Morphometric analysis of the distal humerus of some Cenozoic Catarrhines: the Late Divergence Hypothesis revisited.

Washburn's Late Divergence Hypothesis (LDH) makes a series of predictions about the phyletic affinities of extant hominoids and also predicts the locomotor behavior of the "formative ape," the common ancestor to the African apes and humans, and of the earliest hominids. The present study explores some of these predictions in the light of distal humerus morphology. Multivariate analysis of distal humerus metrics, corrected for the within-groups distortion of shape by size, was used to compare a broad sample of 22 modern anthropoid taxa with 15 fossils from the Fayum, Rusinga Island, Ft. Ternan, Neudorf an der March, Kanapoi, Kromdraai, Lake Turkana, and Hadar. The results of this analysis support some aspects of the LDH, while other aspects are unresolved. Specifically, the distal humeri of the large hominoids are very distinct from those of other anthropoids; however, Hylobates is not associated clearly with either hominoids or monkeys. This suggests that "brachiation" (sensu Hylobates) is too specialized a behavior and cannot explain the common set of characteristics known to unify the hominoids. Among the large apes, there is no discrimination that can be made between the African apes and Pongo, and Homo appears only slightly closer to Pongo than to the other apes. This offers little insight into the question of whether humans did, or did not, go through a "knuckle-walking" stage in their ancestry. The Oligocene and Miocene distal humeri sustain Washburn's assertion that the "formative apes" resembled the suspensory quadrupedal platyrrhines like Ateles. The Miocene P. africanus specimen highlights the uniqueness of Hylobates in showing that this fossil "bridges" the morphology of the acrobatic cebids and the morphology of the larger apes; it appears to be clearly "intermediate" between the two groups. Hylobates, by contrast, has affinities with no fossil, does not connect with the apes, and is generally isolated except for its connection with colobines. Among the Plio-Pleistocene fossils, the Hadar sample proves to be quite primitive, and may be close to the point where hominids and pongids diverged. The Kanapoi distal humerus (KP 271), far from being more "human-like" than Australopithecus, clearly associates with the hyperrobust Australopithecines from Lake Turkana. The stratigraphically late Kromdraai distal humerus is the only hominid to be projected near Homo. The Plio-Pleistocene hominids generally evince a pattern consistent with the prediction of a late separation of hominids and pongids. But, the results of this study, like many before it, fail to resolve the central question of whether the last common ancestor of the African apes and humans "knuckle-walked." It is suggested here that this may be an unresolvable problem.