Body mass estimation in hominins from humeral articular dimensions.

OBJECTIVES: While many attempts have been made to estimate body mass in hominins from lower limb bone dimensions, the upper limb has received far less attention in this regard. Here we develop new body mass estimation equations based on humeral articular breadths in a large modern human sample and apply them to 95 Plio-Pleistocene specimens. MATERIALS AND METHODS: Humeral head superoinferior and total distal articular mediolateral breadths were measured in a morphologically diverse sample of 611 modern human skeletons whose body masses were estimated from bi-iliac breadth and reconstructed stature. Reduced major axis regressions were used to compute body mass estimation equations. Consistency of the resulting estimates with those derived previously using lower limb bone equations was assessed in matched Plio-Pleistocene individuals or samples. RESULTS: In the modern reference sample, the new humeral body mass estimation equations exhibit only slightly lower precision compared to the previously derived lower limb bone equations. They give generally similar estimates for Pleistocene Homo, after accounting for the different shape of the humeral head articular surface in archaic Middle and Late Pleistocene Homo, except for distal humeral estimates for Late Pleistocene specimens, which average somewhat below lower limb estimates. Humeral equations give body mass estimates for australopiths that appear much too high, except for Australopithecus sediba. A chimpanzee-based distal humeral articular formula appears to work well for larger australopith specimens. DISCUSSION: The new formulae provide a more secure foundation for estimating hominin body mass from humeri than previously available equations.

Lower limb articular scaling and body mass estimation in Pliocene and Pleistocene hominins.

Previous attempts to estimate body mass in pre-Holocene hominins have relied on prediction equations derived from relatively limited extant samples. Here we derive new equations to predict body mass from femoral head breadth and proximal tibial plateau breadth based on a large and diverse sample of modern humans (avoiding the problems associated with using diaphyseal dimensions and/or cadaveric reference samples). In addition, an adjustment for the relatively small femoral heads of non-Homo taxa is developed based on observed differences in hip to knee joint scaling. Body mass is then estimated for 214 terminal Miocene through Pleistocene hominin specimens. Mean body masses for non-Homo taxa range between 39 and 49 kg (39-45 kg if sex-specific means are averaged), with no consistent temporal trend (6-1.85 Ma). Mean body mass increases in early Homo (2.04-1.77 Ma) to 55-59 kg, and then again dramatically in Homo erectus and later archaic middle Pleistocene Homo, to about 70 kg. The same average body mass is maintained in late Pleistocene archaic Homo and early anatomically modern humans through the early/middle Upper Paleolithic (0.024 Ma), only declining in the late Upper Paleolithic, with regional variation. Sexual dimorphism in body mass is greatest in Australopithecus afarensis (log[male/female] = 1.54), declines in Australopithecus africanus and Paranthropus robustus (log ratio 1.36), and then again in early Homo and middle and late Pleistocene archaic Homo (log ratio 1.20-1.27), although it remains somewhat elevated above that of living and middle/late Pleistocene anatomically modern humans (log ratio about 1.15).

Clavicular curvature and locomotion in anthropoid primates: A 3D geometric morphometric analysis.

OBJECTIVES: As a component of the primate shoulder, the clavicle is expected to reflect locomotor adaptations. Whereas previous work has generally focused on clavicular length and torsion, the shape of clavicular curvature may better distinguish taxa and provide additional information about upper limb use in locomotion. This study uses three-dimensional geometric morphometrics to analyze shape differences in the curvatures of the clavicle in different locomotor groups of anthropoid primates. METHODS: Sliding semi-landmarks were placed on clavicles of 10 Anthropoid primate species (total n = 85) that display a range of locomotor behaviors. Landmarks (k = 39) were chosen to capture the overall curvature of the clavicle in three dimensions. RESULTS: The degree of ventral curvature in the clavicle represents a gradient from most-curved in suspensory genera (e.g., Ateles, Hylobates, and Pongo) to least-curved in genera that are rarely suspensory (e.g., Papio and Gorilla). This curvature may allow an increased range of craniodorsal movement without the clavicle impinging on the thoracic outlet. An inferior curvature of the medial clavicle is found in hominoids and brachiators. This curvature could help stabilize the shoulder and prevent superior dislocation of the clavicle in suspension. Finally, a superior curvature in the lateral part of the clavicle, most pronounced in quadrupedal monkeys, may be related to the relative position of the scapula and sternum. CONCLUSIONS: Patterns of clavicular curvature in anthropoid primates reflect locomotor behavior and successfully distinguished among taxonomic and locomotor groups. In the future, this method could be used to assess locomotor behavior in fossil primates. Am J Phys Anthropol 158:257-268, 2015. © 2015 Wiley Periodicals, Inc.

Body mass estimation from knee breadth, with application to early hominins.

OBJECTIVES: The estimation of living body mass from skeletal dimensions is an important component of many studies of early hominins and more recent human archaeological remains. Most previous investigations have concentrated on weight-bearing elements of the lower limb, in particular the femoral head. In this study, we develop new body mass estimation equations derived from measurements of the knee in a modern sample of known body mass, and use them to estimate body mass in 11 fossil hominin specimens (including Au. africanus, Au. afarensis, and early Homo). MATERIALS AND METHODS: The reference sample consisted of 100 living subjects who participated in the Baltimore Longitudinal Study of Aging. Mediolateral breadth measurements were taken from radiographs of the knee, and regressed against recorded body weight to generate body mass estimation equations. RESULTS: Knee dimensions were generally found to be good predictors of body mass in the modern human sample, with median absolute percent prediction errors of 7 to 9% (comparable to or better than previously reported equations derived from the femoral head). Taxon-average estimated body masses were 46.1 kg for Au. afarensis, 38.4 kg for Au. africanus, and 53.6 kg for early Homo. DISCUSSION: Estimates for early Homo were similar to or smaller than those generated previously from the femoral head. Estimates for australopiths, however, were larger than those generated from femoral head equations. This result is consistent with other evidence that the femoral head was relatively unloaded in australopiths compared with Homo, possibly due to subtle differences in gait. Am J Phys Anthropol 158:198-208, 2015. © 2015 Wiley Periodicals, Inc.