"What are the prospects for estimating admixture in populations in regions of the world where archaic fossils would not be well preserved, or preserved at all (e.g. the tropics…)?" The prospects are very poor. Attempts have been made in Sri Lanka to retrieve DNA from remains that are only a few thousand years old. No luck. DNA degrades rapidly in tropical environments. "Are there ways to ferret out evidence of admixture without having the genome of an archaic population to refer to?" There are indirect ways, but they all involve assumptions that are open to criticism. Ironically, the best assumption is to use the reconstructed Neanderthal genome as a proxy for the genomes of other archaic populations in Asia and Oceania. These other archaic groups were probably more similar to the Neanderthals than to modern humans.

I'm not sure that heritability is not a statistic for individuals. If a son and a father exhibit the same personality disorder, is it wrong to consider heritability estimates for that disorder if we wish to understand its mode of transmission (genetic vs learning/cultural conditioning). 
 
In particular, I'm thinking here of the case of Marc Lépine, who was responsbile for the mass murder of 14 women twenty years ago. Both he and his father showed similar psychological characteristics, in particular low thresholds for ideation and expression of violence (a 'quick temper' in plain English). The most popular explanation is that Marc inherited these characteristics through social and cultural conditioning, even though his father left his life at the age of 7 and even though Marc hated his father (to the point of changing his name and reproaching his mother for not doing enough to protect him from his father's rage). 
 
To my knowledge, no one has ever suggested that the mode of transmission here was genetic. In fact, most people seem to argue for some kind of early childhood conditioning. Yet heritability for violent behavior in boys is moderate to high when we look at adoption and twin studies. I'm aware of the weaknesses in using heritability estimates. But is this a reason for not mentioning heritability when discussing individual cases, like Marc Lépine?

Razib, 
 
I don't follow your reasoning. Biological difference is not a function of genetic relatedness. You can get major differences between sibling species and minor differences between distantly related species. The relevant factor is difference in adaptive landscape and in attendant selection pressures. 
 
pconroy, 
 
There is no evidence that Neanderthals had tools, such as darning needles or the like, to make tailored clothing. At best, they could have made ponchos, which would have still left much of their body surface exposed. This is in contrast to early modern humans, whose sites show much evidence of tailored clothing. 
 
A second line of evidence is that the human body louse (which has adapted to living in human clothing) seems to go back only 50,000 years in time. I discuss both of these points in the following blog post: 
http://evoandproud.blogspot.com/2007/11/neanderthal-redheads.html

The assertion is reasonable. The distribution of MC1R receptors varies among primate species. Even if the MC1R allele is identical in two species, it may have different effects because the receptors may be distributed differently (see Mundy & Kelly, 2003, Am J Phys Anth, 121, 67-80). 
 
I would also submit that MC1R receptors should be distributed differently between Neanderthals and modern humans. If Neanderthals were furry, cutaneous MC1R receptors would have been subject to different selection pressures.

"is there any doubt that neaderthals were light skinned?" 
 
In a human body, the Neanderthal MC1R allele would lead to red hair and light skin. But it's unclear whether the same would occur in a Neanderthal body. The metabolic pathways for expression of MC1R may have been quite different. 
 
This question may be academic anyway, since Neanderthals probably had little exposed skin, i.e., they were as furry as bears. They almost certainly did not have tailored clothing, unlike early modern Europeans. At best, they might have made ponchos from animal hides, which would have still exposed much of their body surface to sub-zero temperatures. In addition,the human body louse (which lives in clothing) seems to have arisen only 50,000 years ago.

Interesting post. It looks like a rare-color effect, i.e., the scarcer a color variant becomes in a population, the stronger will be the preference for it. 
 
I know of three estimates of hair color distribution in English and white American samples. The discrepancies seem to arise over 1) whether strawberry blondes are classified as blondes or redheads and 2) whether black hair is lumped in with brunette hair or treated as a separate category. 
 
68.1% brunette, 26.8% blond, 5.1% red. (Rich & Cash, 1993)  
68% brunette, 25% blond, 1% red, 6% black (Takeda et al., 2006) 
74% brunette, 18% blond, and 8% red (Mather et al., unpublished) 
 
Whereas the most common variant is brown hair, male preference is strongest for blonde and black-haired women. I suspect preference for red hair also exceeds the actual representation of redheads in the population. 
 
