SURVEY: HUMAN EVOLUTION
The long march of everyman
Dec 20th 2005
From The Economist print edition
It all started in Africa
OUT of Africa, always something new”, wrote Caius Plinius Secundus, a Roman polymath who helped to invent the field of natural history. Pliny wrote more truly than he could possibly have realised. For one fine day, somewhere between 85,000 and 60,000 years before he penned those words, something did put its foot over the line that modern geographers draw to separate Africa from Asia. And that something—or, rather, somebody—did indeed start something new, namely the peopling of the world.
Writing the story of the spread of humanity is one of the triumphs of modern science, not least because the ink used to do it was so unexpected. Like students of other past life forms, researchers into humanity's prehistoric past started by looking in the rocks. The first fossilised human to be recognised as such was unearthed in 1856 in the Neander Valley near Dusseldorf in Germany. Neanderthal man, as this skeleton and its kin became known, is now seen as a cousin of modern humans rather than an ancestor, and subsequent digging has revealed a branching tree of humanity whose root can be traced back more than 4m years (see article).
Searching for human fossils, though, is a frustrating exercise. For most of their existence, people were marginal creatures. Bones from periods prior to the invention of agriculture are therefore excedingly rare. The resulting data vacuum was filled by speculation scarcely worthy of the name of theory, which seemed to change with every new discovery. Then, in the 1980s, a geneticist called Allan Wilson decided to redefine the meaning of the word “fossil”. He did so in a way that instantly revealed another 6 billion specimens, for Wilson's method made a fossil out of every human alive.
In retrospect, Wilson's insight, like many of the best, is blindingly obvious. He knew, as any biologist would, that an organism's DNA carries a record of its evolutionary past. In principle, looking at similarities and differences in the DNA sequences of living organisms should allow a researcher to reconstruct the family tree linking those organisms. In practice, the sexual mixing that happens with each generation makes this tedious even with today's DNA-analysis techniques. With those available in the 1980s it would have been impossible. Wilson, however, realised he could cut through the problem by concentrating on an unusual type of DNA called mitochondrial DNA.
Mitochondria are the parts of a cell that convert energy stored in sugar into a form that the rest of the cell can use. Most of a cell's genes are in its nucleus, but mitochondria, which are the descendants of bacteria that linked up with one of humanity's unicellular ancestors some 2 billion years ago, retain a few genes of their own. Mitochondrial genomes are easy to study for three reasons. First, they are small, which makes them simple to analyse. Second, mitochondria reproduce asexually, so any changes between the generations are caused by mutation rather than sexual mixing. Third, in humans at least, mitochondria are inherited only from the mother.
In 1987, Rebecca Cann, one of Wilson's students, applied his insight to a series of specimens taken from people whose ancestors came from different parts of the world. By analysing the mutational differences that had accumulated since their mitochondria shared a common ancestor, she was able to construct a matriline (or, perhaps more accurately, a matritree) connecting them.
The result was a revelation. Whichever way you drew the tree (statistics not being an exact science, there was more than one solution), its root was in Africa. Homo sapiens was thus unveiled as an African species. But Dr Cann went further. Using estimates of how often mutations appear in mitochondrial DNA (the so-called molecular clock), she and Wilson did some matridendrochronology. The result suggests that all the lines converge on the ovaries of a single woman who lived some 150,000 years ago.
There was much excited reporting at the time about the discovery and dating of this African “Eve”. She was not, to be clear, the first female Homo sapiens. Fossil evidence suggests the species is at least 200,000 years old, and may be older than that. And you can now do a similar trick for the patriline using part of the male (Y) chromosome in the cell nucleus, because this passes only from father to son. Unfortunately for romantics, the most recent common ancestor of the Y-chromosome is a lot more recent than its mitochondrial equivalent. African Adam was born 60,000-90,000 years ago, and so could not have met African Eve. Nevertheless, these two pieces of DNA as they have weaved their ways down the generations have filled in, in surprising detail, the highways and byways of human migration across the face of the planet.
Sons of Adam, daughters of Eve
Detail, however is not the same as consensus, and there are two schools of thought about how people left Africa in the first place. Appropriately, some of their main protagonists are at the rival English universities of Oxford and Cambridge. The Oxford school, championed by Stephen Oppenheimer, believes that the descendants of a single emigration some 85,000 years ago, across the strait of Bab el Mandeb at the southern end of the Red Sea, are responsible for populating the rest of the world. The Cambridge school, championed by Robert Foley and Marta Mirazón Lahr, agrees that there was, indeed, a migration across this strait, though probably nearer to 60,000 years ago. However, it argues that many non-Africans are the descendants of at least one subsequent exodus.
