Here is an image showing how "race realists" and "anti-race realists" view human biological variation:



The first shows a genetic continua, with relatively abrupt, or sharp breaks between local populations. The second agrees with the first, but adds sharp breaks between each continent (groups of populations as races) and imposes a hierarchical clustering, e.g. each local population falling into a much larger regional grouping (race).

I would consider myself to be a race realist and yet my races can be cut from a continuum. Steve Sailer and Michael Levin, two other self-described realists, allow for the same. Obviously different meanings of "realism" are afloat. For clarity, we can simply call your type "Krom-race realism". I have already noted my problems with this:

1) Historically intraspecific races were not thought this way. Discontinuities evidenced species, continuities races.
2) Presently, many self-proclaimed "race realists" do not have a problem with continuities.
3) Presently, taxa subspecies can be cut from continua.

You add another:

I don't really like the term "race realism" because it implies an ontological position. Instead the race dispute in biology is whether race as a concept is an accurate or productive (useful) way to describe/capture human biological variation, not whether race "exists" or is "real".

I actually employ "realism" in an ontological sense. You do not. What I mean is that "the concept references something in nature". What you mean is that "that which is referenced, at least when it comes to humans, is worth my attention".

Maybe you should call "Krom-race realism" something else. Do you at least apply the formulation consistently? If climatic zones run seamlessly into one another are they not real. Are clines (i.e., character gradients) by definition also not real? Or do races alone need to show "abrupt, or sharp" breaks to be "real"? Why?

This is philosophy not science. The debate about race and 'natural kinds' has no relevance to biology.

I don't care for the "nature kind" nonsense either. It adds no clarity to the discussion. The concept of "natural division", however, does -- since the distinction between natural and artificial division is pretty clear and well understood in biology.

Can you show anything exists?

Yes, if you define "exists" in a way that allows you to show this. I do this in my section I. By this ontology, for example, dinosaurs don't exist because while the concept is biologically coherent no referent is currently to be found.

Would you agree a definition of a race is a meta-population? "a regional group of connected populations [of a species]".

As noted, I distinguish between a general concept/definition and specific ones. My general is: "intra-specific natural divisions". Insofar as "a regional group of connected populations [of a species]" is actually a natural divisions, it would constitute a specific definition of race. As I discussed in section III-A, these types of zoological definitions lend themselves to misreading. Which is why e.g., Mayr and O'Brien attempt to clarify the definition by qualifying which types of populations could be races -- e.g., members must have a shared natural history, a shared set of inherited traits, etc.

What it seems to come down to, is that you see significant-enough "breaks" in genetic/phenotypic variation at the continental level, to justify a racial classification in the sense of Caucasoid/Mongoloid/Negroid, while I am saying these are incredibly weak and do not support the race concept as being useful.

It doesn't seem as if you understood anything at all. While I see breaks on the continental level, I don't require races, in general, to have these -- If I did I wouldn't have a general concept. I made that point numerous times. Krom races seem to just be an exclusive subtype of my general races.

Why I reject your race position, is Linnaean races like "Homo Europaeus" basically correspond to a slope of 1 degrees.

Genetic clusters between continents capture < 2% of variation.

This is obviously far less than what Ehrhard had in mind.

Handley, Lori J. Lawson, et al. (2007). "Going the distance: human population genetics in a clinal world." Trends in Genetics. 23(9): 432-439.

If you wish to continue this discussion comment here. I do not wish to clog up this review thread anymore, especially as you are not providing any thoughtful critiques. (Relocate the comments above which you wish me to address.)

Again, you obviously did not bother to read through my paper. You said, for example, "Linnaean races like "Homo Europaeus" basically correspond to a slope of 1 degrees". As I discussed, Linnaeus did not have a race concept. For him, "Homo Europaeus" was an environmentally induced inconstant variety of the 100% homogeneous in nature species of Homo sapiens. By his systematics, humans were identical by biological nature, so human varieties would have explained 0% of what we would now call "genetic" variation. Try again.

Your arguments all these early writers supported the trivial/minor variation are wrong, so your race concept is completely flawed from the start. If you actually read these early sources properly they set major/great differences between the races, not trivial variation. I could give you 100+ more quotes like above. They are all in This is Race by Earl Count (1950), a compendium on early literature on race.

