(urth) Mystery of Ascia

[Gerry Quinn]

From: "Lee Berman" <severiansola at hotmail.com>
>
> Gerry we are a bit closer to understanding on these issues. I 
> misunderstood what you were
> getting at with "natural country type". I thought you meant there could be 
> no prevalent
> variation of a gene in a certain region. Now I understand you are arguing 
> that there is more
> than one possible adaptation that can help a population to live in a given 
> environment. And
> of course that is true. Large body size, Hunter's response, brown adipose 
> tissue, epicanthic
> folds, large nose, light skin; all these have been argued as cold weather 
> adaptations in humans
> and a given population in a cold environment can have any of them (though 
> I don't know of any
> which have all of them in high frequency).

Exactly.  Different peoples migrating to a country would have different 
genetic capacities for such adaptations; combined with random chance this 
would result in various possibilities, rather than a specific type that 
would be forced by the conditions of a particular country.


> A couple misundertandings on your part:
>
>> There isn't such a thing as a "Swedish version" of a gene. While gene 
>> pools are mercurial, genes are not.
>> Changing one single base code among tens of millions on the gene for 
>> insulin is what makes people diabetic.
>> Genes are copied faithfully person to person. Mutations are almost always 
>> harmful
>
>>Not true.  Here, for example, is Wikipedia on the subject of alleles:"It 
>>is now appreciated that most
>>or all gene loci are highly polymorphic, with multiple alleles, whose 
>>frequencies vary from population
>>to population, and that a great deal of genetic variation is hidden in the 
>>form of alleles
>>that do not produce obvious phenotypic differences."
>
> First it should be understood that "allele" is not a synonym for "gene". 
> An allele is defined as a gene
> variant. Second it should be understood that "phenotype" refers to the 
> outward expression of a gene.

An allele is a gene variant, yes, originating presumably in a mutation. 
Alleles vary from population to population.  Alleles which are frequent in 
Sweden but rare elsewhere can reasonably be described as Swedish versions of 
genes.


> So, Wikipedia is saying that there can be a lot of gene variation where 
> the difference doesn't DO
> anything. This relates to the idea of "junk DNA". If a gene is not 
> expressed, it it not subject to
> natural selection. It can just sit there and be copied or mutated from 
> generation to generation and it
> doesn't matter.

Wikipedia is saying nothing of the sort.  They are not talking about junk 
DNA.


> With certain traits which don't have a selection pressure on them, like 
> eye color or blood type, there
> are various alleles like blue/brown or A-B-O. But for stuff that matters, 
> like the code for insulin or
> the code for having 5 fingers alternate alleles (diabetes and 6 fingers) 
> are selected against and are
> rare.

What makes you think there is no selection pressure on blood types or eye 
colour?  As for insulin, the insulion molecule itself is a small peptide 
which seems to be highly conserved, but there are certainly common allele 
variants in regulatory genes related to it.  (Incidentally, the prevalence 
of these varies with race.)


> But this is not what we are talking about. We are talking about high 
> frequency genes which have obvious
> phenotypic effects and which are subject to selection and, as a result, 
> show regional variation in
> frequency like skin color (btw, there is probably no adaptive advantage to 
> blue eyes and blonde hair;
> they are just accidental by-products of the gene for low melanin 
> production in the dermis). There is no
> "African" melanin or "Swedish" melanin" or "Turkish" melanin or whatever. 
> It is the same pigment in all
> humans (and many other animals as well). There is no irreversible variant 
> of melanin that will make one
> population of humans forever different than others.

First, you are leaping from "no obvious phenotypic effects" to "no 
phenotypic effects".  Probably many if not most alleles have subtle 
phenotypic effects.

Second, melanin is emphatically *not* a single molecule - it is a complex 
polymer with thousands of variations.  There are three main types in humans. 
It seems reasonable to assume that the exact type produced will vary 
accordsing to allele combinations of melanin-related genes.

I don't know what you mean by "irreversible variant of melanin".


> Anyway, those are just side issues. The main issue you are not getting is 
> that there can be no inherent
> "irreversibility" of gene frequency within a species. By definition it is 
> not possible because
> the potential for admixture exists. With admixture and selection, any 
> particular gene in a regional
> population of a species can go from 0% to 100% and back to 0% as long as 
> the gene exists somewhere
> on earth for that species.

That's irrelevant.  The issue is whether a particular *constellation* of 
alleles can be replicated with finite probability.


> You can't argue that artificially isolating a group of Swedes and a group 
> of Africans in a cage for
> infinity years constitutes an example of Dollo's Law of irreversibility. 
> It makes no sense.

I *already stated* that I was incorrect to describe the form of 
evolutionarey irreversibility I am talking about in terms of Dollo's Law.


> A scientific
> law must describe what happens in nature. It must have an element of 
> predictability. For your experiment
> to create irreversibility your caged individuals must become different 
> species. So different that there is
> never a possibility of interbreeding. This has already been discussed and 
> it must be understood or there is
> no point in discussion.

They do *not* have to become different species or be unable to interbreed to 
create irreversibility of the sort I am describing.  I agree that a single 
allele if strongly selected for could go from around 0% to around100% and 
back again.  That does not mean that is is statistically feasible to recover 
a particular constellation of traits controlled by many genes all with 
variant alleles.  If you put a tribe of Swedes or Yoruba in a cage and 
subject them to selective pressures so that they adapt to their new 
environment, then even if you change the conditions back to those of Sweden 
or Yorubaland, the original allele frequencies and the constellations of 
traits they govern will not be recoverable.  They will never again be 
exactly like the original Swedes or Yoruba.  This is a form of evolutionary 
irreversibility.


> I get the impression you were hoping Dollo's Law might make it impossible 
> for lighter skinned people to
> ever look like their African ancestors even if global warming turns the 
> whole planet into a tropical
> rain forest or something like that. It just isn't the case. As long as we 
> remain one, potentially
> interbreeding species, it is possible for the whole of humanity to end up 
> falling anywhere in the range of
> the spectrum of variability. We could all become light skinned, all dark 
> skinned, all curly haired, all
> straight haired, all blue eyed, all brown eyed, etc.

How many times do I have to repeat that I am not talking about Dollo's Law? 
Do you even read the posts you respond to?  But yes, no current people will 
ever come to look exactly like the African ancestors of humans.  While any 
single gene could revert to ancestral forms, it is a statistical 
impossibility for large numbers of genes to revert together.  Thus, a form 
of irreversibility.

- Gerry Quinn