Abstract
Vermilingua is a peculiar group of xenarthran placental mammals with well-established phylogenetic relationships based on morphological and molecular data, but until now, no combined analysis of those data types. Here we perform the first total-evidence phylogenetic analysis of Vermilingua and estimate divergence times for living and fossil anteaters with a tip-dating approach, using the Fossilized Birth-Death model in a Bayesian framework. Using combined analyses, we recovered as monophyletic the two families, Cyclopedidae and Myrmecophagidae, as in previous studies, although our revised morphological data alone do not support the monophyly of Cyclopedidae. We show how the combination of molecular and morphological data indirectly affects the position and support for fossil taxa even though molecular datasets do not contain direct information for them. The estimated divergence times of Palaeomyrmidon and Protamandua are much older than the ages of those fossils, reinforcing the scarcity of the vermilinguan fossil record. In addition, we obtained evidence that the extinct Neotamandua is an ancestor of the extant Myrmecophaga, not a sister taxon as in previous studies, that did not incorporate the possibility of anagenetic evolution between fossil and living taxa.
Acknowledgments
We would like to acknowledge Martin Ezcurra and Laura Chornogubsky (Museo Argentino de Ciencias Naturales - MACN), and Claudia Guimarães Costa (Pontifícia Universidade de Minas Gerais - PUC-MG) for kindly allowing us to study specimens under their care, Rafaela Missagia (Universidade Federal de Minas Gerais - UFMG), Leonardo Cotts (Universidade Federal do Rio de Janeiro - UFRJ) and William Simpson (Field Museum of Natural History - FMNH) for photographing and sharing photos of vermilinguans, Fredrik Ronquist and Chi Zhang for help interpreting zero tip branch lengths in the implementation of the FBD model in MrBayes, Cayo Dias (UFMG) for writing and sharing the Phyton script used to count branch lengths in MrBayes outputs, and Mark Puttick for assistance in using the MCMCtreeR package. Rafaela Missagia also provided useful comments on an early version of the manuscript. Kenneth Angielczyk (FMNH) kindly helped us to revise a late version of the manuscript, providing many suggestions, which helped to improve our English and the manuscript content as a whole. We are also indebted to two anonymous reviewers and the editor who provided useful comments that helped us to improve the manuscript. We thank Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) for financial support via a monthly scholarship (Finance Code 0001) and Programa de Apoio à Pós-Graduação (PROAP). We also thank FAPEMIG, Fundação O Boticário and CNPq for research grants for this study.
Disclosure statement
No potential conflict of interest was reported by the authors.
Supplemental data
Supplemental data for this article can be accessed here: https://doi.org/10.1080/14772000.2020.1729894.
Associate Editor: Christoph Bleidorn