[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Subject Index][Author Index]

Re: holophyletic groups only & the future


Jaime,
Well, I certainly would never put Kinman markers for every dinosaur. Such markers are only used when I create a paraphyletic group, which I do as little as possible. As I said in my book, "Paraphyletic groups (or semi-paraphyletic groups used herein) are often needed to make classifications simpler, more useful, and anagenetically more informative. However, their overuse is to be avoided and our greatest challenge is to achieve an optimal mix for each taxon (purely cladistic for many groups, particular at lower taxonomic level)." Extinction and fossil gaps make cladistic classification very useful in a majority of cases, but some paraphyletic cuts are unavoidable in a comprehensive classification of life. I challenge any strict cladist to present a purely cladistic classifcation of organisms, even just down to phylum level. There aren't all that many phyla, I would really like to see a cladist try to do this. I don't think it is possible.
As for the bird examples you cited, I am not up-to-date on the controversies over bird classification.
However, in 1994, I placed Strigiformes (owls) as sister group to Caprimulgiformes (nightjars and allies). I never even considered placing owls with falconiformes. I considered owls and nightjars as closer relatives to one another, than either was to hummingbirds, although perhaps more recent evidence indicates otherwise. All vultures remain in Ciconiiformes, until I see good evidence that some vultures are closer to falconiforms.
And when it comes to using DNA sequences, I have seen enough examples of overextrapolation and misinterpretation, that it has to be pretty strong and convincing molecular data before I will let it override morphological evidence that I find more convincing. The analysis of molecular data is still new enough that all the potential problems involved have not been adequately tested. It certainly has had its success stories, but I suspect that overexuberance may be causing as many problems as it is solving. But as I said above, I am not up-to-date on the avian literature.
For the most part my classifications follow majority opinion, except for those cases where I am convinced that the minority is correct. I certainly ignored the majority opinion when I classified scorpions and eurypterids in a new Class Scorpionea (while the only cladistic analysis nested scorpions well within the arachnid class). It was only recently that new fossil evidence confirmed the holophyly of a Class Scorpionea. In some cases I followed the majority and wish I hadn't, especially my temptation to split off the delta and epsilon branches of Class Proteobacteracea (Phylum Eubacteria) from the larger main group of proteobacteria (alpha, beta and gamma branches). Proteobacteria sensu lato is still the majority view, but my doubts are now even stronger (it is probably paraphyletic at best, and the likelihood of polyphyly has seemed to increased in my mind). I evaluate each case separately, and hopefully in the long run I will have more hits than misses. Only time will tell. Been a long day, and too tired to write more, which I am sure will be a relief to some. :-)
--------Ken Kinman
********************************************************* 
From: "Jaime A. Headden" <qilongia@yahoo.com>
To: dinosaur@usc.edu
CC: kinman@hotmail.com
Subject: Re: holophyletic groups only & the future
Date: Mon, 3 Jul 2000 18:14:43 -0700 (PDT)

Ken Kinman wrote:

<... I hit upon the idea of semi-paraphyletic groups,
where you removed a group (like birds from reptiles),
but you had to leave a marker behind to explicitly
show this. It's like removing it formally, but not
removing it in an informational sense. Birds (Aves)
are still shown to be dinosaur descendants, but
without all this silliness about non-avian dinosaurs,
etc.>

  Then why not put a marker for every dinosaur beside
its sister group? This might seem like a daunting and
ridiculous task, but if done, it would be monumental
and may also show how many clades would need to be
clarified paraphyletically or semi-paraphyletically to
show trends, development, etc., while showing the
uniqueness of whatever clade there is being analyzed.

  Let me ask another question: how would you explain
the process of evolution by relating that birds are
elsewhere in the system than with their ancestors or
sister groups? Would you, say, put *Psittaca* next to
Passerae in swung-brackets because while Passeriforms
are considered a mono[holo]phyletic group,
psittaciforms (sensu lato) are said to have descended
from them. In a cladisitic system, we can say that
*Psittaca* and all taxa closer to it than any other
bird group [= Psittaciformes], and place this within
Passerae (all taxa closer to *Passer* [I would use
*Corvus*, actually] (like oscines and sub-oscines)
than to all other bird groups) one could then nest
*Psittaca* (and a bunch of other bird "orders") into
Passeriformes. This would go real well with a lot of
ornithologists, who've been dealing with more
traditional heirarchies for decades and centuries,
even, yet Sibley and Alquist, Olsen, et al. have
suggested some forms or another of this idea, have
suggested spliting vultures to nest with storks and
falconiforms separately, took owls from the "predatory
birds" grouping and put them with hummingbirds and
nightjars, etc. Dropping of traditional "orders" for
most of these groups is becoming a matter of course,
and this is based on genetics -- some would say
arguable, but a recent paper in _Auk_ suggested, on
DNA and less tangible evidence that a subspecies of a
nightjar be elevated to species instead because it
represented an anagenetic population that could not
have descended from within the original species (I
think it is *Aegotheles whitei*).

  Suggesting therefore that removing a descendant from
a group is in anyway natural because it better
conforms to one other side's views of separation or
heirarchy is not a sound practice, because this really
is a "What feels good to me" type of practice.

  Similarly, for one in terminology of descent and
overall phylogeny, the term "non-avian" used for
groups of dinosaurs is applied only in such a frame as
overall or avian sistematics. One could say
non-orithuran for *Archaeopteryx* or non-carinate for
*Hesperornis* in terms of relationship, but no one is
proposing using this terminology whenever. For
instance, unless there is a reference to the
appearance of a feature in a larger group and its
aspect to phylogeny, a biologist need only refer to a
feature in aspects of its biology, or gross anatomy,
or function, and never have to deal with a phrase like
"non-ornithuran avian brachial structure" for some
unique process only enantiornithines and all other
ornithurans might share.

  I would enjoy running my finger down lists of
characters and running to every bird in the
Smithsonian to analyze their appearance, truly, and
testing DNA, and comparing results to determine the
difference between the anagenetic and cladogenetic
phylogenies produced by such results. Purely
character-driven analyses (e.g., Gauthier (1986)
Sereno (1999)) cannot be tested with DNA to produce
comparative studies, and even those DNA studies may
not be accurate, based on mutation of formerly shared
characteristics, and this is a problem when dealing
with end-results that every neontologists and
paleontologist has to: there are no intermediate forms
-- all specimens of life are end products in one way
or another, and each had features in common with
others, but mutated them beyond recognition, so that
while we can hypothesize on the sister groups and
ancestor/descendant relationships of organisms, none
whatsoever can be truly accurate. Best guesses and all
that.

  Not that I'm saying we should drop everything
because "We'll never know for sure" because unless we
do test, we will never figure out the probabilities
and plausibilities inherent to Nature and biology in
all its complexity. Science, simply, is the practice,
or art, of knowing {Lat., _scio_ = to know}. Any
approach to science must be done without assumption of
any basic process or test, to better define the results.

=====
Jaime "James" A. Headden

  Dinosaurs are horrible, terrible creatures! Even the
  fluffy ones, the snuggle-up-at-night-with ones. You think
  they're fun and sweet, but watch out for that stray tail
  spike! Down, gaston, down, boy! No, not on top of Momma!

__________________________________________________
Do You Yahoo!?
Kick off your party with Yahoo! Invites.
http://invites.yahoo.c

________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com