The Context of Human Genetic EvolutionRobert Foley1Human Evolutionary Biology Research Group, Department of Biological Anthropology, and King’s CollegeResearch Centre Human Diversity Project University of Cambridge, Cambridge, CB2 3DZ, UKThe debate on modern human origins has often focused on the relationship between genes and fossils. Althoughmore and more genetic evidence has been accumulating in favor of a recent African origin for modern humans,it has been assumed by many that the fossil evidence remains ambiguous. On the contrary, it has been clear forsome time that the fossil evidence does not support the multiregional model: Fossils and archeology indicate apattern of multiple dispersals from and beyond Africa, against which the genetic data can be compared. Thecontinuing value of paleobiology is in complementing genetic information by revealing the context of humanevolution: locating the dispersals and extinctions of populations in time and space, correlating these events withthe environmental forces that shaped them, and providing an increasingly detailed understanding of themorphology and technology of early humans.Molecular biology has revolutionized the study ofhuman evolution. The importance of fossils as theprimary source of information about our past hasbeen steadily undermined as it has become possibleto infer detailed aspects of recent human historyfrom the distribution and frequency of genes foundaround the world today. To some extent, a fusion ofpaleontological and genetic approaches came aboutlast year with the extraction and sequencing of an-cient DNA from an extinct hominid, the Neander-thal type specimen (Krings et al. 1997), but this islikely to remain a rarity. Ancient DNA, however, hasbeen able to confirm that humans and Neander-thals belonged to different populations over the lastquarter of a million years and that Neanderthals didbecome effectively extinct.These results fit in with an increasingly consis-tent interpretation of human evolutionary genetics.Compared with chimpanzees and other apes, thehuman population is relatively lacking in geneticdiversity (Ruvolo et al. 1993); such genetic variationas does exist occurs primarily within populationsrather than between (Relethford and Harpending1994); African populations are more diverse geneti-cally than those found anywhere else in the world(Vigilant et al. 1991; Cavalli-Sforza et al. 1994; Wat-son et al. 1997); and for the most part, non-Africanpatterns of genetic variation can be treated as a sub-set of African ones. The chronological and demo-graphic context of the processes of diversificationhas been strongly disputed, but several genetic sys-tems indicate that living populations derive from arelatively small population (effective populationsize between 5000 and 50,000 individuals) (Rogersand Harpending 1992; Harpending et al. 1993; Neiand Takahata 1993; Harpending 1994). That sizerepresents a bottleneck in the hominid lineage dat-ing back no more than 200,000 years (Cann et al.1987; Stoneking et al. 1992), with evidence of de-mographic expansion occurring in the last 70,000years (Rogers 1995).This emerging consensus has been interpretedin evolutionary terms as evidence for a recent Afri-can origin (i.e., ∼200,000 years ago) for modern hu-mans, followed by a dispersal out of Africa, withlittle or no interbreeding with other populations ofhominid. Many anomalies and further details re-main to be sorted out. The geographical clades ofthe mtDNA have been questioned (Maddison 1991;Maddison et al. 1992), and it has been suggestedthat much of the genetic evidence is consistent witha number of different models (Templeton 1992,1993). Furthermore, tree building and mismatchtechniques, the primary means of turning geneticinformation into evolutionary history, have beensubject to dispute (Templeton 1992, 1993). Finally,estimates of the coalescence time of human b-globin genes (∼800,000 years) have also been usedto argue against a recent African origin for modernhumans (Fullerton et al. 1997), but they are not, infact, inconsistent with the recent-origin model, asthe coalescence time for a neutral autosomal locuswould be expected to be roughly four times that ofa mitochondrial gene, and selection may also play aconfounding role.Such disputes, though, should perhaps not dis-tract from the general pattern: Genetics supports a1E-MAIL [email protected]; FAX 44 (0) 1223-335460.PERSPECTIVE8:339–347 ©1998 by Cold Spring Harbor Laboratory Press ISSN 1054-9803/98 $5.00; www.genome.org GENOME RESEARCH 339single recent African origin for modern humans.This consensus leads to two important questions:First, what is the relationship between this evidenceand the fossil evidence that it has been claimed,supports the so-called multiregional model? Andmore broadly, what can paleoanthropology contrib-ute to a field where the bulk of the resources is beingput into genetics? The answer to the first question isthat it has been clear for a long time that the mor-phological evidence does not support a multire-gional view. However, this has not percolatedthrough to the genetic community. This apparentlack of communication between the two disciplinescan be used to help provide an answer to the secondquestion: that paleobiology provides a different sortof evidence, one that focuses specifically on thecontext in which human genetic evolution oc-curred. These themes will be developed below.Multiregional EvolutionThe multiregional model of human evolution has along history. It was first proposed by Franz Weiden-reich (1943) in relation to his analyses of the PekingMan fossils but has been further developed and pro-moted in recent years by Milford Wolpoff and AlanThorne (Wolpoff et al. 1984). The essence of thismodel is that since the origin of Homo some twomillion years ago, the human lineage has evolved asa single lineage, without speciation events (Wolpoffand Caspari 1997). Within this lineage there wouldhave been universal directional trends in characterssuch as brain size, but regional differences in popu-lations would have persisted. Over time, therefore,some specific morphological traits would persist inparticular regions, regardless of other global trends;the human lineage thus evolved in several regions(multiregionality). This continuity has been seen toextend to the transition from archaic to modernhominids in Africa, Asia, and Europe, the last ofthese being the link between Neanderthals andmodern Europeans. The