Body weight prediction in early fossil hominids: Towards a taxon‐“independent” approach

The choice of a model taxon is crucial when investigating fossil hominids that clearly do not resemble any extant species (such as Australopithecus) or show significant differences from modern human proportions (such as Homo habilis OH 62). An "interhominoid" combination is not adequate either, as scaling with body weight is strongly divergent in African apes and humans for most skeletal predictors investigated here. Therefore, in relation to a study of seven long bone dimensions, a new taxon-"independent" approach is suggested. For a given predictor, its taxonomic "independence" is restricted to the size range over which the body weight-predictor relationship for African apes and humans converges. Different predictors produce converging body weight estimates (BWEs) for different size ranges: taxon-"independent" estimates can be calculated for small- and medium-sized hominids (e.g., for weights below 50 kg) using femoral and tibial dimensions, whereas upper limb bones provide converging results for large hominids (above 50 kg). If the remains of Australopithecus afarensis really belong to one species, the relationship of male (above 60 kg) to female body weight (approximately 30 kg) does not fall within the observed range of modern hominoids. Considering Sts 14 (22 kg) to represent a small-sized Australopithecus africanus, the level of encephalization lies well above that of extant apes. If OH 62 (approximately 25 kg), with limb proportions less human-like than those of australopithecines, indeed represents Homo habilis (which has been questioned previously), an increase in relative brain size would have occurred well before full bipedality, an assumption running counter to current assumptions concerning early human evolution.