Meeting Abstract
Reconstructing non-mammalian synapsid posture usually takes one of two approaches. One approach models range of motion and muscle moment arms from fossils, and another method compares fossils to extant tetrapod analogues. The former method is constrained by how well body fossils reflect in vivo biomechanics, while the latter is limited by the relevance of extant analogues to fossil synapsids. Non-mammalian synapsids lack both extant representatives and a tight phylogenetic bracket. The only extant synapsids are mammals, which are highly derived and disparate compared to most fossil synapsids. Extant reptilian and amphibian analogues also possess derived morphologies compared to their extinct relatives, despite superficially “primitive” postures. Does the high phylogenetic and morphological disparity of these analogues impact their usefulness? We collected long bone and girdle dimensions from skeletal material for of 45 extant mammal, reptile, and amphibian taxa, and compared these data with analogous measurements from Diasparactus, Ophiacodon, Dimetrodon, and Aulacephalodon. We used principal component analysis to visualize the distribution of the sampled taxa in morphospace, and phylogenetic flexible discriminant analysis to explore categorical estimates of synapsid posture compared to traditional extant locomotor grades. Results indicate that fossil synapsids lack informative locomotor analogues among extant tetrapods. Furthermore, our results suggest that the traditional sprawling-upright dichotomy not only masks the continuum of tetrapod posture, but also ignores phylogenetic and physical constraints relevant to their locomotor habits.
