of which 318 species correspond to the superfamily Brachycephaloidea. All known genera within the superfamily were sampled. The complete dataset included 13,686 molecular characters (mitochondrial and nuclear markers) and 185 morphological characters, which were analyzed together. In addition, I performed additional analyses modifying the complete datasets to evaluate the effects of character and taxon sampling. As results, I found that the superfamily Brachycephaloidea does not represent a monophyletic as previously thought. Likewise, the phylogeny resulting from this analysis showed several taxa to be nonmonophyletic: Brachycephalidae and Craugastoridae as well as genera Craugastor, Psychrophynella, and Pristimantis. From the detailed examination of the impact of both character and taxon sampling on the phylogenetic relationships of the superfamily Brachycephaloidea, I found that the inclusion of a comparatively small phenomic dataset (185 character) as well as the inclusion of two key taxa (Atopophrynus syntomopus and Dischidodactylus duidensis) had disproportionately large impacts on the tree topology. Finally, I provide a new taxonomy for this group. In this, I recognize five families, of which three are morphologically diagnosable (Eleutherodactylidae, Hypodactylidae, and Strabomantidae) whereas the other two are diagnosable solely with molecular evidence (Ceuthomantidae and Craugastoridae). Likewise, 12 out of 26 genera included are morphologically diagnosable.