[expat_panama]

 -- someone in Eurasia 3500 years ago couldn't be the ancestor of every Native American...

The point Hopkin is highlighting is that since (as we agree) that it's easy to believe that from say, 1500BC to 500BC that at least one guy visited the Americas from Asia, and shared genes.  We don't have photos but it's simply not reasonable to say it didn't happen.

Since then, 2,500 / 20 = 125 generations have passed.   Remember that this is 125 doublings -- a factor of some 40 digit number.  This would spread the guy's genes to every single human in the Americas.  This isn't hard to believe considering that it took about that long for the first Eurasians to cover that area.   The only way that his family tree would not have spread would be that maybe the guy had some kind of genetic condition where everyone in his family dies at birth (rim shot), or maybe him and the little woman moved to the moon (cow bell).

But seriously folks, not to worry-- I promise not to take advantage of our newly discovered family ties in order to borrow money.

[SunkenCiv]

Not reasonable to not believe it? Uh, no. It's not only reasonable to not believe it, but what these two authors are saying is simply not possible. As in impossible. And I'm not an isolationist, not at all. They're saying that everyone in Africa -- everyone -- and everyone in the Amazon rain forest, no matter how isolated -- everyone -- has a common ancestor who lived in Asia 3500 years ago.

Ancestral lines can be quite narrow. With (usually) 46 chromosomes, once our great-great-great-great-grandparents are considered -- 64 people -- only 46 of them (at most) have contributed anything to our makeup. With our (usually) 23 pairs, a mating pair of humans can produce 2 ^ 23 possible combinations, in theory not in practice. Isolated populations can be remain viable, with a quite narrow ancestry on a genealogy chart. And I'm a multiregionalism advocate.

The study isn't scientific because it isn't based on actual evidence of any kind, nor is it falsifiable because there's nothing to falsify. Within the past couple of days a post-Olympics study of sorts was quoted on FR, which claimed that in 150 years, women will have faster sprint times than men, based on the curve of improvement in finish times. That doesn't take into account physiological differences (it also played fast and loose with the data). This "study" is no more reliable.

George W. Bush will be reelected by a margin of at least ten per cent

[Keith Pickering]

it's easy to believe that from say, 1500BC to 500BC that at least one guy visited the Americas from Asia, and shared genes. We don't have photos but it's simply not reasonable to say it didn't happen.

Since then, 2,500 / 20 = 125 generations have passed. Remember that this is 125 doublings -- a factor of some 40 digit number. This would spread the guy's genes to every single human in the Americas.

Not at all necesarily true. Among my family, in my generation, two out of three siblings have no children. In my parent's generation, (father's side) one out of two have no children; mother's side, also one out of two have no children. So right now, my four grandparents have a combined total of ONE great-grandchild between them (and probably won't have any more, either.) That's three generations removed. If that kid becomes a priest, or dies young, all four grandparents will be out of the descendents game entirely.

This kind of thing happens a lot, especially in societies with high infant mortality. Which means most of the human race throughout most of its history.