This is a modified version of a paper by Henry Harpending, Stephen Sherry, Alan Rogers and Mark Stoneking (Current Anthropology 1993 pages 483-496) entitled The Genetic Structure of Ancient Human Populations. (Richard Effland)

Given these assumptions, the results show that human populations are derived from separate ancestral populations that were relatively isolated from each other before 50,000 years ago. Major population expansion took place between 80,000 and 30,000 years ago - 80,000 in Africa and perhaps 40,000 years ago among ancestors of Europeans. How long were these ancestral populations isolated from each other? Since all human mtDNAs (mitochondrial DNAs) coalesce approximately 200,000 years ago, mtDNA provides no information about population structure before that time.

The mtDNA coalescence time is consistent with indications from other polymorphisms that the ancient effective size of our species or its ancestral species was of the order of 103 to 104 females. This rather small size seems difficult to reconcile with the evidence of isolated, hence geographically separated ancestral populations. Could Homo Erectus or early Homo Sapiens have occupied large parts of Africa, Europe, and Asia yet have consisted of only several thousands or tens of thousands of individuals? Clark estimates that of the approximately 48 million square kilometers of arable land in the Old World, 20 million square kilometers are in cold-climate areas and "two-crop tropical areas". Since Homo Erectus did not use these areas, the land occupied by these species in the Old World may have amount to roughly 25 million square kilometers. Lee and DeVore, in their summary of a conference on foraging (Hunting and Gathering) peoples, suggest that reliable reports of contemporary density are 1 to 25 persons per 100 square miles.

Upper Paleolithic population densities in Europe were much greater than those of their archaic predecessors. The middle of the Lee and DeVore estimate is about 5 persons per 100 square kilometers. Reducing this by one order of magnitude gives 5 per 1,000 square kilometers or 125,000 as the world population of Homo Erectus. Weiss arrives at an estimate of 500,000 by positing a smaller occupied region of the Old World but densities equal to those of contemporary foraging populations. A total population of 125,000 would correspond to an effective number of females of roughly 25,000. The effective population size of fluctuating population is much closer to the minimum than to the average, so the ecological estimate of 25,000 is not wildly inconsistent with high-end genetic estimates of the long-term effective number of females 5,000.

It is, however, on the edge of credibility because our ecological assumptions favor a small world population of Homo Erectus and the mtDNA data suggest that the ancestral effective size was closer to 500 females...Comparisons of Acheulian assemblages over large areas with assemblages left by modern humans reveal a remarkably uniformity in the Acheulian from the Cape (South Africa) to Korea. Even derived artifact industries such as European Middle Paleolithic (Mousterian) are characterized by remarkably little variability through time or space. The number of readily identifiable stone tool types they contain is relatively small and formal bone artifacts, decorative items, and art objects are all but absent. It is hard to imagine how this apparent cultural uniformity could have persisted without high levels of movement and mate exchange between groups. This mate exchange ought to have maintained unity of the species or at least the macropopulation. If local groups were bands of 25, there would have been 5,000 bands occupying 25 million square kilometers or 5,000 square kilometers per band. If these had been packed into a hexagonal lattice, the average spacing between bands would have been about 75 kilometers; because instead they would have been concentrated around special features of the environment such as river valleys, the real average spacing might have been considerably less, perhaps within the range of dispersal of a large mammal.

Arguments from the homogeneity of cultural remains of Homo Erectus must also contend with the apparent ancient separation of the world into the Acheulian region and the Asian chopper-chopping tool region. While population genetics so far cannot absolutely distinguish between the multiregional hypothesis and the Out of Africa hypothesis of the origins of modern humans from Homo Erectus, it strongly favors the latter. The uniformity of the Acheulian suggests high levels of migration between groups, but the small effective size of the total population of modern human ancestors suggest that they could not have been spread over such a large area and still maintained high rates of gene flow. Even with assumptions that radically favor the multiregional hypothesis we arrive at total population size estimates of Homo Erectus that are too large to accommodate the genetic data.

A scenario that is consistent with both the ecological and the genetic lines of evidence is as follows: Around 100,000 years ago ancestral humans spread into separate regions from a restricted source, but there was not necessarily a dramatic expansion. The data are consistent with an early expansion and subsequent bottleneck or with an early modest growth and slow expansion. Later, starting around 50,000 years ago, dramatic population growth occurred separately within dispersed daughter populations that were genetically isolated from each other. This is the weak Out of Africa hypothesis. In the archaeological record, the time range of 45,000-35,000 years ago is an approximate boundary for an apparently dramatic change in human behavior. In the interval between 50,000 and 40,000 before present (b.p.), the Mousterian, the African Middle Stone Age, and similar industries were widely replaced by clearly more advanced industries, including those of the Upper Paleolithic, in which the degree of geographic and temporal variability is far greater, the number of readily recognizable stone artifact types is much larger, and formal bone artifacts and art objects are common. The suggestion is that culture rather than biology drove the burst of growth of our ancestors. This scenario is also consistent with evidence from genetic studies of protein polymorphisms and nuclear DNA....The present data are clearly inconsistent with the strong Out of Africa hypothesis. If there was indeed a single large expansion from Africa around 100,000 years ago, we should see the signature of it in the mtDNA (mitochondrial DNA), but instead we see indications of multiple later expansions associated with modern technology instead of modern morphology.