By Peter D'Adamo
Orinally published in: 
One of the great chin-scratchers of modern physical anthropology revolves around blood type, in particular why most indigenous populations of the New World have such incredibly high percentages of the gene for type O. Sometimes, especially as you move south of the modern US-Mexico border, the percentages almost reach 100%.
Since almost everyone agrees that human habitation of the New World began with migrations out of the Siberia, across the Bering Sea, and the population on the Russian Asiatic side shows no similar high percentage of type O; if anything the percentage frequency of the type O gene drops as we move further and further north and east. Several theories have been advanced to explain the apparent 'Bering Straits Bottleneck'.
The most often suggested is the genetic drift theory. The basic idea behind genetic drift is easy enough to understand. If you flip a coin two hundred times, there is a very good chance that your results will be somewhere close to 100 times coming up heads, and another 100 times coming up tails. Indeed, the more you flip a coin, the more likely (given that you have an honest coin) the results will be 50% head and 50% tails.
However, suppose that you instead only flipped the coin seven times; would it not be feasible on any given Sunday to flip five heads and two tails? Sure it is. That is how Las Vegas stays in business. Genetic drift is like that: A small population may have an uncharacteristic gene distribution simply because the genetic coin did not flip enough to have things even out.
So the Genetic Drift Theory of the 'Bering Bottleneck Type O Anomaly' posits that a small band of folks swam, walked or boated over the Bering Strait, and because their numbers were so small, the genes for A and B did not come along with the coin flip. This small number of colonist determined the future gene pool for the continent due to their exerting a 'founder effect'.
It's not a bad theory, except that in order to accomplish this, the numbers of Asian immigrants to the New World must be very small; along the order of a dozen or less, so that there is an even slight statistical chance that they could all be type O. However, even if the original colonizers of the New World numbered, say ten or eleven, the odds of those entire ten or eleven colonist being type O is about one in a thousand. Even if the number of colonists is dropped to five the odds only drop to one in thirty-two. () And that also assumes that there was one boatload or band of colonists, when common sense tells us that there must have been numerous attempts, though perhaps not all successful, to migrate to the New World.
The second theory is that of natural selection, which a lot of people equate with evolution, but it's not. Natural selection posits that perhaps a mixture of all blood types were part of the original migration, but for some reason, probably infectious disease, the type A and type B colonists died out. Of the two, Natural Selection is perhaps the stronger theory since there a definite likes between ABO type and susceptibility to small pox, syphilis, E. coli and tuberculosis, all of which probably killed lots of people back then.
However, as any honest exterminator will tell you, it's hard to kill them all.
A.E. Mourant addressed this issue in his book Blood Relations
"Like the absence of B in the Australian aborigines, the lack of B in the northern zone and of A and B in the southern zone raises a problem of world-wide importance. Was the B gene totally absent from the original populations from eastern Asia that ultimately reached Australia and America, or was the gene lost on the way? If so, was this due to genetic drift in relatively small isolated populations, or to natural selection? Early blood-group workers suggested that when man left Asia for Australia and America mutations for the A and B genes had not yet occurred. However, analogous if not identical genes occur in the higher apes at least, and so are several million years old. In the light of the discussion of O frequencies in Europe it is not difficult to see how, as a result of the elimination of A and B fetuses of O mothers, first the gene B (which is rarer than A) could have tended to disappear, and then A itself."()
Now, it has been know for a while () that human and primate ABO genes are somewhat analogous, let's just say that they are similar enough for our purposes, which is to say that the individual genes for A, B and (by default) O are 'old'. However, does it automatically lead us to assume that just because genes share a long history, does that mean we can assume that they will always exist in the percentage numbers? Of course not, we just say that with Genetic Drift: percentages change.
With apologies to Edward Tufte, let's take a look at the following graphic:
What you are looking at is the northeast corner of Asia and the northwest corner of North America at the Bering Strait, across and under-which one day your kids may be able to drive their cars. Not surprisingly, the colors of the map mean things: For example, the darker green the land is colored, the higher the frequency of the gene for type O; the lighter the color, the lower the percentage (less type O genes)
Now, first of all, note that these are indigenous populations, so the modern-day Alaskans and Siberians don't figure here that much here. What sticks out at you? Yup, there is lots of O gene the further east (the right side of ther map) you travel! But what else? Normally we might expect the trail of O genes to drift nicely along, but in our map the distribution is bi-modal: The incidence of O gene is higher at both ends of the map and lower in the middle. You can see that by looking at the bar graphs below, which not only looks at the relative 'percentage if each percentage' but also the percentage of land versus water: Each bar graph is actually a snapshot of one of the sixteen 'slices' of the map, the black lines.
So if anything, the more constricted that land mass became, the less you find the type O gene.
Interestingly, look at the red numbers on the map. They are the percentages of type B gene. Notice as well that the Asian side of the Strait has some of the highest percentage of Type B gene on the planet. What about on the American side?
Virtually no type B gene.
Now, to me this implies that there may well have been two waves, a 'First Wave'� that contained very high percentages of type O gene and which had a relatively easy time getting across the Bering Strait (which may well have still been a land bridge) and who created the 'Founder Effect' in America, and a 'Second Wave' somewhat higher in Type A and much higher in Type B which followed but got stymied by the ecological changes and the closing of the land bridge.
So what I think is that both the Genetic Drift and the Natural Selection theories are correct, but I'm more inclined to move both of their occurrences with regard to blood type further back in time and much further west. In that case, rather than having crossed before the advent of the genes for A and B, our first American colonists would have walked across before the rest of Asia had a chance to recover from the results of its own initial 'flip of the coin.'
By which time there was no more walking there.