"Jim McGinn"  wrote in message
> >
> > What will happen with the beans selected by the scoop? In this
case, > >
smaller beans won't tend to fall out. There will be no tendency for
> > beans to get bigger. However you will get genetic drift effects
>
> I thought you were setting up a scenario to test genetic drift.  Now
> you are just telling us it exists.
>
I set up the examples expecting you to say "both are NS". I would then say
"That's what the argument is about, different definitions of natural
selection".
Of course since I believe in genetic drift I believe it will occur when you
select beans from a bucket using a scoop.
>
> > Bean size takes a random walk,
>
> Now explain how NS cannot involve, "taking a random walk."
>
The phrase was popularised by Stephen Gould. The idea is that a drunk is on
the sidewalk between the wall and the gutter. Each step he takes he might go
randomly left or right, but if he hits the wall he can't go any further
left, and if he hits the gutter he winds up, well, in the gutter.

Now bean size will be partly controlled by a complex of genes. Since the
trait is heritable, taking a scoopful slightly biased towards big beans
means that the beans in next season's bucket will be slightly bigger - one
step of the walk. The next scoop out might bias us to bigger beans again, or
it might bias us to smaller beans, another step in the walk.
Eventually one of the alleles will go extinct - this is the equivalent of us
hitting the gutter. (Gould used the wall as an example of minimal possible
complexity, here we've got two gutters).
Of course, at one locus an allele for big genes might fix, whilst at another
an allele for small beans might fix, both at the same time since there is no
selection. It would be better to think of a group of drunks, all staggering
down the same road.
---
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 8/4/03 11:30:17 AM