Humans as second orangutans: sense or nonsense?

Mark Stoneking,y BioEssays Editorial Board member Extraordinary claims require extraordinary evidence – or so I was taught as a student. But it sometimes appears as if papers making sensational claims can get published with rather less than compelling evidence, especially if said claims are likely to result in attention from the popular press. An apparent example comes from the Journal of Biogeography, with the recent publication of a paper making the astonishing claim that humans are actually more closely related to orangutans than, as (nearly) universally accepted, to chimpanzees. The authors of this study carried out phylogenetic analyses of selected morphological and fossil characters and found a closer relationship of humans with orangutans, rather than with chimpanzees. I do not feel qualified to comment on the adequacy of their analyses, except to note that previous analyses of fossil characters have followed the genetic evidence in grouping humans with chimpanzees, so presumably this new result will be a bone of contention amongst paleoanthropologists. But how do the authors reconcile this result with molecular genetic evidence, going back 40 years now, that unequivocally indicate that African apes (specifically, chimpanzees) are our nearest living relatives? The authors identify what they consider to be four problematic aspects of the molecular genetic evidence. First, they claim that homoplasy (i.e., parallel changes in independent lineages, or reversal within a lineage) and alignment problems (in particular, how to account for insertions and deletions necessary to align sequences of different lengths) can render phylogenetic analyses of DNA sequence data inaccurate. But no specific examples are given as to how homoplasy and/or alignment problems could possibly have resulted in an erroneous conclusion of a human-chimpanzee relationship. Moreover, neither homoplasy nor alignment issues are expected to have any impact over the time scales in question; in particular, there is no ambiguity in aligning coding sequences among the species in question, and phylogenetic analyses of both mtDNA and nuclear DNA coding sequences unequivocally support the human-chimpanzee relationship. So, this is a red herring. Second, the authors criticize molecular genetic analyses of human-chimpanzee relationships for having ‘‘extremely limited’’ outgroup sampling. In fact, the studies cited by the authors would be regarded by most as having more than