The mystery of a missing bone: revealing the orbitosphenoid in basal Epicynodontia (Cynodontia, Therapsida) through computed tomography.

The basal non-mammaliaform cynodonts from the late Permian (Lopingian) and Early Triassic are a major source of information for the understanding of the evolutionary origin of mammals. Detailed knowledge of their anatomy is critical for understanding the phylogenetic transition toward mammalness and the paleobiological reconstruction of mammalian precursors. Using micro-computed tomography (µCT), we describe the internal morphology of the interorbital region that includes the rarely fossilized orbitosphenoid elements in four basal cynodonts. These paired bones, which are positioned relatively dorsally in the skull, contribute to the wall of the anterior part of the braincase and form the floor for the olfactory lobes. Unlike procynosuchids and the more basal therapsids in which the orbitosphenoids are well developed, dense, and bear a ventral keel, the basal epicynodonts Cynosaurus, Galesaurus, and Thrinaxodon display cancellous, reduced, and loosely articulated orbitosphenoids, a condition shared with many eucynodonts. The hemi-cylindrical orbitosphenoid from which the mammalian condition is derived re-evolved convergently in traversodontid and some probainognathian cynodonts.

Aggregations and parental care in the Early Triassic basal cynodonts Galesaurus planiceps and Thrinaxodon liorhinus.

Non-mammaliaform cynodonts gave rise to mammals but the reproductive biology of this extinct group is still poorly known. Two exceptional fossils of Galesaurus planiceps and Thrinaxodon liorhinus, consisting of juveniles closely associated with an adult, were briefly described more than 50 years ago as examples of parental care in non-mammaliaform cynodonts. However, these two Early Triassic fossils have largely been excluded from recent discussions of parental care in the fossil record. Here we re-analyse these fossils in the context of an extensive survey of other aggregations found in these two basal cynodont taxa. Our analysis revealed six other unequivocal cases of aggregations in Thrinaxodon, with examples of same-age aggregations among immature or adult individuals as well as mixed-age aggregations between subadult and adult individuals. In contrast, only one additional aggregation of Galesauruswas identified. Taking this comprehensive survey into account, the two previously described cases of parental care in Galesaurus and Thrinaxodon are substantiated. The juveniles are the smallest specimens known for each taxon, and the size difference between the adult and the two associated juveniles is the largest found for any of the aggregations. The juveniles of Thrinaxodon are approximately only 37% of the associated adult size; whereas in Galesaurus, the young are at least 60% of the associated adult size. In each case, the two juvenile individuals are similar in size, suggesting they were from the same clutch. Even though parental care was present in both Galesaurus and Thrinaxodon, intraspecific aggregations were much more common in Thrinaxodon, suggesting it regularly lived in aggregations consisting of both similar and different aged individuals.

Cranial Ontogeny of the Early Triassic Basal Cynodont Galesaurus planiceps.

Ontogenetic changes in the skull and mandible of thirty-one specimens of Galesaurus planiceps, a basal non-mammaliaform cynodont from the Early Triassic of South Africa, are documented. The qualitative survey indicated eight changes in the craniomandibular apparatus occurred during growth, dividing the sample into three ontogenetic stages: juvenile, subadult, and adult. Changes in the temporal region, zygomatic arch, occiput, and mandible occurred during the transition from the subadult to adult stage at a basal skull length of 90 mm. At least four morphological and allometric differences divided the adult specimens into two morphs, indicating the presence of sexual dimorphism in Galesaurus. Differences include extensive lateral flaring of the zygomatic arches in the "male" morph resulting in a more anterior orientation of the orbits, and a narrower snout in the "female". This is the first record of sexual dimorphism in a basal cynodont, and the first time it is quantitatively documented in a non-mammaliaform cynodont. An ontogenetic comparison between Galesaurus and the more derived basal cynodont Thrinaxodon revealed differences in the timing and extent of sagittal crest development. In Galesaurus, the posterior sagittal crest, located behind the parietal foramen, developed relatively later in ontogeny, and the anterior sagittal crest rarely formed suggesting the anterior fibres of the temporalis were less developed than in Thrinaxodon. In contrast, craniomandibular features related to the masseters became more developed during the ontogeny of Galesaurus. The development of the adductor musculature appears to be one of the main factors influencing skull growth in these basal non-mammaliaform cynodonts. Anat Rec, 300:353-381, 2017. © 2016 Wiley Periodicals, Inc.

Ontogeny of the Early Triassic Cynodont Thrinaxodon liorhinus (Therapsida): Cranial Morphology.

