African Origin of Human Mitochondrial DNA Reexamined

Abstract

Few papers in evolutionary biology have created as much of a stir as did that of Cann et al. (1987), which claimed in part that Africa is a likely source of the human mitochondrial gene pool. Criticisms (Darlu and Tassy, 1987; Saito and Omoto, 1987; Pamilo and Nei, 1988; Spuhler, 1988; Excoffier and Langaney, 1989) and extensions (Wilson et al., 1987, 1989; Wrischnik et al., 1987; Vigilant et al., 1989; Horai and Hayasaka, 1990) of Cann et al.'s watershed paper have focused on the validity of the inferred phylogeny, but have not questioned whether Cann et al.'s phylogeny supports their inference of African origin. The purpose of the current paper is not to examine, in general, the question of human mitochondrial DNA (mtDNA) origin. I will focus on the evidence provided by Cann et al.'s data and their analytical methods. I will show that their data do not unequivocally support an African origin of human mtDNA, as there are equally parsimonious migration pathways that suggest different regions of origin. My main message will be a methodological one: in answering an evolutionary question using phylogenetic analysis, the full implications of data are not revealed unless one considers all equally well supported hypotheses. METHOD OF INFERRING ANCESTRAL REGION