I talked about some pieces of this puzzle earlier: Patterson and Reich found an Amerindian-like component in Europeans, especially northern Europeans. Their first calculations showed such admixture in all Europeans other than Sardinians and Basques: later calculations found that all Europeans had this admixture, with the Sardinians and Basques having the least. Europeans (those they looked at) averaged around 25-35% of this Amerindian-like component. It might be higher in Scandinavia and the eastern Baltic. This is all theory, based on existing populations.
Eske Willerslev and Kelly Graf have brand-new ancient DNA results that bear on this question. They sequenced genetic material from the skeleton of a boy that died near Lake Baikal about 24,000 years ago. Turns out that the kid was related to modern Europeans and Amerindians, but not to East Asian (excuse me, Oriental) populations like the Chinese or Japanese. The kid had U mtdna, and an R y-chromosome.
We already knew that the old European hunter-gatherers extended very far to the east, as far as the Urals: looks as if this population extended well into Siberia. Amerindians – the main old original population, all the way down into South America – are a mix of this ancient northern population (about a third) and some Chinesian population.
Amerindian groups show the same relatedness to this Mal’ta culture boy all the way down into South America, so the admixture happened in early days – either before the Paleo-Indians arrived south of the glaciers, or very shortly thereafter. Judging from some anomalous early skeletons, like Kennewick Man, that don’t look at all like East Asians or modern Amerindians, it may have happened in America. Sequencing Kennewick Man would likely answer that question. Looking at the Y-chromosome data, we may be able to get an idea of who conquered who. Judging from a quick and incredibly superficial look, Amerindian Y-chromosomes look closer to some European lineages than to East Asian ones.. while Amerindian mtDNA lineages (except for X) don’t look close to European mtdna lineages.
Then we could talk about Joe Greenberg and Merrit Ruhlen’s impression that Amerindian languages have a very ancient connection with Indo-European,