Marriage, we're told by the president and a lot of other people, can only be between one man and one woman. Anything else would go against thousands of years of tradition and nature itself. If the president's DNA could talk, I think it might disagree.

In the 1980s, geneticists began to study variations in human DNA to learn about the origin of our species. They paid particular attention to the genes carried by mitochondria, fuel-producing factories of the cell. Each mitochondrion carries its own small set of genes, a peculiarity that has its origins over two billion years ago, when our single-celled ancestors engulfed oxygen-breathing bacteria. When a sperm fertilize an egg, it injects its nuclear DNA, but almost never manages to deliver its mitochondria. So the hundreds of mitochondria in the egg become the mitochondria in every cell of the person that egg grows up to be. Your mitochondrial DNA is a perfect copy of your mother's DNA, her mother's DNA, and so on back through history. The only differences emerge when the mitochondrial DNA mutates, which it does at a fairly regular rate. A mother with a mutation in her mitochondria will pass it down to her children, and her daughters will pass it down to their children in turn. Scientists realized that they might be able to use these distinctive mutations to organize living humans into a single grand genealogy, which could shed light on the woman whose mitochondria we all share--a woman who was nicknamed Mitochondrial Eve.

Alan Wilson of the University of California and his colleagues gathered DNA from 147 individuals representing Africa, Asia, Australia, Europe, and New Guinea. They calculated the simplest evolutionary tree that could account for the patterns they saw. If four people shared an unusual mutation, for example, it was likely that they inherited from a common female ancestor, rather than the mutation cropping up independently in four separate branches. Wilson's team drew a tree in which almost all of the branches from all five continents joined to a common ancestor. But seven other individuals formed a second major branch. All seven of these people were of African descent. Just as significantly, the African branches of the tree had acquired twice as many mutations as the branches from Asia and Europe. The simplest interpretation of the data was that humans originated in Africa, and that after some period of time one branch of Africans spread out to the other continents.

Despite the diversity of their subjects, Wilson's team found relatively little variation in their mitochondrial DNA. Although their subjects represented the corners of the globe, they had less variation in their genes than a few thousand chimpanzees that live in a single forest in the Ivory Coast. This low variation suggests that living humans all descend from a common ancestor that lived relatively recently. Wilson's team went so far as to estimate when that common ancestor lived. Since some parts of mitochondrial DNA mutate at a relatively regular pace, they can act like a molecular clock. Wilson and his colleagues concluded that all living humans inherited their mitochondrial DNA from a woman who lived approximately 200,000 years ago.

The first studies by Wilson and others on mitochondrial DNA turned out to be less than bulletproof. They had not gathered enough data to eliminate the possibility that humans might have originated in Asia rather than Africa. Wilson's students continued to collect more DNA samples from a wider range of ethnic groups. Other researchers tried studying other segments of mitochondrial DNA. Today they have sequenced the entire mitochondrial sequence, and the data still points to a recent ancestor in Africa. All mitochondrial DNA, it now appears, came from a single individual who lived 160,000 years ago.

More recently, men offered their own genetic clues. Men pass down a Y chromosome to their sons, which remains almost completely unchanged in the process. Y chromosomes are harder to study than mitochondrial DNA (in part because each cell has only one Y chromosome but thousands of mitochondria). But thanks to some smart lab work, scientists began drawing the Y-chromosome tree. They also found that all Y chromosomes on Earth can be tracked down to a recent ancestor in Africa. But instead of 170,000 years, the age of "mitochondrial Eve," they found that their "Y-chromosome Adam" lived about 60,000 years ago.

This discrepancy may seem bizarre. How can our male and female ancestors have lived thousands of years apart? Different genes have different history. One gene may sweep very quickly through an entire species, while another one takes much longer to spread.

In 2001 I wrote an essay on this odd state of affairs for Natural History. At the time, scientists weren't sure just how real the discrepancy was. After all, both estimates still had healthy margin of errors. If mitochondrial Eve was younger and Y-chromosome Adam was older, they might have missed each other by only a few thousand years. On the other hand, if the gap was real, there were a few possible explanations. In one scenario, a boy 60,000 years ago was born with a new mutation on his Y chromosome. When he grew up, its genes helped him reproduce much more successfully than other Y chromosomes, and his sons inherited his advantage. Thanks to natural selection, his chromosome became more common at a rapid rate, until it was the only chromosome left in our species. (This selective sweep might have been just the last in a long line of sweeps.)

Now comes a fascinating new paper in press at Molecular Biology and Evolution. Scientists at the University of Arizona suspected that some of the confusion over Adam and Eve might be the result of comparing the results of separate studies on the Y chromosome and mitochondrial DNA. One study might look at one set of men from one set of ethnic backgrounds. Another study might look at a different set of women from a different set of backgrounds. Comparing the studies might be like comparing apples and oranges. It would be better, the Arizona team decided, to study Y chromosomes and mitochondrial DNA all taken from the same people. Obviously, those people had to be men. The researchers collected DNA from men belonging to three populations--25 Khosians from Southern Africa, 24 Khalks from Mongolia, and 24 highland Papuan New Guineans. Their ancestors branched off from one another tens of thousands of years ago.

The results they found were surprisingly consistent: the woman who bequeathed each set of men their mitochondrial DNA was twice as old as the man whose Y chromosome they shared. But the ages of Adam and Eve were different depending on which group of men the scientists studied. The Khosian Adam lived 74,000 years ago, and Khosian Eve lived 176,500 years ago. But the Mongolian and New Guinean ancestors were both much younger--Adam averaged 48,000 years old and Eve 93,000 years.

You wouldn't expect these different ages if a single Y chromosome had been favored by natural selection, the Arizona team argues. Instead, they are struck by the fact that Khosians represent one of the oldest lineages of living humans, while Mongolians and New Guineans descend from younger populations of immigrants who left Africa around 50,000 years ago. The older people have an older Adam and Eve, and the younger people have a younger one. The researchers argue that some process has been steadily skewing the age of Adam relative to Eve in every human population.

Now here's where things may get a little sticky for the "one-man-one-woman-is-traditional-and-natural" camp. The explanation the Arizona scientists favor for their results is polygyny--two or more women having children with a single man. To understand why, imagine an island with 1,000 women and 1,000 men, all married in monogamous pairs, just as their parents did, and their grandparents, and so on back to the days of the first settlers on the island. Let's say that if you trace back the Y chromosomes in the men, you'd find a common ancestor 2,000 years ago. Now imagine that the 1,000 women are all bearing children again, but this time only 100 men are the fathers. You'd expect that the ancestor of this smaller group of men lived much more recently than the common ancestor of all 1,000 men.

Scientists have proposed that humans have a history of polygyny before (our sperm, for example, looks like the sperm of polygynous apes and monkeys, for example). But with these new DNA results, the Arizona researchers have made a powerful case that polygyny has been common for tens of thousands of years across the Old World. It's possible that polygyny was an open institution for much of that time, or that secret trysts made it a reality that few would acknowledge. What's much less possible is that monogamy has been the status quo for 50,000 years.

People are perfectly entitled to disagree over what sort of marriage is best for children or society. But if you want to bring nature or tradition into the argument, you'd better be sure you know what nature and tradition have to say on the subject.