First skulls of the Early Eocene primate Shoshonius cooperi and the anthropoid-tarsier dichotomy

Abstract

THE phylogenetic relationships of living tarsiers and extinct omomyid primates are critical for deciphering the origin and relationships of primate higher taxa, particularly anthropoids^1–6. Three competing phylogenetic hypotheses are: (1) tarsiers are most closely related to early Cenozoic Omomyidae^5–8, particularly genera such as Necrolemur from the late Eocene of Europe^9–11; (2) tarsiers share a more recent common ancestry with anthropoids than they do with any known omomyid^2–4,12,13; (3) tarsiers and/or omomyids are most closely related to strepsirhines¹⁴. The anatomy of four skulls of the early Eocene omomyid Shoshonius cooperi — the first cranial material recovered for this genus—strongly suggests that Shoshonius shares a more recent common ancestry with Tarsius than do either anthropoids or other Eocene omomyids for which cranial anatomy is known. If the primate suborder Haplor-hini (anthropoids, omomyids, tarsiids) is monophyletic, the phylogenetic position of Shoshonius requires that anthropoids and Tarsius diverged by at least the early Eocene, some 15 million years before the first appearance of anthropoids in the fossil record^15–17.