Abstract
Living reptiles include more than 20,000 species with disparate ecologies. Direct anatomical evidence from Neodiapsida, which includes the reptile crown-group Sauria and its closest extinct relatives, shows that this diversity originates from a single common reptilian ancestor that lived some 255 million years ago in the Paleozoic. However, the evolutionary assembly of crown reptile traits is poorly understood due to the lack of anatomically close outgroups to Neodiapsida1–7. We present a substantially revised phylogenetic hypothesis, informed by new anatomical data derived from high-resolution synchrotron tomography of Paleozoic reptiles8–12. We find strong evidence placing the clade Millerettidae as a close sister to Neodiapsida, which uniquely share a suite of derived features among Paleozoic stem reptiles. This grouping, for which we name the new clade Parapleurota, replaces previous phylogenetic paradigms by rendering the group Parareptilia as a polyphyletic assemblage of stem reptiles, of which millerettids are the most crownward. Our analysis presents hypotheses that resolve long-standing issues in Paleozoic reptile evolution, including the placement of captorhinids on the amniote stem lineage and firm support for varanopids as synapsids, which taken together provide a greatly improved fit to the observed stratigraphic record. Optimizations of character evolution on our phylogenetic hypothesis reveals gradual assembly of crown reptile anatomy, including a Permian origin of tympanic hearing, the presence of a lower temporal fenestra in the amniote common ancestor, with subsequent modifications on the reptile stem lineage, leading to the loss of the lower temporal bar. This evolutionary framework provides a platform for investigating the subsequent radiations of the reptile crown group in the Early Triassic, including the lines leading to dinosaurs (including birds), crocodilians, lepidosaurs, and extinct marine reptiles.
Competing Interest Statement
The authors have declared no competing interest.