Mike <syvanen@ucdavis.edu> wrote:

The origin of a novelty is mutation,

recombination or crossover. It can
go to fixation, as Larry never tires of reminding us, for a number of
reasons other than selection. Why should the ability to give *rise* to
such novelties require group selection? I mean, there is no selection
for the ability to mutate; mutations just happen. Some regions of a
genome may be subjected to less selection for error correction,
structural stability via, I suppose, methylation and the like. The
conclusion that therefore they have been selected for the ability to
mutate strikes me as horse-after-cart thinking. It is, oddly, overly
selectionist, given that the *default* state of genes is to mutate
unless corrected. So we ought to say there has been selection for those
genes that need to be highly conserved, to be conserved, not that those
genes that are highly unconserved are so because it was selected for.

I am justifying this scenario by the potential advantage that could
accrue to a population in the future. Now we must address the question
as to how the population happened to already have those mutator
activities. We answer this question by looking to the past, ie
existing populations arose because they were descendants from other
successful struggles for existence. Not logically connected to this
general arguement for group selection but an empircal fact, is that the
genes controlling these mutator activities are invariably heavily
involved in horizontal gene transfer events. From this we can further
argue that HGT, and not vertical evolution per se, ensures the
survival of these mutator systems.

I know Usenet is a quick and dirty medium, but that is too quick for me.
Mutating regions

are still favourably mutating because either (i)
unsuccessful mutants have been eliminated, or (ii) the rate of failure
of mutants is not so high that selection for error correction or genomic
stability has occurred significantly. We don't need to say that it is
*favourable* to mutate in those regions, necessarily.

Of course, some mutation clearly *is* favourable, such as the V region
of immune cells. And that clearly *is* a case of group selection - those
groups of somatic cells that did it well survived better than those that
didn't, and so the related gametes were more reproduced.

The one paper that most influenced my thinking about the importance of
group selection was one by Stephen Stanley that came out in PNAS
(1979?). A short and sweet group selection arguement by a
paleontologist. Unfortunately after a flood of critism he retreated
and modified his ideas but I think he got it right the first time.
Some people just cannot hold up against social pressure.

And some peopel recognise valid criticism :-) but I haven't seen that,
so I'll chase it up if I get the time. Thanks, as always.