Perspectives in Biology and Medicine

GENE FREQUENCIES, DNA SEQUENCES, AND HUMAN ORIGINS HENRY HARPENDING* Genetics has become an important source of information about the origins of anatomically modern humans. In particular, a famous paper by R. Cann, M. Stoneking, and A. Wilson [1] examined mitochondrial DNA sequences from a world sample of humans. They suggested that the common ancestress of these mitochondrial sequences lived about 250,000 years ago and that this recent common ancestress supported a view that modern humans had a relatively recent origin in Africa. Their conclusions seemed to provide strong support to one side of an acrid debate in human paleontology, and the commentary that grew up around the paper, in particular the idea that the common mitochondrial ancestress was "Eve," brought the controversies about human origins and the role of genetics in human prehistory into the popular and semipopular press. Two entirely different kinds of genetic data and theory enter into the debate about human origins. One uses the frequency of genetic markers in different populations to reconstruct the branching history ofthe populations , while the other uses mitochondrial DNA sequences to reconstruct the branching history of a particular molecule. In this essay I will discuss the current state of both approaches after describing, in outline form, the context of current controversies about human origins. Since any inference in science is only as good as the underlying model on which it is based, I will pay particular attention to the assumptions proponents of each approach must make. My conclusions will be that the fundamental assumption of genetic marker studies is almost certainly violated so that they do not, in fact, tell us much about the history of our species. DNA sequences, on the other hand, do provide important information that is relevant to the issues, but the real implications of *Department of Anthropology Pennsylvania State University, University Park, Pennsylvania 16802.© 1994 by The University of Chicago. AU rights reserved. 003 1-5982/94/3703-0869$0 1 .00 384 Henry Harpending ¦ Human Origins DNA sequence data have been missed almost completely by commentators and textbook writers. Theories ofModern Human Origins Creatures that are unambiguously members of our own species appear in the fossil record approximately 40,000 years ago in Europe, while creatures that are ambiguously our own species appear in the Levant and in Africa approximately 100,000 years ago. Older precursors are called Homo erectus, while later precursors are usually called archaic Homo sapiens or (especially in Europe and the Near East) Neanderthals . Collectively, these precursors are found in Africa, Europe, Asia, and Indonesia. They were a successful group, having persisted without much apparent change for a million years or so. Current nomenclature places the major taxonomic difference between early precursors, Homo erectus, and later precursors, archaic Homo sapiens. Many anthropologists regret this usage, wishing instead that the later precursors were assigned to erectus and that sapiens were reserved for unambiguous anatomically modern humans. I will avoid all such semantic issues by using "precursors " and "moderns" for ancestral and modern human respectively. What was the process by which precursors in the Old World were replaced by moderns? According to one model, the worldwide population of precursors evolved modern characteristics in situ so that there was regional continuity of populations [2]. The genetic basis of modern appearance and behavior spread by mate exchange among the widely separated precursor populations, but the level of gene flow was low enough to preserve distinctive anatomical attributes that are shared between precursor and modern populations in the same regions. For example , "shovel-shaped incisors" are found in both Asian precursors and moderns. According to this model, then, the famous Homo erectus from China, Peking Man, may have descendants today, and these descendants are in Asia or in the New World. Other labels for this scenario are the candelabra model, the Weidenreich hypothesis, and multiregional evolution [3]. While various versions of this model differ in the supposed magnitude of gene flow among regional precursor populations, all agree that it was enough to ensure that genetic material determining modern traits could spread, but that it was not enough to swamp regional peculiarities. The opposed Garden of Eden model posits that anatomically modern humans arose in a...