'Australopithecus sediba' has been hypothesized to be a close relative of the genus 'Homo'. Here we show that MH1, the type specimen of 'A. sediba', was not optimized to produce high molar bite force and appears to have been limited in its ability to consume foods that were mechanically challenging to eat. Dental microwear data have previously been interpreted as indicating that 'A. sediba' consumed hard foods, so our findings illustrate that mechanical data are essential if one aims to reconstruct a relatively complete picture of feeding adaptations in extinct hominins. An implication of our study is that the key to understanding the origin of 'Homo' lies in understanding how environmental changes disrupted gracile australopith niches. Resulting selection pressures led to changes in diet and dietary adaption that set the stage for the emergence of our genus.

The craniofacial skeleton is often described in the clinical literature as being comprised of vertical bony pillars, which transmit forces from the toothrow to the neurocranium as axial compressive stresses, reinforced transversely by buttresses. Here, we review the literature on bony microarchitecture, 'in vivo' bone strain, and finite-element modeling of the facial skeleton of humans and nonhuman primates to address questions regarding the structural and functional existence of facial pillars and buttresses. Available bone material properties data do not support the existence of pillars and buttresses in humans or 'Sapajus apella'. Deformation regimes in the zygomatic complex emphasize bending and shear, therefore conceptualizing the zygomatic complex of humans or nonhuman primates as a pillar obscures its patterns of stress, strain, and deformation. Human fossil relatives and chimpanzees exhibit strain regimes corroborating the existence of a canine-frontal pillar, but the notion of a zygomatic pillar has no support. The emerging consensus on patterns of strain and deformation in finite element models (FEMs) of the human facial skeleton corroborates hypotheses in the clinical literature regarding zygomatic complex function, and provide new insights into patterns of failure of titanium and resorbable plates in experimental studies. It is suggested that the "pillar and buttress" model of human craniofacial skeleton function be replaced with FEMs that more accurately and precisely represent in vivo function, and which can serve as the basis for future research into implants used in restoration of occlusal function and fracture repair.