Terry Gray wrote:
> It would be good to sit down and chat with you sometime.
> Make sure you call me up the next time you pass through
> Colorado.
I'll be lecturing at the University of Colorado-Boulder, the Air Force
Academy, and Colorado State on September 20-21. Watch the
Discovery Institute and ARN web pages for details. I hope to
see you then!
> I've asked this question before but you've never answered it to my
> satisfaction.
>
> Let admit the possibility of polyphyly and focus our attention on
> branches/bushes a little higher up. Now does the genetic
> evidence and the known mechanisms of modern, metazoan
> lateral transfer suggest that all insects are descended from a
> common ancestor (or ancestral population)? How about
> all arthropods?
Lateral gene transfer (LGT) would be negligible in the Metazoa,
relative to prokaryotes. Please note that LGT is a *hypothesis*
invoked to explain the anomalous phylogenetic distribution
of genes and their protein products amongst the prokaryotes.
The anomalous distributions include such basic housekeeping
genes as aminoacyl tRNA synthetases. These aren't the
resistance plasmids (for instance) that we know may pass
horizontally between current bacterial populations. We're talking
about swapping the basic wiring and plumbing here. Whether
LGT on that scale is plausible is a significant unaswered question.
Early promiscuous LGT is a hypothesis in need of testing; it is
not an observation.
About the arthropods. Here's a funny story: take it as a
cautionary tale.
The Ecdysozoa [the molting animals] are a fairly new clade in
metazoan systematics. I mean, who would think of classifying
arthropods (e.g., flies) and nematodes (worms) together?
But the clade is now a widely-accepted group in animal
phylogeny. Anyway, a couple of months ago, I was reading
Fryer's (1996) long critical assessment of arthropod monophyly,
and ran across the following:
Philippe et al. also draw attention to 'artifacts' that lead to
such 'topological errors' as the portrayal of Drosophila and
Aedes [both insects] as a sister group of nematodes, which
they attribute to the rapid evolution of both groups...(1996:46)
I have the book containing Philippe et al.'s chapter, so I looked
up the citation. Sure enough, Philippe et al. write, of their own
18S rRNA results:
The tree displays a number of interesting features and
also immediately illustrates one major source of artefact:
several species or groups of species display much longer
branches, indicating either a two- or three-fold higher rate
of evolution in those sequences or inaccurately determined
sequences. Such inequalities are known to generate
topological errors in the positioning of the corresponding
taxa....A clear example is provided by two insects, Drosophila
and Aedes, which emerge as a sister group to nematodes
(the latter also all having long branches)....Both for Drosophila
and for Caenorhabditis it is known that the problem lies not
in the quality of the sequences but in their rapid rate of
evolution, as has been pointed out in several previous papers.
(Philippe et al., 1994:18)
But today, in the year 2000, the grouping of flies with worms isn't
an artefact (i.e., error) at all. Drosophila and C. elegans are now
thought to be cousins within the Ecdysozoa. Both animals molt,
although by classical morphological criteria they have precious
little else in common.
Imagine going to a big family reunion. It's picture time, and
the photographer would like to keep the families together. So
where should you stand? With this guy? Well, your 18S rRNA
resembles his, but that's an artefact. He's not a close relative.
Stand over there, with those people. Wait a minute -- that
first guy was your close relative after all! We made a mistake
about our mistake. Go back and stand with him. OK,
everybody say "cheese" --
There's a serious point here. A few years ago, I attended a
guest lecture (in a graduate course on systematics taught by
Paul Sereno) given by Scott Lanyon, who at the time was a
molecular systematist at the Field Museum here in Chicago.
Lanyon began by saying, "When I see a published molecular
phylogeny, the first question I ask myself is, 'How much of this
should I believe?'" I've taken that advice as something of a
motto ever since. Molecular methods are NOT a cure-all for the
puzzles that have bedevilled systematics since Darwin's time.
So when I hear that molecules have affirmed the monophyly
of the arthropods (or whatever), I say maybe, maybe not:
how much of the result should I believe? The ongoing instability
of metazoan phylogeny is not confidence-inspiring.