As I said, my primary concern in the paper is with the intraspecific race concept - not with how human groups were once perceived or even with specific human racial classifications. (This is why my front cover depicts two strawberry races.) The race concept is and was always a pan-species one, one which originally described "constant varieties". Now, as I have said, races were not seen as necessarily entailing large or significant differences. Species often were not either. Thus, there was a species splitter problem, which I discussed -- see: Stamos (2012).

You seem to be confusing historic concepts of intraspecific race with historic perceptions of human race differences. Yet, how human races were perceived is irrelevant to how the intraspecific race concept was understood. As for humans, I did note that monogenists (those who argued that human groups represented intraspecific races, not species) downplayed the magnitudes of differences. Those who argued that the groups represented species, on the other hand, emphasized the size of the differences. Thus, we have your quote from Blumenbach:

"There seems to be so great a difference between the Ethiopian, the white, and the red American, that it is not wonderful, if men of even of great reputation have considered then as forming different species of mankind...."

But Blumenbach doesn't agree that groups represent different species. He emphasizes:

In the first place, then, there is an almost insensible and indefinable transition from the pure white skin of the German lady through the yellow, the red, and the nations, to the Ethiopian of the very deepest black, and we may observe this, as we said just now in the case of stature, in the space of a few degrees of latitude...It is scarce worth while to notice the well-known differences which occurs in the inhabitants of one and the same country, whose skin varies wonderfully in colour, according to the kind of life they lead. The face of the working man or the artisan exposed to the forces of the sun and weather, differs as much from the cheeks of a delicate female, as the man himself does from the dark American, and he again from the Ethiopian.
https://books.google.com/books?id=u9QKAAAAIAAJ&pg=PA105&lpg=PA105&dq=There+seems+to+be+so+great+a+difference+between+the+Ethiopian,+the+white,+and+the+red+American&source=bl&ots=Jwn9DVnpf-&sig=hGaXJ6gKdSJM5s8UAWmrwNp2W54&hl=en&sa=X&ei=b65CVZvCNMuyoQSf24GYAQ&ved=0CCcQ6AEwAg#v=onepage&q=There%20seems%20to%20be%20so%20great%20a%20difference%20between%20the%20Ethiopian%2C%20the%20white%2C%20and%20the%20red%20American&f=false

Now, as for your other points, it's difficult to address them, since they are so conceptually tangled. For example, you say that human continental races only explain 3% (or so) of genetic variance in e.g., high mutation rate microsatellites and imply that this is much less than what was originally thought*. But originally, races would not have been thought about in terms of gene variance since the gene concept had to wait for Mendel -- and no one would have know what simple sequence repeats were. Originally races were thought about in terms of morphological variance (transmitted along genealogical lines). Were morphological differences, generally, perceived as being larger than they are now? Magnitudes were not quantified in the 1700s and early 1800s so we don't really know. I imagine, though, that the morphological differences that were perceived corresponded with the ones that existed. Now, as for the portion of "genetic variance" which continental divisions capture, which is an issue which you seem to be fixated on, this depends on the markers which we are interested in. Insofar as were are interested in phenotypic differences, it would seem to me that we should concern ourselves with SNP variance. As I noted in section IV, this is non-trivial. But I guess that you didn't read that section either.


But, yes, so long.

[*Another point: as discussed, genetic variance at a typical loci -- and so average genetic variance -- will be neutral and so will tend to index time of divergence between populations. It is low in humans because the divergence time is low (for major human races 5 digits). But would early races thinkers have thought it to be high (had they thought in terms of DNA)? If we mean race in the intraspecific sense -- which is what is being discussed -- no, because in this model groups were thought to have very recently diverged from a common stock and spread across the world in the imagined short time since creation (thus divergence time = a few thousand years = <<< Fst). Again, you are not thinking through what I wrote in the paper. Once you grasp that race was equated with constant varieties contra Linnean species/inconstant varieties you will see how untenable the historic (intraspecific) race = deeply divergent groups position is. The case against race is largely based on atrocious historiography.]

I will make public the list of reviewer requests I have sent out. I am primarily emailing researchers who have written about the biological concept.

(1) Adam Hochman -- kindly notified me that he had other commitments (received 4/18)

(2) Michael O. Hardimon -- pending (sent 4/14)

(3) Jeremy Pierce -- No reply (sent 4/08)

(4) Michael Levin -- Accepted, then no reply after (sent 3/13)

(5) Michael Woodley -- Replied, noted he liked it, but said to publish elsewhere (sent 3/13)

(6) Neven Sesardic -- Replied, busy (3/11)

(7) Philosopher Michael James (who wrote the Stanford article on race)-- Politely declined (4/21)

(8) After initially accepting (4/20), Robin Andreasen politely reneged (4/30).

