Phylogeny and classification of hominoidea as inferred from DNA sequence data.

Abstract

It is now clear that the molecular record rather than the fossil one is a more suitable indicator in clarifying the phylogenetic relationship among extant species and in dating the branching events among them (e.g., refs. 1) and 2) ) . In a previous paper in this Proceedings, j' we developed a statistical method to date the splittings in the evolution of Hominoidea by using mitochondrial DNA (mtDNA) sequences. Our estimate was dependent on the assumption that the divergence between primates and ungulates occurred 90 million years (Myr) ago. However, we now know that 65 Myr is much more reasonable for this separation, because the mass extinction of dinosaurs occurred at this age4~ and because there is no fossil evidence that indicates an earlier date.' A molecular clock of mtDNA calibrated by this revised reference time gave dates of 13.7±1.4, 11.6±1.2, 3.6±0.5, and 2.5±0.4 Myr ago (:± refers to 1SD) for the separation of gibbon, orangutan, gorilla, and chimpanzee, respectively, from the line leading to man.e) Unless interspecific transfer of mtDNA7~ occurred during the evolution of early hominids, this dating rejects a widely believed hypothesis that Australopithecus a f arensis that lived same 3.7 Myr ago is ancestral to man after the human-ape splitting. It is known that the introgression occurred between mice8~ and between frogs9> 1-1.5 Myr after speciation. If it happened in the early hominids, then the human-ape splitting may go back to about 4-5 Myr ago.° Future sequencing of nuclear DNA should make the situation clear. Such a close relationship between man and chimpanzee casts a serious problem on traditional classifications of Hominoidea. The widely accepted classifications are shown in Fig. 1, E and F. In these schema, man is separated from all the extant apes and is placed in the family Hominidae as its only living member. A closer phyletic relationship of chimpanzee and gorilla to man rather than to orangutan has been pointed out from molecular evidence by Zuckerkandl