> The fact that the races end up living in different areas is strong evidence that when put in close proximity, different races don't want to interbreed. That'

> The fact that the races end up living in different areas is strong evidence that when put in close proximity, different races don't want to interbreed. That's true, but (1) that's not what I was talking about; I was talking about geographic inhomogeneity in original migration patterns and (2) this argument applies to ethnic/language/religious groups in general, not just races. More on this below. > That means they all started out living in white areas. Sort of. https://en.wikipedia.org/wiki/List_of_U.S._states_by_historical_population https://en.wikipedia.org/wiki/List_of_U.S._states_by_histori... shows that in 1850 South Carolina had 668k people of whom 385k were enslaved (hence not choosing where they lived, exactly). I didn't look up the number of free blacks, but I expect it was quite small given the laws about re-enslavement at the time. Anyway, more like the white people were living in a black area there. ;) But the real point is that for a while there various parts of the country did _not_ have appreciable black populations because slaves had not been brought there while your typical plantation had a much larger black population than white population. Voila, _involuntary_ geographic inhomogeneity. > The reason for segregation is not relevant That really depends on the question you're trying to answer. > We'd have to compare this to how white men behave when they have white women as slaves Sure. But the worst case for the "enthusiastic" hypothesis is that in this situation they always end up with kids, right? There may be value in comparing to this worst case, even. > There's also no doubt that it's far less than would occur if there were no preference one way or the other for interracial breeding Sure. I don't think anyone reasonable is arguing there was no preference on the matter. But if we're going to talk about sub-species, exact frequencies and cutoffs do matter if the whole thing is to make sense other than as a mood-affiliation labeling exercise ("_these_ two groups are different sub-species, but these two other groups are just less likely than average to intermarry/interbreed"). There is good data showing that breeding across religious and language boundaries is less common (and was even less common in the past) than breeding within those boundaries. And yet very few people nowadays will argue that, say, the French and the Germans are separate sub-species, or were separate sub-species 100 years ago. Yes, I realize that 100 years ago people were in fact likely to make just such an argument. :) Heck, inter-political-party marriage is a lot less common than within-political-party marriage in the US. And the rate of inter-political-party marriage, in the US, is dropping, while the rate of interracial marriage is rising. According to http://fivethirtyeight.com/features/how-many-republicans-marry-democrats/ http://fivethirtyeight.com/features/how-many-republicans-mar... the former is sill more common than the latter but not _that_ much more common. I expect the lines to intersect within the next decade or two if current trends continue... I guess we could argue that sub-species boundaries are malleable and they can merge or separate over a timescale of a few generations. That would be a valid take on the whole thing, I guess, but then I question the usefulness of the "sub-species" labeling; it seems to not have a useful definition in this situation from my point of view.