(9) I didn't direly ask Johnathan Kaplan because he told one of my reviewers (3/20) that he didn't want to engage with "racists, nor people who spout racist nonsense while hiding behind mealy-mouthed denials", that the reviewer should have fun hanging out with other racists on websites like ODP/OBG and engaging in "terribly projects", and that I was a
"intellectually dishonest racist". I presumed, based on this, that he would refuse a review request.

(10) Justine Smith -- politely declined (5/1)

(11) Joshua Glasgow -- politely declined (5/1)

(12) Kenan Malik -- politely declined (5/3)

One could make the same kind of statistic argument for US white-black IQ difference. The overlap is quite substantial and most variation is within populations. Does this mean that one should make a Lewontin-style argument for that division too? It works for gender/sex differences too, and pretty much any division below 1.5 d or something like that. There is a saying in philosophy for this kind of argument, "that which proves too much, proves nothing".

Yes, by lewontin's logic, socioeconomic racial differences ( = racism) don't exist, as they typically are around 0.5 SD (i.e., ~7% of the total variance is between ethnoracial groups).

One could make the same kind of statistic argument for US white-black IQ difference. The overlap is quite substantial and most variation is within populations. Does this mean that one should make a Lewontin-style argument for that division too? It works for gender/sex differences too, and pretty much any division below 1.5 d or something like that. There is a saying in philosophy for this kind of argument, "that which proves too much, proves nothing".

Yes, by lewontin's logic, socioeconomic racial differences ( = racism) don't exist, as they typically are around 0.5 SD (i.e., ~7% of the total variance is between ethnoracial groups).

This is a very long paper. I think a lot of people are discouraged by its sheer size. Perhaps crowdfund it and make a payment to the reviewer conditional on receiving a review (no matter whether it's positive or negative).I could act as intermediary.

I will make public the list of reviewer requests I have sent out. I am primarily emailing researchers who have written about the biological concept.

(1) Adam Hochman -- kindly notified me that he had other commitments (received 4/18)

(2) Michael O. Hardimon -- pending (sent 4/14)

(3) Jeremy Pierce -- No reply (sent 4/08)

(4) Michael Levin -- Accepted, then no reply after (sent 3/13)

(5) Michael Woodley -- Replied, noted he liked it, but said to publish elsewhere (sent 3/13)

(6) Neven Sesardic -- Replied, busy (3/11)

(7) Philosopher Michael James (who wrote the Stanford article on race)-- Politely declined (4/21)

(8) After initially accepting (4/20), Robin Andreasen politely reneged (4/30).

(9) I didn't direly ask Johnathan Kaplan because he told one of my reviewers (3/20) that he didn't want to engage with "racists, nor people who spout racist nonsense while hiding behind mealy-mouthed denials", that the reviewer should have fun hanging out with other racists on websites like ODP/OBG and engaging in "terribly projects", and that I was a
"intellectually dishonest racist". I presumed, based on this, that he would refuse a review request.

(10) Justine Smith -- politely declined (5/1)

(11) Joshua Glasgow -- politely declined (5/1)

(12) Kenan Malik -- politely declined (5/3)
[/quote]

(13) Quayshawn Spencer -- politely declined, busy (5/4)

(14) Matthew Kopec -- politely declined, was hesitant given the journal's unestablished reputation (and hereditarian leaning) (5/4)

(15) Koffi Maglo -- politely declined (5/4)

(16) Roberta Millstein -- politely decline (5/5)

This is a very long paper. I think a lot of people are discouraged by its sheer size. Perhaps crowdfund it and make a payment to the reviewer conditional on receiving a review (no matter whether it's positive or negative).I could act as intermediary.

Davide,

I am trying to get the paper reviewed by someone who is actually somewhat familiar with the topic. When that fails -- that is, when I exhaust my twoscore long list of those who have recently written on the race concept (as such) -- I will insist that someone on the OBG editorial board review it, which will possibly mean you. As you reviewed John Harvey's book on race and as you conduct pop. genetic research you are at least qualified to review my paper. If it comes to that, I would be happy to compensate you or whomever for time spent. Maybe 100 Euros or so, since this is what paid reviewers get; there would, of course, be no expectation that you would approve.