The cranial morphology of 68 Thrinaxodon liorhinus specimens, ranging in size from 30 to 96 mm in basal skull length, is investigated using both qualitative and quantitative analyses. From this comprehensive survey, we determined that nine cranial features, including five in the temporal region, separated the sample into four ontogenetic stages. A bivariate analysis of 60 specimens indicated that the skull generally increased in size isometrically, with the exception of four regions. The orbit had negative allometry, a result consistent with other ontogenetic studies of tetrapods, whereas the length of the snout, palate, and temporal region showed positive allometry. The last trend had strong positive allometry indicating that during ontogeny the length of the sagittal crest increased at a much faster rate than the rest of the skull. The large number of changes in the temporal region of the skull of Thrinaxodon may indicate a greater development of the posterior fibres of the temporalis musculature from an early ontogenetic stage. For example, the posterior sagittal crest developed much earlier in ontogeny than the anterior crest that formed in adults, and bone was deposited dorsally creating a unified posterior sagittal crest rather than having a suture that spanned the entire depth of the skull roof. In combination with the isometric height of the zygomatic arch and the almost complete absence of the zygomatic arch angulation, these ontogenetic changes suggest that there was greater development of the temporalis relative to the masseter muscles, indicating a strong posterodorsal movement of the mandible in Thrinaxodon.

Functional implications of dicynodont cranial suture morphology.

Cranial suture morphology of Lystrosaurus and the generalized dicynodont Oudenodon was investigated to determine the strain environment during mastication, which in turn may indicate a difference in cranial function between the two taxa. Finite element (FE) analysis indicated that less strain accumulated in the cranium of Lystrosaurus during orthal bite simulations than in Oudenodon. Despite the overall difference in strain magnitude, moderate to high FE-predicted strain accumulated in similar areas of the cranium of both taxa. The suture morphology in these cranial regions of Lystrosaurus and Oudenodon was investigated further by examination of histological sections and supplemented by observations of serial sections and computed tomography (CT) scans. The predominant type of strain from selected blocks of finite elements that contain sutures was determined, enabling comparison of suture morphology to strain type. Drawing from strain-suture correlations established in extant taxa, the observed patterns of sutural morphology for both dicynodonts were used to deduce cranial function. The moderate to high compressive and tensile strain experienced by the infraorbital bar, zygomatic arch, and postorbital bar of Oudenodon and Lystrosaurus may have been decreased by small adjustive movements at the scarf sutures in those regions. Disparities in cranial suture morphology between the two taxa may reflect differences in cranial function. For instance, the tongue and groove morphology of the postorbital-parietal suture in Oudenodon could have withstood the higher FE-predicted tensile strain in the posterior skull roof. The scarf premaxilla-nasal suture of Lystrosaurus provided an additional region of sutural mobility in the anterior surface of the snout, suggesting that Lystrosaurus may have employed a different biting regime than Oudenodon. The morphology of several sutures sampled in this study correlated with the FE-predicted strain, although other cranial functional hypotheses remain to be tested.

Comparative feeding biomechanics of Lystrosaurus and the generalized dicynodont Oudenodon.

Differences in cranial morphology among the Dicynodontia have been correlated with changes in masticatory function, and hence, dietary preference. Although the derived masticatory apparatus of dicynodonts allowed propaliny, it has previously been hypothesized that Lystrosaurus primarily utilized powerful orthal jaw movements to process fibrous vegetation. Cranial specializations of Lystrosaurus, such as shortened and deepened cranium and a mobile premaxilla-nasal suture, are thought to have increased the efficiency of its masticatory system compared with generalized Permian dicynodonts. Here we aim to test this assertion using biomechanical modeling techniques. We use finite element analysis (FEA) and a study of cranial functional morphology to compare the biomechanical performance of the crania of Lystrosaurus and Oudenodon, a generalized dicynodont, during orthal bite simulations. Muscle forces were estimated for each dicynodont using the dry skull method and applied to each cranium to produce a reaction force at a bite point. Patterns and average magnitude of Von Mises stress in each dicynodont cranium and in segmented regions of interest were assessed. During orthal bite simulations, higher stress occurs throughout the Oudenodon cranium, indicating that the cranium of Lystrosaurus is more resistant to normal, static feeding loads. Despite this difference in stress magnitude, patterns of stress are similar within both taxa. The FE-stress results, along with mechanical advantage of adductor musculature, a broad symphyseal contact, and other cranial features suggest that Lystrosaurus may have used a snapping bite to cope with tough fibrous vegetation.