The pattern of hominin postcranial evolution reconsidered in light of size-related shape variation of the distal humerus.

Previous research suggests that some hominin postcranial features do not follow a linear path of increasing modernization through geological time. With respect to the distal humerus, in particular, the earliest known hominin specimens are reportedly among the most modern in morphology, while some later humeri appear further removed from the average modern human shape. Although Plio-Pleistocene humeri vary widely in size, previous studies have failed to account for size-related shape variation when making morphometric comparisons. This study reexamines hominin postcranial evolution in light of distal humeral allometry. Using two-dimensional landmark data, the relationship between specimen size and shape among modern humans is quantified using multivariate regression and principal components analysis of size-shape space. Fossils are compared with modern human shapes expected at a given size, as well as with the overall average human shape. The null hypothesis of humeral isometry in modern humans is rejected. Subsequently, if one takes allometry into account, the apparent pattern of hominin humeral evolution does not resemble the pattern described above. All 14 of the Plio-Pleistocene hominin fossils examined here share a similar pattern of shape differences from equivalently-sized modern humans, though they vary in the extent to which these differences are expressed. The oldest specimen in the sample (KNM-KP 271; Australopithecus anamensis) exhibits the least human-like elbow morphology. Similarly primitive morphology characterizes all younger species of Australopithecus as well as Paranthropus robustus. After 2 Ma, a subtly more human-like elbow morphology is apparent among specimens attributed to early Homo, as well as among isolated specimens that may represent either Homo or Paranthropus boisei. This study emphasizes the need to consider size-related shape variation when individual fossil specimens are compared with the average shape of a comparative group, particularly when specimens fall near an extreme of the comparative size distribution.