[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Subject Index][Author Index]
Re: Re: dino dna
>Sorry Tom, I have to strenuously disagree. If you have read my other
>morphology is at best representative of only 2% of any organisms' genome. My
This is contrary to what I learned in Bob DeSalle's class, but I'll take
your word on it for now.
>data indicates a mammalian-dinosaur common ancestor at the age I indicate.
>However, this is not an absolute. This is based on a small portion of a
>clock gene, which has been proven to be a reliable clock.
None of the alledged 'molecular clocks' I've seen in the press have been
scaled independantly of morphology/stratigraphy (at least, none of the
clocks used in large scale phylogenetic comparisons). For example, some of
these clocks are dated to the sauropsid-synapsid split, an event
demonstrably shown to have occurred by the Pennsylvannian. How a clock
scaled to these trees could be used to contradict these trees, I just don't
> Until more DNA
>is available, there is always a grey area. Albeit, this data is important
>because it agrees with divergence times shown by fossil record. Morphological
>similarities do not have to be present, and their absence in no way implies
>that the gene tree that I have, based on the sequence data, is wrong.
>Morphology does not accurately reflect eovlutionary history of any organism.
Spoken like a true molecular biologist. I'll reply, "genes do not
accurately reflect the evolutionary history of any organism", a statement
with just about as much weight a priori.
However, for the last thirty years or so, morphologists have had an
important tool for rejecting trees: parsimony. If two cladograms disagree,
one or both can be falsified by combining the information from both trees
into a single analysis, which can, at best, support only one. When I
approach molecular phylogenists with this observation, and ask them "how do
you resolve differing trees from differing genes?" (a very common result),
they have not given me an answer. Now, maybe you have one? I'd like to
know how alternative trees with the same taxa but alterantive sequences can
be resolved, other than the "my sequence is better than their sequence" rote.
>Morphology and physiologogy, for all intensive purposes, are unlinked.
Anatomists worldwide would disagree, but perhaps at a cellular level, you
have a point. However, the effect hypothesis is at work here, and metazoans
demonstrate attibutes at the organismal level which cannot be determined in
>Molecular data reflects far more physiological adaptation than does
>By the way, fossil morphology does not, and cannot define a species, so that
>should be a clue to you when you begin to compare my gene trees to your
>morphological trees. I do not wish to be offensive, but you need to take a
>wider viewpoint. You are not to blame for your attitude, you are just caught
>in a transition because hitherto, genetics has not played any meaningful role
>in the study of extinct lifeforms. It is beginning to, and this technology
I think you are making an a priori assumption here (actually, more than one,
but I'll let that pass...). Although most of my training is in organismal
level biology, I have taken genetics courses in my graduate career
(including one from Bob DeSalle, you for most of the people on this net is
most famous for the project of sequencing fossil insect DNA from amber). I
postualte instead that it is you, and many other molecular biologists I have
met, who have the narrow, reductionist viewpoint: i.e., that organisms are
nothing more than the way DNA makes more DNA. Such an approach misleads
biologists, since it ignores the macroscopic evidence. For example, what
hypotheses about the ancestral states of birds might DNA analyses yield?
Perhaps a bunch of AGTCs, maybe even a few genes, but not much more than
that. Morphological studies, however, have provided testable hypotheses
about the ancestral birds, ones which will only be resolved by additional
Thomas R. Holtz, Jr.
Dept. of Geology
University of Maryland
College Park, MD 20742