[For a discussion of the arthropods in particular, see G. Fryer
1996, section headed "Molecular Approaches and Their
Limitations"]
> How about all marsupials?
Don't know.
> I can see why the "new" evidence is confusing with respect
> to the very ancient lineages and this is very, very interesting.
> But is it really so confusing with respect to metazoan sequences?
> I don't get that picture at all in reading the literature, if anything
> the genomic arguments for are sewing up rather nicely the
> arguments for common ancestry. The paper in Science about
> a year ago on large scale chromosomal mapping giving
> tremendous insight into the relationship between various
> mammalian orders is a prime example.
Interesting you should mention that _Science_ paper. I have the
poster from that issue (15 October 1999) hanging in my office.
It shows the chromosomal banding patterns for 20 mammalian
species, with zebrafish as the outgroup. Across the top of the
poster are black lines, connecting the various species, representing
a phylogenetic hypothesis. When one turns to the article (O'Brien
et al. 1999) on which the poster is based, one reads the following
(p. 463):
The hierarchical phylogeny among the mammal orders,
dating back to at least 60 to 100 million years ago
[represented on the poster], is not yet resolved. The
foldout included with this issue of _Science_ presents
an amalgam of opinions about placental-eutherian orders
that are supported by some but not all data examined to
date. For example, the superordinal group Glires, which
associates rodents and hares, is strongly supported by
a number of unusual morphological traits and fossils, but
numerous molecular studies of nuclear and mitochondrial
genes do not support the association. In contrast, multiple
molecular studies have aligned the superorder Afrotheria
(six orders derived from Africa, including hyraxes, elephants,
elephant shrews, sea cows, aardvarks, golden moles, and
tenrecs), but strong morphological association of these
orders is not apparent....Finally, the causes of genome
exchanges, their dichotomous rates, and the reasons for
species-level fixation are not well understood, a caveat
that would affect the assumptions of phylogenetic analysis.
None of these aspects are fatal, but considering the
limitations will be critical in applying genome differences
to the evolutionary history of mammalian orders.
Ask yourself Lanyon's question -- how much of the result should
I believe? In particular, without the assumption of the common
descent of the mammals, do the chromosomal data establish
that the mammals share a common ancestor?
No. Common descent is not a hypothesis being tested by the
data. Rather, the data are being interpreted with common
descent as a given.
> I just think that you are mixing apples and oranges when you
> bring in all the lateral gene transfer stuff from bacterial genomics
> into these discussions of common ancestry of animal phyla
> and/or classes. Doolittle and company may find the task of
> reconstructing lineage well-nigh impossible by these new
> findings, but I serious doubt if any of them are ready to suggest
> that evolution (change through time via "ordinary" biological
> mechanisms") hasn't occurred.
Of course you're right. An intriguing question to ask would be
what sort of data might cause Doolittle and company to question
"evolution." I can't guess.
> Does your critic of this methodology undermine DNA fingerprinting
> as a forensic method?
No. But we know DNA fingerprinting works because we have an
independent means of assessing the reliability of the method.
If DNA fingerprints placed non-relatives together -- i.e., persons
known to be unrelated by data independent of DNA -- we wouldn't
trust the technique. Over the time scales and taxonomic distances
treated by phylogenetic methods, by way of contrast, we have
neither an independent means to assess reliability, nor any reason
(other than the question-begging) to assume that our extrapolations
are reliable.
That's it for me, friends -- back to work.
Paul Nelson
Senior Fellow
The Discovery Institute
www.discovery.org/crsc
References
Fryer, G. 1996. Reflections on arthropod evolution. _Biological
Journal of the Linnean Society_ 58:1-55.
O'Brien, Stephen J. et al. 1999. The Promise of Comparative
Genomics in Mammals. _Science_ 286:458-481.
Philippe, Herve et al. 1994. Can the Cambrian explosion be inferred
through molecular phylogeny? In Michael Akam et al., eds., _The
Evolution of Developmental Mechanisms_ (Cambridge: The
Company of Biologists Ltd.), Development 1994 Supplement,
pp. 15-25.
This archive was generated by hypermail 2b29 : Wed Jul 26 2000 - 10:13:58 EDT