Cercopithecoid cervical vertebral morphology and implications for the presence of Theropithecus in early Pleistocene...

While it is recognized that the overall configuration of the vertebral column, as well as the size and shape of individual vertebrae, differ within and between primate taxa, relatively little is known about the degree to which vertebral morphology reflects a phylogenetic signal or the degree to which vertebral elements can be used in accurate taxonomic classification. Isolated vertebrae are occasionally found in fossil assemblages, and proper taxonomic identification is necessary to make inferences about the animal's biology and place it in a broader phylogenetic and evolutionary context. Recently, three large primate cervical vertebrae (C3, C5, and C6) from Pirro Nord, Italy (early Pleistocene, late Villafranchian) were attributed to the genus Theropithecus based on size comparisons with extant cercopithecoid primates (Rook et al., 2004, J. Hum. Evol. 47, 267-277). These fossils were suggested to indicate an early dispersal of this genus out of Africa around 1.6-1.3 Ma possibly co-incident with early Pleistocene dispersals of Homo. Because of the potential importance of these fossils for interpreting Theropithecus evolution and the relatively few morphological data on primate cervical vertebral morphology, we examined the size and shape of cervical vertebrae in a large sample of extant cercopithecoid taxa (n=106). Specifically, we evaluated whether subfamily and genus level assignments can be made on the basis of isolated cervical elements. Discriminant analyses reveal that scaled shape variables are good discriminators of taxonomic affinity at the subfamily level but are poor discriminators at the genus level. Least-squares regressions show that raw linear dimensions of cervical vertebral morphology are good predictors of body mass in the extant sample. Our regression results produce a likely body mass estimate of 22-38 kg for the Pirro Nord cervical vertebrae. Based on these regression estimates, the poor ability to discriminate cervical vertebrae at the genus level, and paleoenvironmental reconstructions of Pirro Nord, it is unlikely that the Pirro Nord fossils can be confidently attributed to the genus Theropithecus. These findings have important implications for recent interpretations of the nature of Theropithecus dispersal out of Africa.