Global Phylogeography of Delphinine Dolphins Using Mitogenomics

Madeleine Becker1,2, Katherine Murphy3, Charles Potter2, Ibrahima Ndong4, Lucy Keith-Diagne4 and Michael R McGowen2, (1)University of Southern California, Biological Sciences, Los Angeles, CA, United States, (2)National Museum of Natural History, Smithsonian Institution, Vertebrate Zoology, Washington, DC, United States, (3)National Museum of Natural History, Laboratories of Analytical Biology, Washington, DC, United States, (4)African Aquatic Conservation Fund, Mbour, Senegal
Abstract:
In this study, we investigated the dolphin subfamily Delphininae, with particular emphasis on the genus Delphinus. Since 1994, short and long-beaked morphotypes of common dolphins have been separated as distinct, sympatric, cosmopolitan species: D. delphis (short-beaked) and D. capensis (long-beaked). An ultra long-beaked morphotype ‘tropicalis’ has also been hypothesized as a subspecies of D. capensis or a species in a monophyletic group with D. capensis. While the D. capensis designation was consistent with skull morphology of short and long-beaked morphotypes in California, genetic and more global assays have called the validity of this species into question. Using a global dataset of whole mitochondrial sequences, this study seeks to resolve some of the taxonomic issues of Delphininae with a concentration on resolving phylogeographic relationships within Delphinus. Here we have compiled the most genetic data for Delphinus populations to date. We found that long-beaked Delphinus samples from the East North Pacific formed a monophyletic group within D. delphis samples, but the ultra long-beaked individual was not included in this clade. RAxML and Mr. Bayes analysis also showed unclear geographic groupings within Delphinus, with some groups showing geographic structure— potentially explained by occasional migration between different ocean basins or persistence of multiple mitochondrial lineages through time due to large population sizes. Our Delphininae tree had high RAxML support, but low Mr. Bayes support at basal nodes, yet clear polyphyly is seen for both Tursiops and Stenella.