Re: limit of selection???

From: Malcolm (malcolm_at_55bank.freeserve.co.uk)
Date: 08/25/04 

Date: Wed, 25 Aug 2004 04:41:39 +0000 (UTC)

"A.C.H." <br.hessels@planet.nl> wrote
>
> The issue i was trying to pose was what this would mean, how far
> selection can push an adaptation, how perfect it'll become.
>
> [moderator's note: I can't be the only one thinking of Fisher's
> Fundamental Theorem here, can I? The rate of evolution is proportional
> to the product of the selection gradient, the genetic component of
> phenotypic variation, and the heritability of the trait. Does no one
> else think this is relevant here? - JAH]
>
>
Yes, I think that's right.
>
> Compare these two situations:
>
> 1) Childbirth is a problem, females die during birth. Evolution by
> natural selection is happening, which is the cause of adaptation
> (easier childbirth).
>
> 2) Women are perfectly adapted to giving birth, childbirth is no
> problem at all, no women die giving birth.
> In this case, the natural selection, which caused the adaptation in
> the first place, falls away.
>
> Therefore the state of perfect adaptation (all women give birth easy)
> cannot be reached, because, paradoxically, the cause of the
> adaptation: natural selection, falls away when you approach this
> state.
>
> So i predict: child birth will be less hazardous, but will never be
> easy.
>
What you've got to remember is that mathematical models of evolution are an
abstraction. A complex behaviour like childbirth is affected by a huge
number of genes. However there is a single gene with a very pronounced
influence on head size, called microcephalin. The isoleucine / glutamine
(IQ) repeats seem to control head size, the more repeats, the bigger the
head. This is a discrete number, and can reach an optimal figure.Normal
humans have 74 repeats, and lower repeats are rare and cause
microcephalisation, a disorder seen in some Pakistani patients.
Of course the microcephalin gene does not on its own "control" head size, it
is purely part of a complex regulatory system. But if this was the only gene
on which evolution works it would presumably quite quickly settle to a level
of 74 repeats, with just a tiny number of deleterious mutations.

But the relevance of Fisher's theorem is that, when everyone is nearly
gentically identical, heritability of a trait drops to a very low figure
(all the variation is caused by the environment), and when something is
nearly perfect, the selection coefficient can be low, if there is not direct
competition. (Women do not compete with each other for the most easy
childbirth, on the other hand they do compete in getting limited resources,
so even a slight advantage may mean that the best woman gets to marry the
King).

Relevant Pages

  • Re: limit of selection???
    ... > selection can push an adaptation, ... The only thing that *physically* happens is that population of animals ... > 1) Childbirth is a problem, ...
    (sci.bio.evolution)
  • Re: limit of selection???
    ... >> escape its selection pressure. ... >> The adaptation is linked to whatever selects it, ... Childbirth is a problem, ...
    (sci.bio.evolution)
  • Re: Science v Religion
    ... You might say that there is adaptation here - the increasing purpleness ... natural selection ). ... I recommend Ernst Mayr's What Evolution ... Similarly wrt purple pansies, regardless of how they are produced (btw ...
    (uk.religion.christian)
  • Re: A critique of the BBC aquatic ape programme and the transcript.
    ... >>of species are the result of natural selection. ... If it's humans, is it because we've had selection to move ... stating that it was an adaptation therefore it's an adaptation. ...
    (sci.anthropology.paleo)
  • Re: Repeat random numbers from a range and text
    ... Now, select the 25 numbers and right click in the selection, and choose ... You did say to allow repeats! ... having to key the each number of the range to a separate row line) ... formulas are keyed to the same worksheet where the random options are ...
    (microsoft.public.excel)