You do realize that inter-racial marriage evens out initial racial differences to a great degree, even in racially-stratified society, right? This argument doesn't make sense.

+1. In the presence of intramarriage, this is perfectly possible and almost certainly what the writers are going for, in light of Valorum's answer. Einpoklum should not be downvoting answers just because they disagree with the way the story is written.

@einpoklum On a timeframe of a few thousands years ? Likely not. Genetics don't work by averaging characteristics from both parents. Biracial families can have a lot of different skin tones, sometimes very contasted, even between siblings. And even if 3500 years were enough to select a somewhat local homogeneity, nothing guarantees the differences would appear on skin color or other 'racial' characteristics as we know them today.

@einpoklum Also please note that until now, the show featured cities and locales from the same continent. The lack of 'ethnic' contrast between Tar Valon and the Two Rivers is consistent with what you would see between Paris and Rome in medieval times. The contrast is much more apparent with the Aiel, as they indeed have had very few mixed lineages with the people of the wetlands since the Breaking. And the show and books are both giving a clear 'racial' signifier: red hair.

@Vallahga - Also, while I am not specifically an expert on genetics, I think einpoklum's reasoning may not be correct. If physically notable recessive traits such as red or blond hair and blue eyes can remain in populations for millenia (and they have) without either disappearing or dominating, should we automatically assume that the alleles responsible for characteristics like skin color would simply "even out," even in the absence of high intramarriage within ethnic groups? Maybe they would, but it does not seem like an obvious assumption.

In other words, as with hair color, I am not convinced that "homogenous distribution of allele frequency across geographic regions" is at all equivalent to "identical alleles between any two given individuals."

I've read the whole series (although not the extra material), and I don't remember anything remotely like the minor spoiler you mentioned. Perhaps I missed it? Could you tell me where that was?

@Adamant mostly it does. Bear in mind, almost everything we arbitrarily file as an "ethnic group" today is an admixture of two or more distinct groups that existed as recently as a few thousand years back in our history. Not many places are that isolated and stable.

@hobbs - Of course, ethnic groups are largely arbitrary. However, if the alleles responsible for things like blond hair can continue existing over the course of millenia, I am not sure why you would believe that individual phenotypic variation across a wide range of features would disappear over a mere few thousand years. Would modern ethnic groups be recognizable? Probably not (but you still need someone to play the characters). Would there still be lots of variation in characteristics like skin tone and hair color? I think so.

@Adamant there would be plenty of variation... between different places. Substantially less in a given place. Some places with as many blondes as Minnesota, some with as many blondes as China.

@Adamant to learn more see bit-player.org/2020/questions-about-trees particularly the bits about how "neutral drift" can still result in plenty of homogeneous clumps.

@hobbs - It doesn't sound like you actually are disagreeing with me. That's exactly what Valorum's answer suggests to be the case: there is phenotypic diversity across the world, but some places tend to have phenotypic differences.

Also, while they are interesting and thought-provoking, I would be careful extrapolating with those simulations. The long-time behavior of a system with parameters chosen to represent trees may or may not say much about the short-term behavior of other species.

@Adamant Valorum's answer is fine. I'm discussing this answer, in light of what the question is actually asking. OP says that regional variations are to be expected, while major diversity within a single city is weird. This answer says "no, because it was all mixed together in the past of the setting, and genetics would keep it that way". But realistic genetics does not keep it that way after 100 generations of reduced mobility (I know people aren't trees, but history provides plenty of corroboration).

@hobbs - Can you really say that, though? I would be prepared to say that perhaps the theory in that blog post is the correct way of describing interactions between trees. It's basic, but I have not had time to examine its assumptions. But how relevant are the assumptions to other species? If they are relevant, how do timescales change? That's not clear. As for what you said, how many examples have we really had throughout history of large populations with much more diversity than the modern United States becoming largely phenotypically homogeneous within X number of generations?

It's difficult for me to answer those questions easily, because I did not specialize in that particular area of biophysics/mathematical biology. I do have an ex-colleague whose research is precisely in mathematical population ecology, and perhaps as early as tomorrow I can ask what their hypothesis is as to what would occur.

@Vilx- see here and here at the WoTwiki.

@randomhead Uhh... Yeah, I remember that part, and I see the parallels, but... it doesn't seem like more than speculation to me. Is there an actual authoritative source for these statements or is that just a fan theory?

@Vilx- You can see a relevant comment from Robert Jordan here: theoryland.com/intvmain.php?i=67 (40th question).

@Vallahga - Ahh! Now that IS something! Thank you! :)

@kuhl - With RJs confirmation that this really IS our world, I'm ready to accept these speculations. :)