Yes you DO get loss of genetic diversity with evolution Part 8

NWR on that thread:

Faith writes:
My argument is that natural selection and genetic drift, all the processes that select or isolate a portion of a population, do bring about the change called evolution but also always reduce genetic variability, which is the opposite of what evolution needs.

That's not quite right. Yes, selection reduces variation.

If this is typical and I suspect it is, I've been consistent in saying "genetic variability" but those who argue with me keep wrongly saying "variation" which obscures my point. It's genetic variability that is reduced in the making of a new phenotype, and it's the reduction of genetic variability that eventually brings about the end of a particular line of variation because it runs out of allelic possibilities. Up to that point you can still get further variation of phenotypes by further selection events or just by migration of a small proportion of a population.

As far as I know, genetic drift does not affect variation. And mutation increases variation.

Genetic drift is just one of the ways that a new gene pool is formed and reproductively isolated from a larger gene pool, which develops new phenotypes while reducing genetic variability. And again, I'm talking about GENETIC VARIABILITY OR GENETIC DIVERSITY, not "variation."

And here it comes, the savior of evolution, MUTATION. The ONLY way this inexorable reduction I'm talking about could possibly be prevented -- though only in theory because mutation couldn't succeed at that even if it occurred. Any process that kept putting new alleles into the gene pool would only interfere with the development of the new species that is the predictable result of reproductive isolation of smaller gene pools. You couldn't ever get "speciation" at all, you couldn't get the distinctive variations in the ring species.

You are correct, that if there were only processes that reduce variation, then eventually evolution would run out of variation and would stop. But as long as there are also processes that increase variation, there is no reason to expect evolution to stop.

Actually, this could be taken as a concession of my point and end the discussion with me the winner. At least he does acknowledge that I've got the main process right. "Processes that increase variation" come down to only ONE purely theoretical process, mutation, and as I point out in that thread and here as well, mutation 1) doesn't create viable alleles, and 2) if it did it would prevent the formation of species altogether BECAUSE THE FORMATION OF SPECIES / VARIATIONS DEPENDS ON THE REDUCTION OR ELIMINATION OF ALLELES IN THE POPULATION.

As far as I know, what is mostly noticed is that variation stays fairly constant.

Variation or genetic variability? Phenotypes DO remain fairly constant, the processes that are called evolution that lead to new variations aren't continuously occurring in any observable degree and in that case the genetic variability would also stay constant. But when you DO have evolution, when you are getting new phenotypes, then you are also getting reduced genetic diversity.

A bottle neck, such as caused by isolation of a small population, can result in reduced variation. But the variation is rebuilt during succeeding generations.

A bottleneck is simply the most extreme example of the necessity of reducing genetic variability in order to get a new phenotype. In these cases you usually get complete population-wide homozygosity for most of the new traits. But any new phenotype requires some degree of reduced genetic variability.

And you are wrong that the variation is rebuilt. It has not been rebuilt in the cheetah and shows no signs of ever being rebuilt. And if it were, if that species were to acquire new alleles it would also lose its cheetah character. If you are going to get established species, or speciation, genetic variability must be reduced.

The type of argument you are making could perhaps be used to suggest that the theory overemphasizes selection and underemphasizes the production of new variation. But you won't be able to refute evolution this way, because the empirical evidence shows that variation does build up again if it has been reduced - unless, of course, that particular line goes extinct.

Well, here he's invoked "evidence," but there is no evidence whatever that variation (genetic variability?) builds up again. And there can't be if there is such a thing as a species.