free trial issue                                             subscribe                                                  back issues

Athena Review: Vol.3, no.2: Recent Finds in Archaeology 

Out of Africa at Least Three Times

All the phases of Eve: a new evolutionary model with interbreeding, not replacement

“Science,” writes paleoanthropologist Ian Tattersall (2001), curator at the American Museum of Natural History, “is constantly subject to rearrangement and change as our collective knowledge increases.” Nowhere is this more true than in the field of genetics, which in the past two decades has produced a series of mind-staggering revelations about human evolution; knowledge which, it seems, is only just beginning to unfold.

The pace of upheavals in genetic knowledge began to seriously quicken about 15-20 years ago in both biochemical and informational technology. Biological methods were much enhanced with the advent of recombinant DNA technology, and through gene mapping techniques such as restriction fragment length polymorphism (RFLP) analysis. By pinpointing mutation sites, researchers can catalogue the sequence changes occurring within DNA loci. Using increasingly sophisticated statistical grouping procedures such as PAUP or GEODIS to establish the most parsimonious ancestral trees, geneticists are now able to map the identified segments of polymorphisms (called haplotypes) within and across human populations (for more detail and a DNA glossary see Schurr, this issue, p.62).

The time element of how long genetic strains have been separated can also be calculated by a “molecular clock” of counting numbers of mutations from known, fixed points on the chain where they once were identical. Such methods promised to resolve long-standing problems that could not be clarified through the fragmentary record of fossil humans.

Fifteen years ago, when haplotype studies were much newer, three scientists (Cann, Stoneking, and Wilson 1987) published evolutionary trees of the mitochrondrial DNA (mtDNA) haplotype. mtDNA helps reproduce the cell’s energy-producing mitochondria whose genome is inherited solely through the maternal line. These showed the oldest extant human branches to be in Africa, and (based on the time/mutation rate of the mtDNA haplotype) indicated a major migration out of Africa between 150,000 to 80,000 years ago. Based on the mtDNA evidence alone, this seemed to account for the entire modern human genome, and thus the theory was established (and took root in the popular press) that all modern humans descended from a single African woman, nicknamed “mitochondrial Eve.”

Most paleoanthropologists agree that hominids originated in Africa, then first spread out of Africa to Eurasia and South Asia as Homo erectus about 1.7 to 1.8 myr ago. The “mitochrondrial Eve” theory based on Cann et al.’s highly influential (1987) paper holds that all hominid lines which had descended from these original migrating Homo erectus populations of about 1.7 myr ago became extinct - including all the Asian and Eurasian varieties of Homo erectus, and all archaic Homo species such as Neanderthals who evolved outside of Africa. All were eventually replaced after 150,000 years ago by the African descendants of mitochrondrial Eve.  

This seemed to effectively refute the “lattice” or “trellis” model of multiregional continuity. This view (with many variations) holds that various descendants of Homo erectus throughout the Old World (including Neanderthals) have intermixed their genomes throughout the 1.7 million years since Homo erectus first left Africa. In support of the “regional continuity” side of the human evolutionary debate, geneticists including Alan R. Templeton (1992) of Washington University showed flawed use of the statistical (PAUP) program in the original (Cann et al. 1987) mtDNA paper. Meanwhile a revised, more detailed “mitochrondrial Eve” study was published (Vigilant et al. 1991), restating the total replacement hypothesis after 150,000 BP.

[Fig.1: Main trends of Templeton’s new model of the expansion of human populations, with three “out-of-Africa” migrations starting with Homo erectus about 1.7 myr ago (after Templeton 2002).]

Over the next few years, haplotype trees generated for Y-chromosomal DNA (nuclear DNA found only in males) seemed to basically support the mtDNA-based, recent out-of Africa replacement scenario. While some Y-chromosomal studies showed longer time depths for the out-of-Africa movement (i.e. back to 200,000 years ago), such results based on male descent were still compatible with the mtDNA full genetic replacement model as it applies to the fate of Homo erectus, Neanderthals (see AR 2,4), and the entire hominid fossil record outside of Africa before 100,000-200,000 years ago.  

