1. ^*, Department of Anatomy and ^†Department of Anthropology, Duke University, North Carolina 27710, USA
2. ^‡Museum of Paleontology, University of Michigan, Ann Arbor, Michigan 48109, USA.

Abstract

The anatomy of the posterior basicranium has been repeatedly invoked in systematic definitions of Primates. One widely cited definition^1,2 of the order claims that 'all undoubted primates'³ are distinguished from other mammals by two basicranial specializations: (1) absence of a major vascular foramen on the medial side of the auditory region, and (2) development of the auditory bulla from the petrosal bone. As we show here, specialization (1) does not apply to the paromomyid Ignacius, and is of uncertain incidence in other unquestioned members of suborder Plesiadapiformes (archaic primates from the early Cenozoic of Europe and North America). Specialization (2) cannot be demonstrated without ontogenetic evidence, and all relevant plesiadapiform fossils are adult. In fact, the only plesiadapiform with an arterial pattern remotely resembling that of early primates of modern aspect (or 'euprimates'⁴) is the microsyopid Cynodontomys, but it is often regarded as non-primate because it lacks a petrosal bulla. Although plesiadapiforms resemble euprimates in traits of the cheek teeth and postcranium^5–7, some other (presumably non-primate) groups possess these traits as well. Since the order Primates is not clearly definable by unique specializations, the best grounds for regarding plesiadapiforms as euprimate antecedents are stratigraphic and phenetic. This fact may be best expressed by systematic arrangements that emphasize adaptive grades rather than unsubstantiated clades.