To supplement last September’s conclusion that the peopling of the Americas was initiated by a pretty diverse group of people who camped out in Beringia for a long time, long enough to differentiate from their Asian sister-clades, comes this study published in this week’s PLoS One, “A Three-Stage Colonization Model for the Peopling of the Americas.”
The new study offers up perhaps the largest published alignments of Native American mtDNA, spanning all Native American haplogroups. Over 77 full mitochondrial coding genomes were constructed from 812 concatenated mtDNA hypervariable region (HVR) I and II sequences. They took these mitochondrial genomes, aligned them up, and applied the same algorithims, Bayesian skyline plotting, used in a recent paper to estimate prehistoric population sizes. Bayesian skyline plots are a unique approach to the coalescent modeling, that assume a single migration event, and thus test the generally agreed consensus that there was a single migration of people in the Americas.
The dominant model on the peopling of Americas started with the the ice sheets advancing and sea levels falling about 17,000 years ago. It is during this time that people are thought to first migrate across the Eurasian landmass and into the Americas. It was thought that a very small number of people, maybe even as small as 70 or so, who crossed over. Perhaps they were nomadic hunters, following game herds from Siberia across what is today the Bering Strait into Alaska, and then gradually spreading southward. Based upon the distribution of Amerind languages and language families, a movement of tribes along the Rocky Mountain foothills and eastward across the Great Plains to the Atlantic seaboard is assumed to have occurred some 10,000 years ago.
What this new study found doesn’t indicate that the peopling of the Americas happened in one fell swoop. Rather, it supplements the paper I mentioned above, “Beringian Standstill and Spread of Native American Founders,” and offers up an interesting time frame, a three stage colonization event that is effectively illustrated by the authors. I’ve cut out the figure the authors provided, it’s to your left. The first stage began about 40,000 years ago with a gradual ancestral population expansion of people from an East Central Asian gene pool into Beringia. The second stage of ‘proto-American Indians’ was marked by almost no population growth for about 20,000 years, which confirms the previous Beringian standstill conclusion. The last stage started about 16,000 years ago with a massive rush of people, about 5,000 strong, fleeing out of Beringia and into the “ice free, inland corridor between the eastern Laurentide and western Cordilleran Ice Sheets and/or along the Pacific coast.” This challenges the n=70 founding population estimation.
The authors offer up no discussion about a possibility of bidirectional gene flow, which was shown in the September 2007 Beringian Standstill study that I keep referring too. I’m thinking its cause the skyline plots test for single migration events and not backflows.
Either way, it is a very enlightening study especially because other models, such as the Clovis archaeological model, which says the peopling of the Americas happened in 11,000 years, estimate really rapid colonization events. That’s awfully fast. Pushing back first rush of people to 16,000 years ago can help us better explain how the Clovis culture radiated so fast.
- Kitchen, A., Miyamoto, M.M., Mulligan, C.J., Harpending, H. (2008). A Three-Stage Colonization Model for the Peopling of the Americas. PLoS ONE, 3(2), e1596. DOI: 10.1371/journal.pone.0001596