References 
 
Mather, F., Manning, J.T., & Bundred, P.E. (unpublished). 2nd to 4th digit ratio, hair and eye colour in Caucasians: Evidence for blond hair as a correlate of high prenatal oestrogen. 
 
Rich, M.K., & Cash, T.F. (1993). The American image of beauty: Media representations of hair color for four decades. Sex Roles, 29, 113-124. 
 
Takeda, M.B., Helms, M.M., & Romanova, N. (2006). Hair color stereotyping and CEO selection in the United Kingdom. Journal of human behavior in the social environment, 13, 85-99

It is difficult to get good anthropological data on hair color for a number of reasons: 
 
1. Hair color varies continuously, so it is difficult, for instance, to draw the line between 'blonde' and 'light brown'. 
 
2. Many people dye their hair, particularly those who are prematurely gray. Such people may not always tell the investigator. 
 
3. Hair color changes with age. At 1 year of age, 50% of male white children from Louisville, Kentucky had light hair. At 6 years of age, only 26% of the same sample had light hair. (Matheny & Dolan, Am. J. Phys. Anthrop. 42: 53-56). Hair color continues to darken from 6 to 18 years of age (Steggerda, J. Heredity, 32: 402-403). This darkening seems to proceed faster in males than in females. 
 
4. Physical anthropologists have shunned this field of research since WWII. This is partly because it is seen as having Nazi connotations and partly because physical anthropologists became convinced that genetic markers (e.g., blood groups, enzymes, etc.) provide a more scientific picture of human variability. 
 
On the other hand, I've seen a number of studies that peg the percentage of blondes among 'old-stock' Americans and English peoples at around 20-25%. Since this ballpark figure comes up so often, there must be something to it.

"assume that all southern europeans are olive skinned, so 42% could be mediterranean at most." 
 
Wouldn't that be more than enough to create the correlation that appears in the data? Even if the proportion were as low as 20%, you would still have a strong confounding influence from geographic origin. I've seen data sets tossed out for much less.

Razib, 
 
There probably is some correlation between the blue eye allele and skin color. But this correlation is greatly amplified by the use of a White Australian sample where people of Mediterranean origin make up much of the 'olive-skinned' category. There is also the issue of geographic variation among people of British origin. It's been shown that Welsh subjects are significantly darker in color than English subjects and I suspect a similar skin color cline runs from northeast England to southwest England. 
 
In short, we're probably looking at a very slight reduction in skin pigmentation due to blue eye / brown eye heterozygotes. Could this correlation explain the apparently strong selection for blue eyes? I'm skeptical.  
 
The only way to resolve this question would be to extract DNA from skeletal remains from the late ice age, the Mesolithic and the early Neolithic. If blue eyes increased in frequency over a long enough period of time, you may be right, although such a finding would be inconsistent with the hypothesis that the advent of agriculture created this selection pressure.  
 
My feeling is that this selection took place within a very narrow time frame, probably between the glacial maximum and the end of the last ice age (15,000 - 10,000 BP).

Razib, 
 
The correlation between eye color and skin color may simply be an artefact of geographic origin. Europeans vary clinally for both eye color and skin color along a north-south and west-east gradient, so if the pool of subjects is geographically heterogeneous you will almost certainly get a correlation between eye and skin color. But this doesn't prove a cause and effect relationship. 
 
In this study, the subjects were white Australians. The authors state that the "Subjects were overwhelmingly (over 95%) of northern European origin (mainly Anglo-Celtic)." Fine. But even within the British Isles, there is significant geographic variation for both eye and skin color. And even if only 4% of the subjects were of Mediterranean origin (Italian, Greek, Maltese, etc.), they would still have made up one third of all 'olive-skinned' subjects.

The fallacy lies not so much in the data (which do show more variation within than between populations for gene loci taken individually) but in the way people interpret this finding. 
 
I addressed this point in one of my recent posts: 
 
http://evoandproud.blogspot.com/2008/01/85-truism.html

Whether the figure is 99.9% or 99.5%, we are talking about nucleotide sequences, not genes. In theory, every single gene could differ by 0.5% within our species. (I discuss this point in further detail on my blog). 
 