I'm fairly familiar with the philosophy of science, population genetics as well as of course the differential psychological/sociological literature, but since we coauthor so often, I'm not really an independent, neutral reviewer.

You can always give an informal critique.

Chuck,

Is there any point in looking for new reviewers? The subject itself is controversial and some criticisms will be motivated by blind hostility. If you look hard enough you may find a prestigious reviewer who is willing to review. But is it worth the trouble? Even if the reviewer is fair and impartial, he or she will focus on certain minor details (because of the sheer length of the text).

It may be useful to have some kind of preface that could orient the reader. What general questions are you trying to resolve? What are the obstacles to resolving them?

Where can I find the latest version of your text?

Chuck,

Is there any point in looking for new reviewers? The subject itself is controversial and some criticisms will be motivated by blind hostility. If you look hard enough you may find a prestigious reviewer who is willing to review. But is it worth the trouble? Even if the reviewer is fair and impartial, he or she will focus on certain minor details (because of the sheer length of the text).

It may be useful to have some kind of preface that could orient the reader. What general questions are you trying to resolve? What are the obstacles to resolving them?

Where can I find the latest version of your text?

Peter,

The latest version is here: https://osf.io/2wsuh/ Some of the formatting screwed up because I switched to word 2007. I added a more detailed abstract, which summarizes the main points and some more discussion in defense of one of my main points -- i.e., early subspecies concepts = genealogically understood constant varieties.

As for this question, "Is there any point in looking for new reviewers?", I was hoping to get confirmation concerning some of the points and clarification concerning some of the arguments which I critiqued -- thus I have been emailing mostly those whom I have critiqued.

As for, "It may be useful to have some kind of preface that could orient the reader. What general questions are you trying to resolve? What are the obstacles to resolving them?"

I think the abstract now makes this clear. Though, I could elaborate more, I guess.


I'm stuck on this question, though -- which I can't seem to get an answer to:

............................................................................

Dear Blah blah blah

I am writing to see if either of you could help with a population genetic question, a clear answer to which has been eluding me.

For context, I am writing a paper on the biological race concept in relations to our species.

I would like to determine, based on genetic differentiation values (e.g., SNP Fst values), what the expected between-group-variance would be for an average quantitative trait owing to neutral divergence.

Would it be:

(a) ~Fst (treating this as an F-ratio that could be converted into mean standardized differences)
(b) ~2*Fst ("")
(c) Something else.

Much of the literature discusses the matter in terms of Qst, defining this as GB/(GB + 2GW). And Qst is said to come out to Fst.

But it's not clear to me if, for diploids (where roughly half of the genetic variance is trapped within individual), the expected quantitative trait F-ratio would be equal to 1* or to 2* Fst.

I stumbled across this discussion by Whitlock (2008) ( "Evolutionary inference from QST")

"Does FST = QST for neutral traits?

The calculation of QST for a trait requires two quantities: the additive genetic variance of the trait within a population (VA,within) and the genetic variance among populations (VG,among). For diploids, QST is calculated as

Qst = GB/(GB + 2GW)

For haploids, the same equation applies, but without the '2' in the denominator. [That '2' for the diploid case comes from the fact that the quantitative genetic variance among populations is proportional to two times FST (Wright 1951).] "

...

It seems to say that the expected between population quant variance would be 2*Fst.

This strikes me as being rather high, though.

To take human continental natural divisions as an example, the Fst SNP value is ~ 0.12.

This would then given F-ratio = 2*0.12 = 0.24 which would be equivalent to an average d-value > 1.00.

From what I recall the typical phenotypic F-ratio, though, is around 0.10 (i.e., in dental and craniometric traits).

Blah blah blah...

.............................

You wouldn't have any idea, would you? I asked Henry Harpending who is more familiar with the Qst literature (inferences of neutral phenotypic divergence from genetic data), but he wasn't sure. I got the basic idea. But I wasn't able to be precisely clear in my discussion because I couldn't resolve this issue -- so I had to talk around it.

(Ya, I understand your apples to oranges point -- I would still like to know what the predicted quantitative trait variation would be due to neutral divergence. See this paper for context: Leinonen, et al. (2013). QST-FST comparisons: evolutionary and ecological insights from genomic heterogeneity.)

I'm stuck on this question, though -- which I can't seem to get an answer to:

............................................................................

Dear Blah blah blah

I am writing to see if either of you could help with a population genetic question, a clear answer to which has been eluding me.

For context, I am writing a paper on the biological race concept in relations to our species.

