Re: in vitro evolution?

R. Joel Duff (
Sat, 13 Dec 1997 09:58:40 -0600 (CST)

Paul Arveson wrote:
> Thanks for your good, technical discussion on _Isoetes_. They are
>an ancient genus, and I didn't realize they were so genetically complex.
> I wanted to ask you (and others) about some of the recent articles in
>Science regarding 'experimental demonstrations of evolution.' I don't have
>the references, I have seen 2 or 3 I have seen in the past few months,
>written by 'evolutionary biologists'. Their experiments all claim
>demonstrations of evolution in vitro, with no micro/macro qualifications.

I am not sure I can give you much of a response on this one. I have not
kept up on the experimental evolutionary research literature, especially in
microbiology. My area is molecular phylogentic reconstruction mainly of
plants which is not experimental at all. I think what we are dealing with
though isn't necessarily experimental evidence demonstrating evolution
(passing or splitting of one genus to another) but observational evidence
of the presumed mechanisms of evolution so we can call that microevolution.
Also there is some confusion between evolution of systems and organisms.
The two can and can't be separated as I see it. Evolution can't come about
without genetic change but genetic change does not necessitate organismal
>In one of the experiments, E. coli were deprived of certain sugars, and
>over the course of several generations, strains emerged that were able to
>metabolize another sugar. This they described as a form of 'natural
>selection' and hence evolution. Couldn't we also interpret this as
>amplification of a metabolic system based on pre-existing DNA?

It really sounds like a sitation where a similar biochemical pathway is
already in existence and a single or several muations are acting to switch
the train to a new track through the modification of some enzyme along the
>I wonder whether perhaps these 'evolutionary biologists' are begging the
>question, assuming what they are trying to establish experimentally. I say
>this as a purely scientific question; I have no intense need to falsify
>evolution but I certainly have been sensitized to the biases (pro or con)
>that exist on the question. Is it fair to criticize the conclusions of
>these experimenters? Is there an experimental test that would demonstrate
>'real' evolution and that can be done on a short time scale?

"real" evolution is so hard to define but I don't think so. I do think
that one can take organisms split them into two lines, place each in
drastically different environmental conditions and in relatively few
generations produce a significant amount of change such that one can
produce genetic isolation but this is artificial selection and it could
well be that strong selectional pressure is not as significant force in
evolutionary history as has been supposed, it is just the easiest to
demonstrate and model. I suspect that other factors more difficult to
model combine to form a more potent evolutionary force than has been
previously appreciated. These is so much talk about allopolyploidy and so
forth (even I mentioned it several times) and it may be the easiest example
of evolution but it is not an important factor in animal evolution AFAIK
(as far as I know). I think we can show clear cases of gene evolution
(observationaly and experimentaly) and we assume that given X amount of
gene evolution this should add up to organismal evolution. I can't
emphasize enough though, that one can have a lot of evolution in a single
gene without resulting in organismal evolution while a single mutation in
another gene might result in immediate genetic isolation despite the fact
there has been little genetic divergence overall. This has lead to many
difficulties in assigning organisms to various taxonomic levels based on
genetic divergence because there is a great differnce in divergence in
different organisms in the same gene.

Look at my aphid/bacteria example. The aphids are described to 11 genera
and I have no doubt there are significant differences morphologically as
well as genetically between them, but the bacterial endosymbiont of all 11
genera is desribed to one genus even though they have coevolved (diverged
at similar points) with the aphids. One would think there should be 11
bacterial genera. In fact there is MORE genetic divergence between the
bacteria (for the SSU gene) than there is for the 11 aphid genera (the gene
tree of the aphids corresponds to generic relationhips while while the
identical gene tree corresponds to species relationships in Buchnera - the
symbiont). But classification and systematics of bacteria is a whole
different field. Genera of bacteria cover a much much wider field of
genetic and morphological variability than in plants and animals and is
often based on different criteria especially since the biological species
concept doesn't really apply.

Those are some of my ramblings on the topic.


Joel and Dawn Duff / | ' \ Spell Check?
Carbondale IL 62901 ( ) 0
e-mail: \_/-, ,----'
or ==== //
or / \-'~; /~~~(O)
* * * * * * / __/~| / | * * *
\\\/// \\\/// =( _____| (_________| \\\///