Abstract
To evaluate the role of the formation of the Central American land bridge in diversification of the American avifauna, Johnson and Weckstein (2011) reconstructed a phylogeny, using DNA sequence data from four gene regions, of 24 nominal species in three putative genera of New World doves. Although a systematic revision of these doves was not a primary purpose of their study, Johnson and Weckstein (2011) provided information that helps to re-evaluate the presumed relationships among the taxa included. Their analysis supported a hypothesis of monophyly for a group containing seven species currently (American Ornithologists’ Union [AOU] 1998, Gibbs et al. 2001, Dickinson 2003, Remsen et al. 2012) placed in the genus Zenaida Bonaparte, 1838, and for a group of 12 taxa in 6 species (of as many as 11 currently recognized species) in the genus Leptotila Swainson, 1837 (Gibbs et al. 2001, Dickinson 2003). However, the 11 species (of up to 16 [Dickinson 2003] or 18 [Gill and Wright 2006]) of the genus Geotrygon Gosse, 1847, occurred in three separate lineages (Johnson and Weckstein 2011: fig. 1), revealing that the genus is polyphyletic. Unfortunately, the type species of Geotrygon, G. versicolor (Lafresnaye), was excluded from their analysis because of the incomplete sequence data available for that species. The purposes of the current study are to provide molecular phylogenetic data for that species and for G. chrysia Bonaparte to determine which lineage includes G. versicolor, and to provide a taxonomic revision of the group. For the few species of the genus for which genetic data are not yet available, only tentative placement based on inference is possible.