Both schools agree that the Bab el Mandebites spread rapidly along the coast of southern Arabia and thence along the south coast of Asia to Australia, though Dr Oppenheimer has them turning inland, too, once they crossed the strait of Hormuz. But it is in describing what happened next that the two versions really part company, for it is here that the descendants of the Oxford migration run into the eruption of Toba.
That Toba devastated South and South-East Asia is not in doubt. Thick layers of ash from the eruption have been found as far afield as northern Pakistan. The question is whether there were people in Asia at the time. One of the most important pieces of evidence for Dr Oppenheimer's version of events is some stone tools in the ash layer in Malaysia, which he thinks were made by Homo sapiens. Molecular clocks have a regrettable margin of error, but radioactive dating is a lot more accurate. If he is right, modern humans must have left Africa before the eruption. The tools might, however, have been crafted by an earlier species of human that lived there before Homo sapiens.
For Dr Oppenheimer, the eruption was a crucial event, dividing the nascent human population of Asia into two disconnected parts, which then recolonised the intermediate ground. In the Cambridge version, Homo sapiens was still confined to Africa 74,000 years ago, and would merely have suffered the equivalent of a nuclear winter, not an ash-fall of up to five metres—though Dr Ambrose and his colleagues think even that would have done the population no good.
The Cambridge version is far more gentle. The descendants of its subsequent exodus expanded north-eastwards into central Asia, and thence scattered north, south, east and west—though in a spirit of open-mindedness, Sacha Jones, a research student in Dr Foley's department, is looking in the ash layer in India to see what she can find there.
Which version is correct should eventually be determined by the Genographic Project, a huge DNA-sampling study organised by Spencer Wells, a geneticist, at the behest of America's National Geographic Society and IBM. But both already have a lot in common. Both, for example, agree that the Americas seem to have been colonised by at least two groups. The Cambridge school, though, argues that one of these is derived ultimately from the first Bab el Mandeb crossing while the other is descended from the later migrants.
Both also agree that Europe received two waves of migration. The ancestors of the bulk of modern Europeans came via central Asia about 35,000 years ago, though some people in the Balkans and other parts of southern Europe trace their lines back to an earlier migration from the Middle East. But the spread of agriculture from its Middle Eastern cradle into the farthest reaches of Europe does not, as some researchers once thought, seem to have been accompanied by a mass movement of Middle Eastern farmers.
The coming together of two groups of humans can be seen in modern India, too. In the south of the subcontinent, people have Y-chromosomes derived almost exclusively from what the Cambridge school would interpret as being northern folk (and the Oxford school as the western survivors of Toba). However, more than 20% of their mitochondria arrived in Asia with the first migration from Africa (or, according to taste, clung on along the south-eastern fringes of the ash plume).
That discovery speaks volumes about what happened when the two groups met. It suggests that many modern south Indians are descended from southern-fringe women, but few from southern-fringe men—implying a comprehensive conquest of the southerners by the northerners, who won extra southern wives.
This observation, in turn, helps explain why Y-chromosome Adam lived so much more recently than mitochondrial Eve. Displacement by conquest is one example of a more general phenomenon—that the number of offspring sired by individual males is more variable than the number born by individual females. This means that more males than females end up with no offspring at all. Male gene lines therefore die out faster than female ones, so those remaining are more likely, statistically, to converge in the recent past.
Successful male gene lines, though, can be very successful indeed. Students of animal behaviour refer to the top male in a group as the “alpha”. Such dominant animals keep the others under control and father a large proportion, if not all, of the group's offspring. One of the curiosities of modern life is that voters tend to elect alpha males to high office, and then affect surprise when they behave like alphas outside politics too. But in the days when alphas had to fight rather than scheme their way to the top, they tended to enjoy the sexual spoils more openly. And there were few males more alpha in their behaviour than Genghis Khan, a man reported to have had about 500 wives and concubines, not to mention the sexual opportunities that come with conquest. It is probably no coincidence, therefore, that one man in every 12 of those who live within the frontiers of what was once the Mongol empire (and, indeed, one in 200 of all men alive today) have a stretch of DNA on their Y-chromosomes that dates back to the time and birthplace of the great Khan.