Red hair going extinct???

Every few years it seems that a new meme declares that "blondes will go extinct!" or that "red hair will go extinct!" I've only been blogging for 5 years, and this story has already cycled multiple times. A co-blogger of mine told me that he did some digging and it seems that this meme is of old vintage, with "blondes going extinct!" stories dating back to the 19th century. The current craze (as evidenced by blogs) seems to have started at an Australian newspaper. But, it is sourced originally to National Geographic Magazine.

First, the story doesn't appear on National Geographic Magazine's website that I can tell. Perhaps it is in the print issue? A reader who has a copy of the current issue might want to post their finding in the comments (I will probably go the bookstore tomorrow and check myself). Let's assume that the story is correct. What are they actually trying to say here? They are actually just restating inferences derived from the Hardy-Weinberg Equilibrium in a panmictic population; this isn't really a "discovery." Let me clarify what I mean.

1. Red hair seems to be correlated in large part with two loss of function mutations on the MC1R locus.
2. This means that to a first approximation one can conceive of it is a recessive Mendelian trait.
3. A population in Hardy-Weinberg Equilibrium (HWE) is governed by the relation: p² + 2pq + q² = 1, assuming a diaellelic locus where random mating, migration, mutation, selection and drift are not operative dynamics.

If one assumes that q² represents the proportion which is homozygous for an allele which only expresses when within that genotype, then one can intuit in HWE as the frequency of q decreases a greater and greater proportion of that allele will be found paired with the p allele (since p is simply the balance left over from the proportion of q). The stories commonly report that only 1-2% of the world's population are redheads currently, so one can imagine that q is already extant at low frequencies. In reality it seems that more than one allele results in loss of function, but in terms of phenotypic effect they are equivalent.

Since the frequency of these alleles differ by population there is a strong deviation from random mating; e.g., the typical Scot marries another Scot, who are already likely to have a much greater chance of carrying the loss of function allele than a non-Scot. As intermarriage across nationality's occurs the proportion of loss of function alleles masked in heterozygote states should increase, so the realized proportion of the redhair phenotype will approach q² in direction proportion to how close to random mating we are. I am very skeptical that the National Geographic Magazine article, if it exists, presented a detailed demographic model so that one could posit that red hair became "extinct" at a particular date. In fact, I strongly suspect that anyone who says that red hair (or blonde hair) is going extinct is in need of a few envelopes or napkins to do some basic algebra, or at least stipulate what low proportion is "close enough" for them to consider a trait "extinct." Finally, as I noted last year, part of the appeal of the stories is probably the disquiet of white people at the "passing of the great race." The reality is that I don't think that the white race is going away anytime soon, but surely a non-trivial proportion of whites are concerned at some level that the colored folk are swallowing them up, and the blonde and red hair stories are simply subtle manifestations of that. These traits are after all amongst the most distinctive of the northern European peoples, and so they can be emblematic of the threat posed by declining birth rates and miscegenation.

Addendum: For regular readers, I am aware that the technical exposition was retarded. That's for Google, not for you my wise souls!