[Taxacom] Robust argument

Richard Pyle deepreef at bishopmuseum.org
Sun Jun 28 20:07:33 CDT 2009


Thanks, Bob -- and no worries! (same to you, Richard Z.)

I guess we've underscored Robin Leech's point about the need for precision
in language.  As I am not, myself, a phylogenist (phylogeneticist?) -- I'm
much more interested in doecumenting the existence and richness of
biodiversity than attempting to infer patterns of historical "relationships"
(whatever that means) -- and therefore I am quite confident that I am the
most culpable in abusing the phylogen- series of words.  I'm very happy
going with "evolutionary history", or some other such term(s).

I believe the archives of this list contain my views about the potentially
misleading nature of what I call "stick-figure" graphs (i.e., cladograms)
for representing a much less clean, much more complex pattern of
reproductive events and the flow of information across many, many
generations of many, many, individual organisms. But neither am I a
clado-basher.  My (admittedly under-informed) sense is that the people who
engage in that form of inquiry are (more or less) on the right trajectory,
and over time are improving methods of analysis and interpretation.

Just a couple of further clarifications:

> Trees like this are both useful and defensible for 
> classifications. I think I agree with you that for this 
> limited end, there's a lot of overlap between sequence 
> information and morphological information, and if you've got 
> the whole genome to play with, then build trees with it and 
> don't worry about morphology. If you include morphological 
> data as well, you risk blurring the inference with 
> redundancy.

That's actually not where I was intending to go with my comments, but I
agree with you nonetheless.  


> The tree I'm talking about is a drastic simplification of 
> 'evolutionary history' sensu me, which includes lots of 
> events and patterns which - as we both agree - are not 
> inferrable from sequence data. To try to infer this vastly 
> more complicated  history, you need lots of morphology, 
> because morphology is so much more sensitive to selection 
> than sequences, and lots of biogeography, because history 
> contains addresses, and whatever fossils you can find.

The thing is, I *was* referring to 'evolutionary history' sensu you -- but I
just can't seem to get the points across clearly.

Let me lay it out this way:

1) I believe that, in the context of modern taxonomy, both morphological
evidence and molecular evidence *can* be informative for elucidating
evolutionary history (sensu you).

2) I believe that the aspects of morphology that have real value in
elucidating evolutionary history (sensu you) have a genetic foundation.
That is, the informative aspects of morphology are based on blueprints
transferred across many successive reporductive events.  (This is what I
meant by the informative aspects of morphology being *somewhere* in the
genome.)

3) At this point in history, I believe we do not come even close to fully
understanding how to extract and interpret the information that exists
within genomic data; which is why I still regard morphological data as more
useful for elucidating evolutionary history (sensu you) in many cases.  This
is quite a different thing from saying that the information exists in the
morphology, but does not exist in the genome.

4) I believe that when we have achieved the point of maximal genetic
information extraction (both technologically, and intellectually), we will
still find ourselves FAR short of the whole evolutionary story -- mostly
because so much information has long been obfuscated and lost (through the
ravages of time, extinction, entropy, mutation, selection, Heisenburg
Uncertainty Principle, etc., etc., etc.) from the collective global set of
extant genomes AS WELL AS all other lines of evidence (see item #6 below).

5) I believe that if/when we do (evenutally -- maybe in my lifetime, maybe
not) have a perfect understanding of how to extract information about
evolutionary history (sensu you) from the collective global genome, we will
probably no longer find morphological analysis useful for elucidating
evolutionary history (sensu you); except when comparing fossils to each
other, and to extant organisms.

6) Regardless of whether any or all of the above beliefs are true or not, I
adamantly believe that the genome is *not* (and almost certainly never will
be) the only source of information for elucidating evolutionary history
(sensu you). I've already listed examples of other non-genetic sources of
information in earlier posts (as have others).

> This is what Michael Ivie so nicely wrote about here, when 
> puzzling over Montserrat. It's what awes me about the 
> chemical ecology of terrestrial arthropods and their plant 
> associations. The phylogeneticists have their sequence 
> inference hammer and they're banging away at every nail in 
> sight. Evolutionary historians sensu me are still thinking 
> about what tools we need and how we might use them. The 
> tree-builders are pasting their simple black-and-white 
> abstractions on every wall in the gallery, and it annoys me 
> sometimes - OK, it annoys me a lot of the time - that they've 
> hidden the richly detailed and beautiful oil paintings 
> underneath, and are getting astronomical prices for their 'art'.

I'm with you on all of that!!

> Privilege of getting old: you're expected to curmudge now and then.

It's also the privilege of the middle-aged and under-slept....

:-)

Aloha,
Rich






More information about the Taxacom mailing list