I would like to determine, based on genetic differentiation values (e.g., SNP Fst values), what the expected between-group-variance would be for an average quantitative trait owing to neutral divergence.

Would it be:

(a) ~Fst (treating this as an F-ratio that could be converted into mean standardized differences)
(b) ~2*Fst ("")
(c) Something else.

Much of the literature discusses the matter in terms of Qst, defining this as GB/(GB + 2GW). And Qst is said to come out to Fst.

But it's not clear to me if, for diploids (where roughly half of the genetic variance is trapped within individual), the expected quantitative trait F-ratio would be equal to 1* or to 2* Fst.

I stumbled across this discussion by Whitlock (2008) ( "Evolutionary inference from QST")

"Does FST = QST for neutral traits?

The calculation of QST for a trait requires two quantities: the additive genetic variance of the trait within a population (VA,within) and the genetic variance among populations (VG,among). For diploids, QST is calculated as

Qst = GB/(GB + 2GW)

For haploids, the same equation applies, but without the '2' in the denominator. [That '2' for the diploid case comes from the fact that the quantitative genetic variance among populations is proportional to two times FST (Wright 1951).] "

...

It seems to say that the expected between population quant variance would be 2*Fst.

This strikes me as being rather high, though.

To take human continental natural divisions as an example, the Fst SNP value is ~ 0.12.

This would then given F-ratio = 2*0.12 = 0.24 which would be equivalent to an average d-value > 1.00.

From what I recall the typical phenotypic F-ratio, though, is around 0.10 (i.e., in dental and craniometric traits).

Blah blah blah...

.............................

You wouldn't have any idea, would you? I asked Henry Harpending who is more familiar with the Qst literature (inferences of neutral phenotypic divergence from genetic data), but he wasn't sure. I got the basic idea. But I wasn't able to be precisely clear in my discussion because I couldn't resolve this issue -- so I had to talk around it.

(Ya, I understand your apples to oranges point -- I would still like to know what the predicted quantitative trait variation would be due to neutral divergence. See this paper for context: Leinonen, et al. (2013). QST-FST comparisons: evolutionary and ecological insights from genomic heterogeneity.)

First of all, the set of 315 Fst values that I calculated using VCFtools (which employs Weir and Cockeram Fst formula) on 1000 Genomes phase 3 data for 26 populations can be seen here (https://docs.google.com/spreadsheets/d/1n-C061ZAVCjtN_D9RZLCZJYuur-DTSIVP2xUx1HCA2w/edit?usp=sharing ). I report Fst for 1st and 21st chromosomes (columns C and D). They are practically identical (r=0.995) so either can be used to represent the whole genome. Note that these include SNPs and indels. If you use these Fst values in your paper, please cite my last article (http://dx.doi.org/10.6084/m9.figshare.1393160 ) because they are in the supplementary material there.

THERE IS INDEED MUCH CONFUSION ON INTERPRETING FST AS RELATIVE BETWEEN POPULATION VARIANCE.
It appears that the expected BETWEEN population variance should be 2*Fst, after correcting for the inbreeding coefficient.

Sarich and Miele (2004) write "Lewontin had noted that 85% of the genetic variability was among individuals within populations, and only an additional 15% was added when individuals in different populations were compared [...]. The point is that we are diploid organisms, getting one set of chromosomes from one parent and a second from the other. To the extent that your mother and father are not especially closely related, then, those two sets of chromosomes will come close to being a random sample of the chromosomes in your population. And the sets present in some randomly chosen member of yours will also be about as different from your two sets as they are from one another. So how much of the variability will be distributed where?
First is the 15% that is interpopulational. The other 85% will then split half and half (42.5%) between the intra- and interindividual within-population comparisons. The increase in variability in between-population comparisons is thus 15% against the 42.5% that is between individual within-population. Thus, 15/42.5=32.5%... "

So your hunch was right in telling you that Fst underestimates the between-population comparisons by about a factor of 2 (in reality it's a little less than 2 if we account for inbreeding).

Sarich, V. and Miele, F. (2004). Race. The reality of human differences. pp 168-169.

"I would like to determine, based on genetic differentiation values (e.g., SNP Fst values), what the expected between-group-variance would be for an average quantitative trait owing to neutral divergence."

It would be a bit less than Lewontin's figure of 15%. Selective value varies considerably among genes, with most being close to selective neutrality.

