Is the bonobo growth trajectory the ancestral one for the Hominoidea?

We collected a total of 96 traditional landmarks and semilandmarks on the face and cranial base of 268 adult and sub-adult crania for a geometric morphometric analysis of five different hominoid species--Homo sapiens, Pan paniscus, Pan troglodytes, Gorilla gorilla and Pongo pygmaeus. The standard relative warp (RW) morphospace was sheared so that pooled within-group size allometry, the axis assumed to express the 'common' ontogenetic shape change, lies horizontally. In the resulting analysis there are differences in size-adjusted mean forms--these appear in sheared RW's 2,3 ...--and there are also differences in the within-taxon ontogenetic trajectories (which appear as within-group correlations of RW's that must, per definition, be uncorrelated in the pool). When visualized, both correspond to regionalized shape differences. The shape differences associated with the second sheared RW, which separates Homo from Pongo and the African apes, are already manifest in the youngest forms we have. These ape ontogenies seem parallel, but the vector for H. sapiens is different (P=0.002). Shape features of the third component, which distinguishes among the African apes, develop during postnatal ontogeny, and thus express actual divergence of ontogenies. Assuming that Hominoidea is monophyletic, the average growth trajectory is a reasonable estimate of the ancestral ontogeny. This trajectory is closest to those observed in our two samples of chimps, particularly to that for bonobo. In the second and third dimensions, furthermore, the bonobos lie nearly at the grand mean. We suggest therefore that bonobo is a particularly good model species for speculations about hominoid ancestral ontogeny. Research supported by the Austrian Science Foundation Project P14738 and a Ph.D. grant of the University of Vienna.