The fossil record stays silent: Confusions and conundrums for hominin pelvis evolution

The evolution of the hominin pelvis is commonly modeled as a series of stages driven largely by the requirements of bipedal locomotion, reproduction, thermoregulation, and pelvic floor muscular support. These patterns are complicated by variation in canal dimensions in relationship with different changes in overall pelvic breadths. To quantify this complexity, we examine pelvic canal dimensions relative to body mass and bi-iliac and biacetabular breadths in a global human skeletal sample (N = 229) and 10 hominin fossils. Relative to measures of pelvic breadth and body mass, there is substantial variation in canal dimensions in hominins and both sexes of Homo sapiens, with weaker relationships with anteroposterior than mediolateral dimensions. Human females are more variable in anteroposterior dimensions than males, but not in mediolateral dimensions. While anteroposterior dimensions are relatively short in most Australopiths, canal breadths are not relatively larger than Homo (except vs. body mass). Mediolateral reduction and anteroposterior expansion are not consistent across Homo, nor within Homo sapiens. Due to substantial variation within both sexes of Homo sapiens, hominin fossils, and the reconstructions within a single assemblage, clearly no one (or 2) pelves can stand in for pelvic form for a hominin grade (e.g., Australopiths, early Homo, and so forth), or perhaps even species. Pelvic shape and "dimorphism" evolution within the hominin lineage did not follow a single trajectory. Given the dearth of Middle Pleistocene Homo pelves from warmer climates (lower latitude), and the variation produced by different pelvic reconstructions, our picture of pelvic form in later Homo remains incomplete.