Neonate aggregation in the Permian dicynodont Diictodon (Therapsida, Anomodontia): Evidence for a reproductive function for burrows?

Abstract In the late 1980's the discovery of late Permian helical burrow casts containing articulated skeletons of the small herbivorous therapsid Diictodon feliceps led to conjecture that they may have been used for oviposition/parturition and shelter for infants. Here we present new fossil evidence in support of this interpretation and discuss the possibility that some of the burrows were specially excavated as brood chambers. A re-investigation of the original helical burrow site recovered several more burrow casts containing scattered yet still-associated skeletons of Diictodon. Mechanical preparation of a complete terminal chamber revealed a disarticulated but anatomically-associated adult Diictodon skeleton along with a single tiny (5 mm long) humerus of an infant dicynodont. Nearby outcrops yielded a second association of an adult Diictodon skull (skull length 93 mm) on top of a tiny semi-articulated Diictodon skull and skeleton (skull length 19 mm) with a second infant mandible and a skeleton of the gracile therocephalian Ictidosuchoides longiceps. Synchrotron imaging of this putative burrow-fill confirmed that the humeral morphology of the infant skeleton closely matches the tiny humerus in the unequivocal burrow cast. The common occurrence of Diictodon remains within the casts, combined with their specialized limbs for digging and histological data that indicates uninterrupted growth to ca. 70% of adult size, strongly suggests that they dug underground primarily for thermo-regulation. Moreover, our new fossil evidence of behaviourally-associated neonate and adult Diictodon within these structures indicates that the terminal portions of underground burrows were facultatively used as brood chambers.