The Journey of Man: A Genetic Odyssey (22 page)

Several anthropological traits – most famously the dental pattern known as sinodonty – are found in north-east Asia and the Americas. By the mid-twentieth century anthropologists such as Carleton Coon had even begun to classify Native Americans as ‘Mongoloid’ in their racial checklists. The problem was that no one knew exactly how long the Native Americans had been living there, and when they had split from their Asian cousins. In the 1950s carbon dating was used to infer an age of 11,000 years for the archaeological site at Clovis, New Mexico. The remains at Clovis contained leaf-shaped stone spear points in the same layer as extinct mammoth bones, which immediately suggested great antiquity to its discoverers. Over the next two decades, sites dating to roughly the same time period were excavated throughout North America. The pattern that seemed to be emerging from the archaeological record was that humans had occupied the Americans no earlier than 12,000 years ago.

In the 1970s and 1980s, though, three archaeological digs – one in North America and two in South America – turned up evidence for a human presence before Clovis. The Meadowcroft Rockshelter in Pennsylvania yielded artefacts that were originally dated using radiocarbon to roughly 14,000 years ago, pre-dating Clovis by 3,000 years. The care with which Meadowcroft was excavated was impressive, and while the dates for the earliest occupation have been revised downward (to around 12,500 years ago), they are accepted by many anthropologists. The site at Monte Verde in southern Chile yielded similar dates to those at Meadowcroft, roughly 13,000 years ago, although nearby hearths have been estimated to be as old as 33,000 years. The earlier date has not been widely accepted, and thus Monte Verde is thought to date – like Meadowcroft – to around 13,000 years ago.

The age of the remains at Monte Verde suggests that humans must have been in North America at least several hundred years prior to reaching Chile, and so pushes back the date of settlement a bit. But it was the final site that was the real bombshell. In a 1986 paper in the scientific journal
Nature
, archaeologist Niede Guidon summed up the find in the title: ‘Carbon-14 dates point to man in the Americas 32,000 years ago’. It was the result of her excavation of the cave of Boqueirão de Pedro Furada in north-eastern Brazil, and it seemed to pull the rug out from under the post-13,000 consensus. Careful examination,
though, has failed to confirm Guidon’s results. The charcoal from the site, which provided a radiocarbon date and was thought by Guidon to be the remains of a fireplace, could have been produced by a natural fire. Furthermore, most of the crude stone artefacts discovered there do not look convincingly human in origin – they could easily have resulted from natural breakage. These doubts have led palaeoanthropologist Richard Klein to suggest that ‘Furada may soon join the long list of dubious claims [for early human settlement in the Americas]’.

In summary, most of the reliable archaeological evidence points to a settlement of the Americas within the past 15,000 years. There is one small problem with this scenario, however: it was at this point that the ice age was at its most intense, and if early humans came from Siberia – as the anthropological and archaeological material suggests – they would have had to traverse the harshest environment on earth just as it reached its nadir. It implies a journey of unimaginable hardship for a species that only recently left its tropical homeland. Surely this would have been impossible? It is at this point that the genetic data provides us with more clues.

Doug Wallace, a geneticist at Emory University in Atlanta, had helped to pioneer mitochondrial DNA analysis of human populations when he was at Stanford University in the early 1980s. By the time he moved to Emory in the mid-1980s, he had become focused on the origins of Native Americans. In particular, he was trying to use mtDNA as a tool to track the origin of the first Native Americans back to particular populations in Asia. The first major publication of this work, in 1992 with Antonio Torroni, showed that Native Americans could be divided into at least two waves of migration. The earliest led to the settlement of both North and South America, while the later wave of migration left genetic traces only in North America. Their estimates of when these migrations took place varied widely, and could have occurred any time between 6,000 and 34,000 years ago. The results confirmed, though, that Native Americans and north-east Asians shared a recent common mitochondrial ancestry.

But how do these results fit with our Y-chromosome data? This question was answered in 1996 by Peter Underhill and his colleagues. Underhill found a single nucleotide change on the Y-chromosome, later named M3, that was common throughout the Americas. While
their sample of Native Americans was by no means exhaustive, over 90 per cent of South and Central Americans they examined were M3, while around 50 per cent of North Americans had this lineage. Clearly, it was the major Native American Y-chromosome founder, defining the American clan.

The only problem was that M3 was not found in Asia. This could have been due to its age, which Underhill and his colleagues estimated to be as little as 2,000 years. The age estimate was extremely uncertain, however, as it was made in the early days of Y-chromosome analysis, and the mutation rate of the single microsatellite used to assess M3 diversity (using the same method that we used to date M173 in Europe) was uncertain. Thus it could also have been as much as 30,000 years old. Clearly more work was needed.

This came in 1999, when Fabricio Santos and Chris Tyler-Smith at Oxford and Tanya Karafet and Mike Hammer at the University of Arizona independently reported that the ancestor of M3 was defined by a marker called 92R7, named for an undefined nucleotide change on the Y-chromosome. They found that 92R7 was present in populations throughout Eurasia, with a distribution from Europe to India. In combination with other nucleotide changes this pinpointed Siberia as the source population for Native Americans, confirming Wallace’s results from mitochondrial DNA. It was difficult to assess the age of the 92R7 lineage, however, as it was so widespread. What was needed was an additional marker on the lineage that would focus attention on the precise populations that could have given rise to the first Native Americans.

