The Oreopithecus’ thumb: a strange case in hominoid evolution

‘‘And now, Doctor, perhaps you would kindly attend to my thumb, or rather to the place where my thumb used to be.’’ (Sir Arthur Conan Doyle, 1892). We (Moyà-Solà et al., 1999) proposed the hypothesis that the hand of Oreopithecus bambolii (Gervais, 1872), an endemic, insular ape from the late Miocene (7e10 Ma) of Tuscany and Sardinia (Italy), does not fit the patternof extant apesand fossilhominoids ofmodern aspect (e.g., Dryopithecus). Instead, we argued that Oreopithecus exhibits a series of features that reflect an improved grasping capability, including firm pad-to-pad gripping,which apes are unable to perform. In these aspects, the Oreopithecus hand approximates the pattern of early hominids, presumably as a response to similar functional demands. Susman (2004) presented a critical appraisal of this work in which he attempted to demonstrate that Oreopithecus had a standard ape hand anatomy adapted to power grasping, which presumably would serve locomotor requirements in suspensory behaviors. His arguments were mainly based on the assumption that we made a set of critical errors of identification that invalidate our conclusions. According to Susman, the unusual characters we described in Oreopithecus do not exist, some because of misidentification and/or errors of measurement (i.e., a long thumb), others because of misinterpretation of accidental morphology and/or incompleteness of the specimens (i.e., pit for the flexor pollicis longus). The main discrepancy involves the identification of the thumb elements and morphology. In this reply, we focus on these aspects and demonstrate that our original conclusions on the hand anatomy of Oreopithecus remain valid.