Body mass estimation from footprint size in hominins.

Although many studies relating stature to foot length have been carried out, the relationship between foot size and body mass remains poorly understood. Here we investigate this relationship in 193 adult and 50 juvenile habitually unshod/minimally shod individuals from five different populations-Machiguenga, Daasanach, Pumé, Hadzabe, and Samoans-varying greatly in body size and shape. Body mass is highly correlated with foot size, and can be predicted from foot area (maximum length × breadth) in the combined sample with an average error of about 10%. However, comparisons among populations indicate that body shape, as represented by the body mass index (BMI), has a significant effect on foot size proportions, with higher BMI samples exhibiting relatively smaller feet. Thus, we also derive equations for estimating body mass from both foot size and BMI, with BMI in footprint samples taken as an average value for a taxon or population, estimated independently from skeletal remains. Techniques are also developed for estimating body mass in juveniles, who have relatively larger feet than adults, and for converting between foot and footprint size. Sample applications are given for five Pliocene through Holocene hominin footprint samples from Laetoli (Australopithecus afarensis), Ileret (probable Homo erectus), Happisburgh (possible Homo antecessor), Le Rozel (archaic Homo sapiens), and Barcin Höyük (H. sapiens). Body mass estimates for Homo footprint samples appear reasonable when compared to skeletal estimates for related samples. However, estimates for the Laetoli footprint sample using the new formulae appear to be too high when compared to skeletal estimates for A. afarensis. Based on the proportions of A.L. 288-1, this is apparently a result of relatively large feet in this taxon. A different method using a ratio between body mass and foot area in A.L. 288-1 provides estimates more concordant with skeletal estimates and should be used for A. afarensis.

Swift and sure-footed on the Savanna: A study of Hadzabe gaits and feet in Northern Tanzania.

Data on footprints and gait of 54 Hadzabe, 6-70 years of age expand understanding of pedal morphology of unshod people and assist the development of ideas about the evolution of hominid bipedality and upright posture. Contrary to published data on gaits and pedal morphology of unshod populations, Hadzabe (also known as Hadza) from northern Tanzania exhibit values of stride length, relative stride length, and walking speeds that exceed those of rural and small-town populations. In all observable plantar features, including foot indices, an interdigital space between the hallux and second toe, fanning of the foot anteriorly, and foot angles (in-toeing and out-toeing), Hadzabe feet are comparable with those of never-shod Machiguengas in Perú. On average, Hadzabe hallucal gaps and ball widths are narrower than those of Machiguengas and other unshod short people. Hadzabe feet are also characterized by valgus halluces versus the varus halluces of never-shod Machiguengas and certain West Africans. Although characterized by a valgus toe, Hadzabe hallucal angles, which do not exceed 20°, are lower than those of Northern Hemispheric urbanites and shod rural populations of the Southern and Northern Hemispheres. Hadzabe also exhibit less medial and lateral rotation of the hip joint than Machiguengas do. The heel and the longitudinal arch impressions of the Hadzabe footprints closely resemble those of the Laetoli bipeds in the manner of weight distribution during locomotion. The striking similarity of footprint impressions, especially the heel and the longitudinal arch, between Hadzabe and Laetoli hominid footprints clearly imply that the pedal features of the Laetoli printmakers are remarkably humanoid. Am. J. Hum. Biol. 9:303-321, 1997. © 1997 Wiley-Liss, Inc.