When it was later shown that M45 marks the same Y lineage as 92R7, the results made much more sense. Here was our central Asian marker, the same one that gave rise to M173 in Europe. It seems that the central Asian clan had made it to the New World as well, picking up the defining M3 marker in the process. This helped to trace a clear migrational route from Africa to the Middle East to the Americas, via the Eurasian steppes, but it still left us with the problem of how to date the first entry into the Americas. It could have happened any time between 40,000 and 12,000 years ago, taking into account both the genetic and archaeological results.

A recent analysis of the M45 lineage by Mark Seielstad and myself
has defined a further marker, known as M242, which is a descendant of M45. It appears to have arisen in central Asia or southern Siberia around 20,000 years ago, and is distributed across Asia, from southern India to China to Siberia, as well as throughout the Americas. It is found at highest frequency in Siberia, and thus it could be called a Siberian marker. It is also immediately ancestral to M3, and defines an evolutionary order of M45 → M242 → M3 that traces a migration from central Asia to the Americas within the past 20,000 years. M242 appears to be the oldest genetic marker in the Americas. Thus the Y-chromosome results have given us much the same picture of the founding of the Americas as the mtDNA results, but have narrowed down the date of entry considerably. Clearly an entry prior to 20,000 years ago is inconsistent with the genetic results, since M242 was still in central Asia at that time. A more recent migration from Siberia is overwhelmingly likely, consistent with the archaeological evidence.

The picture that seems to be emerging from the genetic analysis of Native Americans is that of a migration by the Siberian clan from southern to eastern Siberia within the past 20,000 years. This initial move established a population at the north-eastern edge of Asia. Adapted to a hunting life on the central Asian steppes, they would have subsisted almost entirely off of the large mammals of the far north – musk ox, reindeer and mammoth among them. Consummate hunters, with finely crafted microlith tools, portable dwellings and clothing capable of withstanding the intense cold, these well-adapted tundra dwellers would have gradually extended their range eastward. As the ice age moved toward its lowest temperatures, and more moisture became tied up in the ice caps, sea levels would have dropped by over 100 metres. This would have created a land bridge in Beringia, between Siberia and Alaska, of ice-free land formerly submerged in the Bering Sea. The Siberian clan would have been able to move back and forth across this connection, living a dual Asian-American existence.

Figure 8 M45 is the ancestor of most western Europeans (who have M173) and Native Americans (who have M242 and M3).

However, these first Americans of 15–20,000 years ago had one more obstacle to overcome. They would almost certainly have been barred from southward expansion by a continuous sheet of ice that covered most of northern Canada and eastern Alaska. It was only as the ice age began to abate, after 15,000 years ago, that it would have
become possible to transit the formerly icy interior and enter the North American plains, perhaps via a so-called ‘ice-free corridor’ that some palaeoclimatologists believe ran along the eastern edge of the Rocky mountains. It is around this time that grizzly bears first enter North America from Siberia, showing that humans weren’t the only species to have been stopped by the Alaskan ice. So, the genetic age of 20,000 years, as well as climatological considerations having to do with the extent of glaciation and sea levels, provide an explanation for why we don’t see archaeological remains in the Americas before this time. While archaeologists may someday discover a site that pre-dates 15,000 years ago, the mass of evidence is now in favour of a relatively late initial entry to the Americas. The stones and bones seem to agree with the DNA.

Interestingly, the Native American genetic data allows us to estimate how many people would have made it into the continent during these first migrations. By looking at the number of chromosomes needed to account for the present distribution of genetic lineages in the Americas, and working out how much diversity would have accumulated over the time that they have been in the continents, it is possible to account for all of the mtDNA and Y-chromosome types in Native Americans with a founding population of around ten or twenty individuals. Because some lineages would have gone extinct during the past 15,000 years, as we saw with our French soup recipes, this is certainly an underestimate of the number of individuals who actually made it across. Perhaps a few dozen, or even a few hundred, actually made the trip. Clearly, though, the diversity present in the Americas is a tiny fraction of that found in Eurasia – which in turn is merely a subset of that found in our African forebears. And of those who made it to Alaska, only a few left descendants. The gene pool of Native Americans carries in it a signal of the hardships faced by their Beringian ancestors thousands of years ago as they moved ever deeper into the deep freeze, scraping a living from the frozen wastes of the far north.

After they made it through the rigours of life in the freezer, the plains of North America must have seened like the promised land. Here was a vast grassland – much like the steppe they left long ago in central Asia – filled with large grazing animals. It was as though someone who had been adrift on a raft in the ocean for weeks was suddenly transported into a supermarket. The result was a massive increase in population as these highly efficient Siberian hunters took advantage of their newfound good fortune. In just 1,000 years or so they journeyed all the way to the tip of South America, in the process helping to kill off many of the species that made the plains such good hunting grounds. Three-quarters of the large mammals in the Americas were driven to extinction around this time, among them mammoths and horses – the latter weren’t to reappear in the Americas until the Spaniards introduced them in the fifteenth century. While humans may not have done the job on their own – climate change at the end of the
last ice age almost certainly played a major role – they probably delivered the
coup de grâce
to the gentle giants of the plains.

One of the most contentious issues in the study of Native American origins is deceptively simple: how many waves of migration were there into the New World? If the earliest Americans came from Siberia, did later migrants arrive from further afield? The 9,500-year-old ‘Caucasoid’ skull recently discovered at Kennewick, in Washington state, hints at ancient connections to Europe. Some anthropologists believe that Australian Aborigines migrated to South America, while others think that the Japanese managed to sail across the Pacific thousands of years ago. Can the genetic data help us to sort through these possibilities and weed out the plausible from the simply barmy?