Definition of evolution

arlin at ac.dal.ca arlin at ac.dal.ca
Sat Nov 2 16:49:03 EST 1991


Larry has accused me of putting words in his mouth, and I have to
admit my guilt, in a sense, although I hope it will be understood 
that the intent was not malicious. 

In article <1991Nov1.171612.4442 at gpu.utcs.utoronto.ca>, Larry writes:

>I don't believe that I have EVER said that evolution can be defined solely
>in terms of allele REPLACEMENTS and I don't know of any evolutionary
>biologist who makes such a claim. I always understood the common definition
>to include gene duplications, recombinations, deletions, etc. Arlin seems
>to recognize that many of us take this broad view of "a change in the
>frequency of alleles" but he persists in putting the words "allele
>replacements" in my mouth. 

An allele has to have a home, which we call a locus.  An allele is not
an abstract entity representing in some generalized way a hereditary
determinant. Rather, an allele is a variant sequence at a chromosomal
locus, and the presence of an allele always implies the presence of at
least one other allele that can reside at the same locus.  Whenever
one speaks of an increase or decrease in the frequency of one allele,
as Larry has done numerous times (as in "a change in the frequency of
alleles" above), the decrease or increase in the remaining alleles at
that locus is implied. Thus, to rise in frequency, one allele must
*replace* another.  All or nearly all of the allelic changes in 
evolution are total replacements of one or more alleles by another allele,
though some people define evolution broadly enough to include changes
in frequency that do not constitute complete replacement (others would
exclude such changes on the grounds of being too easily reversible to
be considered "evolution").

Since alleles are *defined* as residing at a locus, it was only
reasonable for me to assume (:-> since Larry has shouted the words
DEFINITION and PRECISE so many times) that when Larry spoke of
allelic changes, or changes in allele frequencies, etc, that he
understood he was talking about a kind of change that involves two or
more variant sequences at a locus, one of which must REPLACE the 
other(s) completely or partially.  However, it is clear now that
Larry did not mean this, and I apologize for any discomfort, 
confusion, or embarassment that my assumption has caused.

I hope we can lay history aside now. It has now been established by
the preceding discussions that some genetic changes in evolution do
not involve allelic changes.  In the case of introductory textbooks
that define evolution as shifting allele frequencies or some such
simple formulation using "allele", it is a matter for debate by
literary psychoanalysts what the authors intend, or whether they can
be convinced to change their minds when the meaning of their statement
is explained to them.  Whether or not they actually mean what they
say, the formulation is wrong.  When I see incorrect statements in
textbooks, I cross them out with a big red pen.

If accepted, this is a BREAKTHROUGH in our current discussion.  
This is NOT just a
matter of semantics, since I've fielded questions from plenty of
correspondents that simply do not understand that some genetic changes
cannot be modelled as allelic changes, even after I've explained
to them what allele replacements really mean.  Usually, it takes a 
while for this point to soak in.  Stewart Schultz has recently 
argued that molecular drive and the origin of new loci could
be modelled as allele replacements (if he still holds this position,
he should speak up soon). 

PROGRESS in this discussion would now be represented by 1) recognizing
that "allele" and "allele replacement" have precise definitions; and
2) that evolution cannot be defined solely in terms of alleles and
allele replacements.  From what I have seen, the next two topics that
demand immediate attention are:

1. Isolation.  Must a definition of evolution encompass speciation?
Must speciation involve isolation?  Is isolation of one population
into two something distinct from genetic change in one population?

2.  The nature of definitions.  We all know that the
*fact of evolution* is that species change (rather than remaining the
same) and that they arise from common ancestors (rather than being
created each independently).  In Darwin's time, change in a species
over time and common ancestry of different species were inferences
made mainly from the fossil record, comparative morphology, and
biogeography. In modern times, we still refer to these evidences, and
we can also make both inferences from comparative molecular biology.
Molecular biology was inaccessible to Darwin, but this didn't stop him
from seeing evolution, in spite of having erroneous theories about
genetics.

What are we trying to do when we make a *genetic* description or
"definition" of evolution to replace the idea of evolution that Darwin
had?  Are we simply stating the same thing in a different way (how)?
Are we saying that genetic change is evolution, while morphological
change is not?  Are we saying that one is more fundamental than the
other, or trying to reduce one set of phenomena to a different set of
phenomena?


I look forward to further discussion of these and other issues, though
I cannot devote as much time in the coming week as I have in the past
week.

Arlin Stoltzfus

Arlin at ac.dal.ca



More information about the Mol-evol mailing list