Gracility of the modern Homo sapiens skeleton is the result of decreased biomechanical loading

Species mean and SD values for each measured trabecular bone variable are listed in Table S1 and definitions for each trabecular bone variable are provided in Table S2. There are no significant differences between the human populations for femoral head superoinferior height or estimated body mass (Table S3). Despite having similar body sizes, visual inspection of the reconstructed volumes of interest (VOI) and 2D cross-sections suggest structural variation in femoral head trabecular bone between these agriculturalist (Norris Farms, Dickson Mounds) and forager (Black Earth, Modoc Rock Shelter) populations from eastern North America (Illinois) (Fig. 1). Scatter plots of individual trabecular bone variables vs. estimated body mass reveal that the foragers and agriculturalists separate from each other in three important structural features: bone volume fraction (BV/TV), trabecular thickness (Tb.Th), and bone surface area relative to bone volume (BS/BV) (Fig. 2). The mean BV/TV for the agriculturalist groups both plot below the lower-bound 95% confidence interval for the nonhuman primate regression line, revealing that these groups have the lowest BV/TV relative to body mass of all primates, aside from Tarsius. In contrast, BV/TV of the forager sample falls within the 95% confidence interval for the nonhuman primates, plotting on top of Pongo. Tb.Th values for the agricultural sample lie well below the nonhuman primate regression line and outside of the 95% confidence interval, indicating that the trabeculae of agriculturalists are relatively thin compared with a primate of similar body mass. In contrast, the Tb.Th of the forager sample falls within the nonhuman primate 95% confidence interval. As a consequence of their lower BV/TV and thick trabeculae, the agriculturalists have relatively high BS/BV. Trabecular separation (Tb.Sp) for all human groups falls just above and outside of the nonhuman primate 95% confidence interval. Both foragers and agriculturalists also have lower trabecular number (Tb.N) relative to body mass compared with other hominoids, plotting just outside of the nonhuman primate 95% confidence interval. All human groups have relatively anisotropic trabecular bone structure.

The ANOVA comparisons for unstandardized residuals substantiate the scatter plot results, revealing significant structural differences between the foragers and agriculturalists, and also among the agricultural populations and most other hominoids (Table 1). No significant differences were found between the Norris Farms agriculturalists and Dickson Mounds agriculturalists for any variable, and are henceforth collectively referred to as “agriculturalists,” unless otherwise specified. Compared with the foragers, agriculturalists display significantly lower mass-corrected BV/TV and Tb.Th, and significantly higher BS/BV. There are no significant differences between the agriculturalist and forager populations for Tb.Sp, Tb.N, or degree of anisotropy (DA). The agriculturalists have significantly lower BV/TV than all other extant hominoids, except Gorilla. In contrast, BV/TV of the forager H. sapiens is comparable to all hominoid groups, other than Pan, a taxon with particularly high femoral BV/TV (see, for example, ref. 18). The Tb.Th values in the forager and agriculturalist humans do not differ from any of the extant hominoids. Both agriculturalists and foragers have significantly higher Tb.Sp and lower Tb.N than Pan. The foragers and Dickson Mounds agriculturalists also have significantly fewer trabeculae relative to their body mass compared with Gorilla (Norris Farms agriculturalists approach significance: P = 0.056). The Norris Farms agriculturalists also have significantly higher BS/BV compared with Pan. Comparisons of DA in humans and other hominoids indicate that both agriculturalists and foragers have significantly more anisotropic trabeculae than all other hominoids; the one exception is a lack of difference between foragers and Pongo (P = 0.069).

The findings of the present study have significant implications for understanding human skeletal form and its relationship to age-related bone loss in contemporary human populations. Osteoporosis is a systemic bone disease with significant global health impacts that is characterized by low bone mass and microarchitectural deterioration, leading to increased bone fragility and fracture risk (36). Although both cortical and trabecular bone, together, determine bone strength, trabecular architecture is particularly susceptible to age-related loss because of the extensive exposure of its metabolically active surfaces and the relatively thin plates and struts (37–40). Because of the documented gracilization of the human skeleton (2) and the diversity of subsistence strategies performed in the past, the hominin fossil and skeletal record is ideal for addressing questions aimed at understanding contemporary human bone biology and health.

The results of the present study indicate that human populations with divergent activity patterns display significantly different trabecular bone structural characteristics in the proximal femur. Mobile foragers possess significantly higher BV/TV and significantly thicker trabeculae compared with more sedentary agriculturalists. The trabecular BV/TV, and consequently the relative stiffness, of the forager femoral head are comparable to most other hominoids. These results, therefore, reveal that more highly mobile human populations have trabecular bone structure in the femoral head that is similar to what would be expected for a nonhuman primate of the same body size. The comparatively gracile trabecular bone of the more sedentary village agriculturalists is composed of significantly thinner trabeculae, resulting in a much higher BS/BV ratio compared with the foragers. These results strongly suggest that BV/TV, Tb.Th, and consequently, BS/BV ratio in the human femoral head are inherently plastic traits influenced by the loading environment. In contrast, the finding of similar trabecular number and spacing in the femoral head of both agriculturalists and foragers, despite divergent loading patterns, suggests a higher level of canalization for these traits.