But haplotype studies have continued to improve in focus, and expand in scope. Now Alan Templeton (2002), long a critic of the sweeping replacement scenario, has synthesized the data from ten different haplotype trees, combining those of the previously studied mtDNA and Y-chromosome DNA with eight new ones, including haplotypes from two X-linked DNA regions, and 6 autosomal DNA regions. His main conclusions (presented in the 7 March 2002 issue of Nature) are, first, that there were at least two major expansions of humans out of Africa after the initial spread by Homo erectus 1.7 myr ago (fig.1), and secondly, that there has been interbreeding, a “ubiquity of genetic interchange between human populations” throughout, not replacement.   One of the major migration periods out of Africa occurred between 840,000 and 420,000 years ago (when Homo erectus were evolving into early archaic Homo types).

Rebecca L. Cann (2002), co-author of the original (1987) “mitochondrial Eve” paper, allows (in a commentary in the same issue of Nature) that there is “strong genetic support” for this earlier population expansion from Africa. The next, beween 150,000 and 80,000 years ago, basically coincides in time with Cann et al.’s (1987) “recent-out-of-Africa” model, but with the important difference that all ten haploytes (including eight bisexually inherited nuclear genes) show there was continued interbreeding, not replacement, and “strongly reject the recent-out-of-Africa replacement event.”

Cann (2002) argues, however, that Templeton treats recombinant gene sequences (inherited by both parents) and non-recombinant sequences (mtDNA and Y-chromosome) as if they provide the same types of inheritance and temporal information. This may create conflicting results since recombinant gene sequences can mutate much faster than non-recombining ones. Templeton does acknowledge this, and only makes explicit inferences and interpretations for those scenarios where he obtained statistically significant results (using a confidence level of 95% or higher).

Templeton’s analyses (based on the GEODIS tree-grouping procedures) further signals a likelihood that the earlier out-of-Africa expansion event between .84 and .42 myr ago was also characterized by interbreeding, and not replacement. If these views prevail, the field is open for erecting a new kind of trellis (multiregional continuity and interbreeding) model - and for reviewing the case for integrating some Homo erectus and Neanderthals into the direct line leading to modern humans.

As Tattersall (2001) also said, “How can we make progress in science if what we believe today cannot be shown tomorrow to be somehow wrong or at least incomplete? Religious knowledge is in principle eternal, but scientific knowledge is by its very nature provisional.” Cann (2002), who views Templeton as “perhaps overambitious in the scale of his analysis,” cautions that larger and better samples are needed to make such global statements on long-term genetic trends. Just as the theory of mitochondrial Eve and of recent total replacement by anatomically modern humans after 150,000 years ago may today (after 15 years) be undergoing replacement, as it is shown to be incomplete, so Templeton’s compelling genetic evolutionary picture (as he might well agree) is only provisional, until the next synthesis of more and better data comes along - probably quite soon.

[Cann, R.L., Stoneking, M., and Wilson, A.C., “Mitochrondrial DNA and human evolution.” Nature 325, 1 Jan 1987; Cann, R.L., “Tangled genetic routes.” Nature 416, 7 Mar 2002; Tattersall, I., 2001 Monkey in the Mirror: Essays on the Science of What Makes us Human. Harcourt Press; Templeton, A.R., “Human origins and analysis of mtDNA sequences.” Science 255, 7 Feb 1992; Templeton, A.R., 2002, “Out of Africa again and again.” Nature 416, 7 Mar 2002; Vigilant et al. Science 253:1503-07, 1991]

Athena Review Image Archive™   |  Paleoanthropology in the News   |  Guide to Archaeology on the Internet   |   Free issue   |   Back issues

Main index of Athena Review   |   Subject Index   |   Travel Pages   |   Galleries and Museums   |  Ad rates   |  Current issue index |

Copyright  ©  1996-2003  Athena Publications, Inc.  (All Rights Reserved).