Molecular analyses of turtle relationships have overturned prevailing morphological hypotheses and prompted the development of a new taxonomy. Here we provide the first genome-scale analysis of turtle phylogeny. We sequenced 2,381 ultraconserved element (UCE) loci representing a total of 1,718,154 bp of aligned sequence. Our sampling includes 32 turtle taxa representing all 14 recognized turtle families and an additional six outgroups. Maximum likelihood, Bayesian, and species tree methods produce a single resolved phylogeny. This robust phylogeny shows that proposed phylogenetic names correspond to well-supported clades, and this topology is more consistent with the temporal appearance of clades and paleobiogeography. Future studies of turtle phylogeny using fossil turtles should use this topology as a scaffold for their morphological phylogenetic analyses. Molecular analyses of turtle relationships have overturned prevailing morphological hypotheses and prompted the development of a new taxonomy. Here we provide the first genome-scale analysis of turtle phylogeny. We sequenced 2,381 ultraconserved element (UCE) loci representing a total of 1,718,154 bp of aligned sequence. Our sampling includes 32 turtle taxa representing all 14 recognized turtle families and an additional six outgroups. Maximum likelihood, Bayesian, and species tree methods produce a single resolved phylogeny. This robust phylogeny shows that proposed phylogenetic names correspond to well-supported clades, and this topology is more consistent with the temporal appearance of clades and paleobiogeography. Future studies of turtle phylogeny using fossil turtles should use this topology as a scaffold for their morphological phylogenetic analyses.