From: Dawsonzhu@aol.com
Date: Tue Oct 14 2003 - 20:03:45 EDT
harper.10@osu.edu wrote:
> Now we have the second prediction. How will the contours of P(x,y) match up
>
> with plotted points? Adaptationism predicts that the local concentrations
> of observed shapes will be near the regions of local optimums in the
> performance. Structuralism predicts that such a situation would be purely
> coincidental and is not to be expected. ID?
>
I would think P(x,y) (the performance) would have to be a time dependent
and environmentally dependent function so it would need more parameters
to describe the affects on x and y. For example, maybe this square structure
becomes more visible when y > ymax (where say "y" is the height of the box
shaped shell).
So in a rocky environment, the box shells would favor having
a small height to avoid become dinner (after all, the nail that sticks out
gets
pounded down). Likewise, perhaps to get dinner, it would have to spread out
over a larger area (x) to filter more nutrients. But in a sandy environment,
there would be the possibility that the box shell could burrow down into the
sand (favoring a large y) and at the same time a large x becomes a liability.
That would represent environmental factors.
But now say some new predator comes in that has adapted a skill for seeing
squares. All these box shells are in trouble now. We have somewhat exhausted
our parameters in x, and y, but say r represents curvature at the corners.
Let's also
presume that the rate at which this organism enters is not too rapid or they
only gradually develop this skill for recognizing box shapes. The reason I
mention
rates is because things like human predation in Australia appears to have
caused
a rapid extinction of a number of animals there. Anyway, the box shells with
curved
corners would then gain the selective advantage.
So my main point is that the dynamics of evolution are where the main
difficultly lies. P(x,y) does not depend so much on x and y, rather it
depends
on both a time dependent and environmentally dependent factors and the
favorable values of x and y represent something of the response to these
factors. Then there are the collisions of asteroids in this hypothetical
world
that destabilize an otherwise perfectly equilibrated system. Here again is
massive
change that happens very rapidly. So you have a variety of uniform and
nonuniform
conditions that occur. It seems like the underlying genetic rate of change
(population
dynamics) would need to be grafted into this to make it express the response
to these
external factors that strongly influence the performance.
On the other hand, trying to minimize the number
of parameters as much as possible does help me see how one might be able
to test this at least semi-analytically. But the complexity of the problem
is and
remains challenging to show clearly.
by Grace alone we proceed,
Wayne
This archive was generated by hypermail 2.1.4 : Tue Oct 14 2003 - 20:04:59 EDT