New gene study enters human origins debate.

In a finding that captured the imagination of a scientists and the public alike, researchers announced in 1987 that an analysis of mitochondrial DNA -- genetic material located outside the cell nucleus and inherited only from the mother -- traced the maternal lineage of all humans back to an African "Eve" who lived about 200,000 years ago. A computer-run statistical analysis of mitochondrial DNA samples drew an evolutionary tree with African roots lying at mitochondrial Eve's feet, suggesting modern humans originated in Africa and rapidly spread elsewhere.

Fatal statistical flaws in this approach later emerged and researchers dropped genetic arguments for the "Out of Africa" theory (SN: 2/22/92, p.123). However, a new type of mitochondrial DNA analysis, described in the August-October CURRENT ANTHROPOLOGY, now presents a more complicated picture of human evolution.

In this scenario, a small subgroup of Homo erectus evolved into modern humans -- probably in Africa -- and slowly trekked to several parts of Europe and Asia beginning around 100,000 years ago. About 50,000 years later, geographically isolated human populations experienced dramatic growth and expansion fueled by the appearance of many cultural innovations.

Genetic analysis offers weaker support for the multiregional evolution theory, a competing view of human origins, contend Henry C. Harpending, an anthropologist at Pennsylvania State University in University Park, and his colleagues, The multiregional theory holds that modern humans evolved simultaneously in several parts of the world for around 2 million years, with contact between separate populations along the way.

Rather than constructing evolutionary trees out of genetic data, Harpending's group analyzed the differences in sequences of mitochondrial DNA both within and between human groups now livig in Africa, Asia, and Europe. According to the researchers, these differences preserve a record of ancient population expansions and separations, which they modeled in computer simulations of mitochondrial DNA change in pairs of populations.

"In living populations, between-group mitochondrial DNA differences far out-pace within-group differences," Harpending holds. But this pattern of change in the structure of mitochondrial DNA does not characterize his computer models of single populations that rapidly grow and split into separate clusters. "Groups of archaic humans apparently remained isolated from each other for tens of thousands of years," Harpending says.

The dating of humanity's common mitochondrial ancestors does not show that our species suddenly evolved around 200,000 years ago, Harpending says. The mitochondrial DNA evidence simply cannot illuminate the structure of human populations before that time, he asserts. But his group estimates that the number of human females at the time mitochondrial Eve lived ranged from 1,000 to no more than 10,000.

This relatively small population shows genetic signs of slight size expansion in Africa around 100,000 years ago, with major size increases occurring on that continent approximately 80,000 years ago, the researchers maintain. Population growth blossomed in Asia and Europe about 50,000 to 40,000 years ago, according to the mitochondrial DNA comparisons.

Up until these growth spurts, stone tools and other articles found at sites throughout Eurasia displayed many similarities; soon thereafter, sophisticated regional cultures appeared, Harpending and his co-workers note. Indeed, cultural change may have sparked marked population increases in dispersed human groups, they argue.

Alan R. Templeton, an evolutionary biologist at Washington University in St. Louis, regards the new analysis of mitochondrial DNA with considerable skepticims. He provided the statistical critique that chopped down earlier evolutionary trees derived from mitochondrial DNA.

"This study is a step in the right direction," Templeton remarks. "But the computer models of population expansion are pretty simple and only test the Out of Africa theory, not multiregional evolution."

Harpending acknowledges that large margins of error exist in his simulations: "We all feel that we need to move beyond mitochondrial DNA as a locus of study."

In a report in the March AMERICAN ANTHROPOLOGIST, Templeton found no evidence for a definite geographic origin for a common mitochondrial ancestor, whom he dates to around 800,000 years ago. Current mitochondrial DNA variations come from dispersed, ancient populations, he contends.

Using a computer program that analyzes the geographic distribution of DNA differences, Templeton concluded that humans experienced size expansions largely within continents, with periodic contact across continents.