The three main hypotheses proposed to explain the relatively gracile skeleton of contemporary humans include: (i) a reduction in physical activity because of increased sedentism and reliance on culture (1–6, 41), (ii) selection for systemic reduction of bone mass in modern humans (14, 16), and (iii) attenuation of strain imposed upon trabecular bone because of larger joint surface areas (12, 13). The morphological differences between the highly mobile foragers and relatively sedentary village agriculturalists clearly point to physical activity as a major determinant of bone mass in the hip joint. The nonhuman primate-like BV/TV in the highly mobile foragers strongly suggests that modern humans are neither systemically gracile nor morphologically constrained by their bipedal anatomy.

Although ancestry and sex are osteoporotic risk factors (42), the influence of these variables have been considered; similar numbers of male and female humans are included in this study, and all individuals derive from geographically proximate archaeological sites in eastern North America (Materials and Methods). It is also possible that diet plays a role in the observed differences between these populations. The dietary shift toward cultivated grains is associated with reduced calcium intake and net calcium deficiencies in some populations reliant on agriculture (cf. 43, 44). The archaeological record provides insight into possible dietary differences between the forager and agriculturalist groups included in this study. The Dickson Mounds and Norris Farms populations appear to have engaged in a mixed-subsistence strategy that included hunting, fishing, and gathering of wild-plant resources in addition to the active cultivation of maize and other plants (45). Isotopic data, however, suggest that the Norris Farms site may have relied less on maize than did Dickson Mounds (45). Analyses of the forager’s subsistence strategies suggest a similarly diverse diet that included fauna, such as large and small mammals, birds, and fish, as well as a diversity of seasonally available plant resources (46, 47). Floral and faunal remains from the Black Earth site indicate a highly mobile population with a large and diverse home-range, used for both seasonal and year-round foraging activities (46). Although dietary differences can contribute to a reduction in skeletal mass among agriculturalists relative to foragers, evidence from other localities (48, 49) strongly implicates activity, rather than diet, as the most important factor influencing skeletal strength and robusticity (the strength or mass of a bone in relation to body size) (41). The results, therefore, further corroborate the suggestion that exercise, perhaps most importantly activity performed during the adolescent growth period (50), is likely to be the best approach for the prevention of bone fragility and heightened fracture risk.

The results of the present study also provide insight into locomotor loading at the hip joint of hominoids. In contrast to the nonhuman apes, all four human populations display a highly anisotropic trabecular structure in the proximal femur. It seems likely that this human-specific trabecular construction relates to the stereotypical loading at the hip joint imposed during bipedal locomotion. In contrast, the proximal femur of quadrupedal and climbing apes is likely to experience more diverse postural and locomotor loads (51). This morphological consistency in humans may provide a tool for reconstructing locomotor behavior in the hominin fossil record.

The findings of the present study are relevant to various fields and inform debates involving the evolution of the human skeleton, bone adaptation, the reconstruction of past activity patterns, and the importance of skeletal loading and exercise for bone health in contemporary human populations. To our knowledge, this study provides the first clear evidence for a strong relationship between patterns of terrestrial mobility and trabecular bone architecture in the hip joint of modern humans. Increased loading magnitude and frequency in mobile human foragers led to relatively robust trabecular bone structure comparable to that seen in other primate species (52). With the emergence of Homo erectus ∼1.8 million y ago, paleontological and archaeological evidence indicate a major shift in hominin evolution characterized by increases in brain and body size (53), improved locomotor efficiency (54), larger home ranges, and increased mobility (55). This long evolutionary history of high levels of terrestrial mobility in the genus Homo, together with the evidence of selection for endurance running (56, 57), strongly suggests that the human locomotor skeleton evolved in a mechanical and physiological context that involved persistent and frequent loading throughout life. This study provides support for the assertion that contemporary humans live in a cultural and technological milieu incompatible with our evolutionary adaptations (58), and highlights the importance of physical activity and exercise for bone health and the attenuation of age-related bone loss, osteoporosis, and fracture risk.

We thank Nina Jablonski and George Milner for helpful suggestions during this project; Kristian Carlson for comments on a draft manuscript; the curators from the museums that loaned primate material; Terrance Martin, Dawn Cobb, and Dee Ann Watt at the Illinois State Museum and Heather Lapham at the Center for Archaeological Investigations, Southern Illinois University Carbondale, for kindly providing access to human skeletal material; Neil Sharkey and Noriaki Okita for providing access to bone analysis software; and Ben Auerbach for providing age estimates for the Dickson Mounds skeletal remains. This research was supported by National Science Foundation Grant BCS-0617097 (to T.M.R.).