"Tim Tyler"  wrote in message
>
> Gould's argument was that *complexity* followed a random walk.
>
> This criticism doesn't apply to that claim.
>
> It /would/ apply to the claim that evolution - or genomes - followed
> a random walk - but Gould never made those assertions in the first >
place.
>
The problem is, what do we mean by complexity? We could take size. This is
highly significant for any animal's adaptive strategy, and there are many
cases of animals becoming smaller as a result of evolutionary pressure as
well as of animals becoming bigger. If we ignore the local pressures on size
then we do have something that looks superficially like a "random walk"
starting from a single-celled base.
What about genome size? For multi-celled eukaryotes the energetic cost of
carrying extra DNA is trivial. However we don't understand the function, if
any, of the non-coding DNA. So genome sizes may be taking a true random
walk.
However what when we restrict ourselves to DNA that does something
(expressed genome size)? Now any random change beyond a small level will
almost certainly be fatal to the organism involved. Only an evolutionary
ratchet with beneficial changes providing the basis for more beneficial
changes can drive big alterations in expressed genome size. Now a beneficial
change could either simplify or complicate the genome. However if it
simplifies then it does so by undoing a change which was previously
beneficial. Sometimes this will happen, but this is not a random walk, as
with the body size example, because simplification does not have a direct
adaptive significance, but only because of the effect on the phenotype.
Evolution seldom acts on a "few proteins" organism directly, so we don't
have the body size situation with niches appearing for organisms of
different sizes.  Only if we propose that simplifications and
complifications are exactly and necessarily balanced in their adaptive
advantage would you have random walk characteristics.
---
þ RIMEGate(tm)/RGXPost V1.14 at BBSWORLD * Info{at}bbsworld.com

---
 * RIMEGate(tm)V10.2áÿ* RelayNet(tm) NNTP Gateway * MoonDog BBS
 * RgateImp.MoonDog.BBS at 4/18/04 6:22:09 AM