One point bothers me: if the 99.9% estimate needs to be adjusted downwards because copy variation was overlooked, shouldn't there be a similar downward adjustment for the 98.5% estimate of human-chimp genetic identity?

Razib, 
 
Fine, your readers might want to read my essay on sexual dimorphism in skin color at: 
 
http://pages.globetrotter.net/peter_frost61z/fwdm.htm 
 
I also have a post on Judith Harris' theory at: 
 
http://evoandproud.blogspot.com/2007/04/parental-selection-human-hairlessness.html

Cleary, human skin has acquired strong erogeneous properties. This statement may seem obvious, even trite, but it needs to be made. In other primates, pelage is much more critical than skin in stimulating sexual and/or parental responses. Yes, there is evidence from other primates that light pink skin elicits nurturance, but the mental algorithm seems to be limited to adult/infant interactions. It may be that this algorithm has been extended to male/female interactions. 
 
I have trouble with the statement "most people agree that sexual selection takes place on the male." This is a biological trend not a biological law. Sexual selection generally acts on males because paternal investment is usually much lower than maternal investment, so at any one time too many males are competing for access to too few females. 
 
In humans, this trend can be reversed, particularly in environments where women are heavily dependent on men for food provisioning.

P-ter, 
 
You're throwing up a number of methodological criticisms that (a) are at the limits of credulity (b) tend to cancel each other out, and (c) don't fundamentally alter the conclusions of the Hawks et al. study.  
 
You state that the false positive rate could be as high as 20% and that these false positives would be concentrated in more recent times. Fine. But wouldn't the same be true for the false negatives? And even if the false positive rate were as high as 20%, are you arguing that this 20% 'tail' could be wagging the 80% 'dog'? In other words, if there were little or no acceleration among the 80% of true positives, you would have to assume a very high rate of acceleration among the minority of false positives. You would also have to assume a highly skewed bipolar distribution in the rate of genetic change over the human genome.

Consider how they're detecting selection-- they look for alleles that are at high frequency (but not fixed) and have extensive LD around them. Since haplotype length decreases with time, by definition, these are young alleles. Any old allele with a strong selection coefficient has long gone to fixation and is not detected (or more carefully, there's much less power to detect it). P-ter, 
 
There are biases pointing in both directions. Recent selective events would disproportionately involve alleles that are just starting to fall or rise in frequency. This variation would be less likely to emerge above the noise in the data. This bias would favor alleles that are older and clearly responding to selection.

Razib, 
 
No, I didn't check out the Excel document. I knew that people of mixed-race used to be classified as 'black' so I assumed that this was still the case. I was wrong. 
 
Britain never did have the 'one-drop rule' (but neither did most of North America). The usual practice, however, has been to identify anyone with visible black ancestry as 'non-white' or even 'black'. I can think of several 'black' entertainers in Britain who, in fact, are only half-black. 
 
Again, given the volatility of these ethnic categories and given the fact that they have been in place only since 2003, how statistically significant could this improvement in test scores be?

Re mixed-race groups: if you go to the Excel tables linked in the Education Department press release, you will see that the various mixed groups (White/Black Caribbean, etc) are separately itemised, distinct from the 'unmixed' groups. The figures quoted for 'Black Caribbean', 'Black African', etc, should therefore not be much affected by mixture. 
 
Both groups would still become 'whiter' over time, i.e., white/mixed-race offspring would tend to be classified as mixed-race and mixed-race/black offspring would tend to be classified as black. The Excel document does not describe the methodology for ethnic classification, although I suspect it's self-report. 
 
Another point: since the mixed-race category was introduced only in 2003, it is probably unrealistic to infer any trends over time (since the ethnic categories are not comparable).

About half of 'black' children in England have British mothers. This is not the case with South Asian minority groups. Over time, this factor alone would tend to converge the 'black' group to the British mean.

Interesting. I assumed that the blue color allele affected skin color only in the homozygous state. 
 
At the risk of quibbling, wouldn?t some of this linkage be an artefact of geographical association? The authors state that the authors were over 95% of northern European origin, most being Anglo-Celtic. But even in the British Isles, both eye color and skin color vary geographically, with people from Wales and southeast England being noticeably darker. Is this genetic linkage or is it simply a geographical association much like what we see between northern and southern Europe.