2D Versus 3D Shape Signals of Climatic Adaptations in the Trunk Skeleton of Recent Humans

The large range of phenotypic variation observable in recent human populations arguably stems in part from long‐term adaptation to local environmental conditions. As a globally distributed species, humans are – and have been – subject to diverse ecogeographical selective pressures, and numerous studies support the idea that humans generally follow body size and shape predictions of Allen's and Bergmann's rules. However, these studies focus on a small number of discrete linear dimensions rather than considering 3D aspects of morphological shape. Here I explore the strength of the relationships between climatic variables and the trunk skeleton to determine how these selective pressures influence bony morphology in recent humans. I collected 2D linear metrics and 3D landmark data of torso skeletal elements (ribs, sterna, vertebrae, pelves) from ten skeletal populations (n=586), which include individuals from equatorial, temperate, and arctic geographical zones, low and high altitude, and the Old and New Worlds. Genetic reference data from populations in the Human Genome Diversity Project‐CEPH were also used to investigate the degree to which trunk shape tracks neutral genetic evolution and the degree to which population history affects the observed relationships between morphology and climate. Partial correlation analyses conditioned on femoral head breadth (a common body size proxy) were used to evaluate the bivariate relationships between climatic and skeletal variables independent of body size. Multivariate methods included Two‐Block Partial Least Squares and Mantel tests of matrix similarity to test associations among the climatic, skeletal, and genetic datasets. Results demonstrate that in the human trunk 2D linear dimensions of skeletal shape exhibit far stronger relationships with climate than do overall 3D aspects of skeletal shape. Temperature and humidity climatic variables are strongly correlated with some linear dimensions of trunk shape, such as bi‐iliac breadth, rib‐tubercle chord length, iliac blade height, and pubis length. Correlations of numerous shape variables with mean annual temperature and mean temperature of the coldest month to the exclusion of mean temperature of the warmest month suggest the relatively greater importance of colder temperatures in morphological adaptation to climate. Altitude appears to constitute a competitive selective pressure to temperature on some rib dimensions but does not appear to strongly influence pelvic shape. Contrary to expectations, 3D shape of torso skeletal elements is extremely weakly correlated with all climatic variables. The pelvis significantly tracks population history but the ribcage does not, indicating a significant role of neutral genetic evolution in shaping patterns of pelvic morphology observed among global human populations. Changing only certain trunk skeletal dimensions in response to ecogeographic pressures may confer selective benefits in terms of thermoregulatory needs, while potentially leaving more flexibility in the rest of the bony frame to respond to selective pressures such as locomotion and obstetrics. Support or Funding Information Grant support for data collection was provided by the Wenner‐Gren Foundation, the Leakey Foundation, and New York University.