DID END-PERMIAN TETRAPODS GASP THEIR LAST BREATHS? TESTING HYPOXIA AS AN EXTINCTION MECHANISM WITH DICYNODONT THERAPSIDS (SYNAPSIDA)

The cause of the end-Permian mass extinction has been difficult to reconstruct, with explanations ranging from an extraterrestrial impact to global warming caused by the eruption of the Siberian flood basalts. Recently, Retallack and others proposed a release of oceanic methane as the ultimate cause of the extinction in the marine and terrestrial realms. They noted that release and oxidation of a large amount of methane would reduce the atmospheric partial pressure of oxygen, and proposed the mechanism of hypoxia as the proximate cause of the vertebrate extinction on land. Furthermore, they cited several morphological features that characterize the surviving tetrapods, including barrel chests and greater separation of the mouth and nasal cavities, as evidence of selection for increased respiratory efficiency that is consistent with their scenario.

Dicynodont therapsids are abundant and well-studied Late Permian and Early Triassic tetrapods. Because at least two major Permian lineages survived into the Triassic, the clade provides an ideal natural test of whether selection for increased respiratory efficiency occurred. We measured the area of the internal nares and the length of the secondary palate in over 250 Permian and Triassic dicynodont specimens to test Retallack et al.’s hypothesis that the internal nares of Triassic dicynodonts were relatively smaller than those of Permian forms (creating a less obstructed airway), as well as the alternative hypothesis that there was no significant difference in nares size between Permian and Triassic taxa. Statistical comparisons were carried out in a phylogenetic context using independent contrasts to account for similarities caused by common ancestry. Our preliminary results imply that nares size is influenced by phylogeny, and may have increased across the boundary. These findings may be consistent with a hypoxia kill mechanism, but also could be correlated with other functional changes in the dicynodont skull.