I never really understood the relevance of Henry's criticism of Lewontin's model (i.e., a large chunk of within-group genetic differences are actually within-individual). This criticism would presumably apply to all diploid species, so Fst would have to be adjusted for all diploid species. Yes, between-group variability would look larger for humans, but it would also be proportionately larger for all diploid species.

Fst has acquired an aura of respectability among population geneticists, yet it doesn't tell us half of what we think it does.

I like your abstract. I still disagree with your positive assessment of pre-Darwinian anthropology. Before Darwin, anthropologists had a lot of false premises that prevented them from thinking clearly (young Earth creationism, belief in the flood and Noah's ark, inheritance of acquired characteristics, belief in pre-Adamite races, etc.)

It would be a bit less than Lewontin's figure of 15%. Selective value varies considerably among genes, with most being close to selective neutrality.

I never really understood the relevance of Henry's criticism of Lewontin's model (i.e., a large chunk of within-group genetic differences are actually within-individual). This criticism would presumably apply to all diploid species, so Fst would have to be adjusted for all diploid species. Yes, between-group variability would look larger for humans, but it would also be proportionately larger for all diploid species.



Fst has acquired an aura of respectability among population geneticists, yet it doesn't tell us half of what we think it does.

[Edited for redundant babbling.]

Peter,

I agree with the last point. As for the first, it's a basic quantitative question for which I am receiving conflicting answers. For example, Davide said ~ 2*Fst and you said ~ Fst. Well, both of you can not be correct.

But why does it matter? For one, I just like to get straight answers. For another, it's relevant to the debate since one of the arguments against the significance of major human races is that between group variance is small relative to that between individuals. This is a valid argument since we are often interested in between group to between individual comparisons -- for example, this is typically how we compare differences in the social sciences (e.g., standardized ones). We don't -- I don't at any rate -- typical compare behavioral, morphological, and outcome differences between e.g., Han and Hausa to that between horses and humans, rather we compare it to that between individuals within populations.

Of course, I agree that the precise relation between Fst and quantitative variance, specifically the question I am asking, is irrelevant to intra-species comparisons. I do discuss such comparisons in one of my sections. And I note the difficulties and intricacies with them, especially in regards to measures of divergence. I could have elaborated more and discussed the points which you have made e.g., regarding some obscure deer super-species. But the inter-species stuff doesn't really garner my interest.

Davide,

An equation I have come across for diploids is 2*Fst/(1-Fst) -- which is supposedly derivable from Wright's 1950 equations. See box 1 here. For haploids it is Fst/(1-Fst). I can't make sense of the math, though -- so I want verification from someone who can. (The relation between quantitative trait variance -- i.e., in the sense of F-ratios -- and genetic variance is not directly discussed, rather Qst and Fst is and derivations of Qst are only noted in passing.) I attached some papers on the issue.

It's really a somewhat trivial matter -- I'm just having trouble getting a clear, definitive answer.

First of all, the set of 315 Fst values that I calculated using VCFtools (which employs Weir and Cockeram Fst formula) on 1000 Genomes phase 3 data for 26 populations can be seen here (https://docs.google.com/spreadsheets/d/1n-C061ZAVCjtN_D9RZLCZJYuur-DTSIVP2xUx1HCA2w/edit?usp=sharing ). I report Fst for 1st and 21st chromosomes (columns C and D). They are practically identical (r=0.995) so either can be used to represent the whole genome. Note that these include SNPs and indels. If you use these Fst values in your paper, please cite my last article (http://dx.doi.org/10.6084/m9.figshare.1393160 ) because they are in the supplementary material there.

THERE IS INDEED MUCH CONFUSION ON INTERPRETING FST AS RELATIVE BETWEEN POPULATION VARIANCE.
It appears that the expected BETWEEN population variance should be 2*Fst, after correcting for the inbreeding coefficient.

Davide,

Would it be possible for you to partition global variance into between continental race, between individual within race, and within individual variance?

See table 4 here for an example.

"To measure the differentiation between populations, the widely used statistic FST [17] and its unbiased estimator [18] were used. FST estimates were averaged over all loci, and 95% confidence intervals (CIs) of the average FST were calculated by bootstrap resampling with 10000 replications...Along with FST, variance components were estimated to reflect intra-individual, inter-individual and inter-population differences in genetic variation."

There appear to be programs which allow for this -- but no one does it. If you need, I will write Nishiyama et al. regarding method/statistical program.

Also, link rot: http://dx.doi.org/10.6084/m9.figshare.1393160