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

Representation of Paraphyletic Groups in Phylogenetic Taxonomy



I've cross-posted this to the newsgroup sci.bio.systematics, which is a
more appropriate forum for this discussion. I suggest that followups be
posted to that newsgroup only, and not to the Dinosaur Mailing List.

To sum up for anyone reading from sci.bio.systematics (ha!), in a previous
post I argued that the methods for representing paraphyletic groups in
phylogenetic taxonomy (e.g., non-avian Dinosauria, ectothermic Amniota,
etc.) are superior to the methods of the traditional system

On Sun, 11 Feb 2001, Ken Kinman wrote:

>      Sorry I missed that post----I would have certainly answered it.   Your
> criticisms may apply to the traditional Linnean System, but they do not
> apply to my modification of the Linnean System (i.e. The Kinman System).

Some of them still do, I think. For example, I mentioned that naming a
paraphyletic group in the traditional system prevents certain other
paraphyletic groups *as well as* certain clades from being named. (E.g.,
naming Class Reptilia [=Clade Amniota minus Clades Aves and Mammalia]
prevents us from naming the paraphyletic group of oviparous members of
Clade Synapsida, or from naming Clade Sauropsida, Clade Diapsida, Clade
Sauria, Clade Archosauria, etc.) Correct me if I'm wrong, but this
phenomenon of "selfish" paraphyletic taxa seems to be present in your
system as well.

>      The only difference is that I would tend to use plus signs instead of
> minus signs (when necessary).  The traditional (paraphyletic) Synapsida is
> simply Pelycosauriformes + Therapsiformes.

One disadvantage here, of course, is that you can't tell it's paraphyletic
by looking at it. "Non-mammalian Synapsida", however, is not only
obviously paraphyletic, but obviously singly paraphyletic. You can tell
that without even having any idea what Mammalia and Synapsida are.

>  The clade Synapsida is *very* simply Reptilea 1, which as my book
> shows is the above two orders + the marker {{Mammalea}}.

What's simpler, "Reptilea 1" (which seems to require consulting an index)
or "Synapsida"?

> Clade Cyanodontia would be something like
> Therapsiformes 13+, and therefore your Cyanodontia-Theria would be my
> Therapsiformes 13+ plus Mammalea 1-10.

Okay "Therapsiformes 13+ plus Mammalea 1-10" or "non-therian Cynodontia"?
I don't see much of a contest here.... (And correct me if I'm wrong, but
aren't those numbers prone to change?)

Once again, your formulation is not obviously paraphyletic (in fact, it
could be polyphyletic if you weren't familiar with the names).
"Non-therian Cynodontia" is obviously singly paraphyletic.

>      The really major difference is that phylogenetic taxonomy formally
> names an enormous number of taxa.  I have gone the other direction, reducing
> formal taxa to only Phylum, Class, Order, Family, Genus, Species
> (eliminating all "formal" intermediate taxa).  So my system is much more
> flexible than the traditional Linnean System, with fewer formal taxa (and a
> WHOLE lot fewer than PT taxonomy).

It seems to me with less ranks, your system is far more vague, and hence
not up to the task of accomodating new data about relationships.
Phylogenetic taxonomy can expand to fit any number of new discoveries.

> This makes it far more practical for non-systematists,

Well, who's this system being designed for? Professionals or middle school
students? (No offense to any middle school students out there, some of
whom may become professionals one day.)

The great thing about PT is that it has the level of detail that
specialists require. Of course most of this detail isn't of use to the
non-systematist, and they don't really need to learn it. Fortunately, the
phylogenetic system is *very* easy to abridge. While a specialist or
interested amateur may appreciate the fine detail of a chart such as the
one at <http://dinosauricon.com/taxa/ornithopoda.html>, an introductory
student can get by on:

--Ornithopoda
  |--_Hypsilophodon_
  `--Iguanodontia
     |--_Iguanodon_
     `--Hadrosauridae

> who can usually just ignore the coding (details of
> relationship) and have a very stable and relatively uncomplicated,
> utilitarian classification.

The diagram above is pretty stable and uncomplicated.

I'd also like to say that, IMHO, stability sometimes reflects stagnation
in the research. In fact, when phylogenies change, traditional systems
seem to feel a pressure to keep the status quo, whereas PT takes it in
stride.

> And it doesn't strip away the "divergence"
> component of traditional classifications.

A component which is very imprecise. As has been argued on this list, you
could make an excellent case that _Triceratops_ has diverged more from the
ancestral dinosaur than _Archaeopteryx_ has. Yet who gets to be in a whole
new class?

At any rate, I don't really see how a traditional classification indicates
divergence better than a cladogram. (It does indicate the author's
subjective feelings about amounts of divergence better, though!)

>       Therefore, I cannot see how you can justify the use of phylogenetic
> taxonomy, which has few advantages over a modified Linnean system,

Extensibility, objectivity, clarity .. I guess those are a few.

> and PT has some terrible disadvantages, such as: (1) hierarchical
> instability,

Only insofar as phylogenetic hypotheses are unstable. And if they are,
shouldn't the classification reflect it?

> (2)  a huge multiplicity of names

?!?! That's a *benefit*, to my way of thinking. We can now discuss groups
that would be unnameable under the rigidness of the traditional system.

> (many which will turn out to be invalid),

Unless given a qualifying clause or synonymous specifiers, phylogenetic
taxa *cannot* be invalid. That's one of the key advantages.

> and admit it or not (3) even an enormous number of ranks-----cladists
> don't always label them as ranks, but indented classifications are
> ranked whether you name the ranks or not.

So what? I fail to see how this is a problem, as long as the ranks are
relative and unnamed, not absolute and labelled.

> And all these disadvantages will be exacerbated
> by the PhyloCode as workers rush in to establish priority for their favorite
> names (with their names attached as the author).

Hopefully the initial companion volume to the first printing of the
PhyloCode will address this issue. And, as with the current codes, appeals
can always be made.

>       The disadvantages of strict PT are so great (as Benton has written in
> his recent review) that it seems far more prudent to modify the Linnean
> System instead, for a variety of reasons.

Linnaean taxonomy was created before the advent of evolutionary theory in
accordance with philosophies (Scala Naturae, Platonic ideals, etc.) that
are simply not relevant in modern biological thought. Phylogenetic
taxonomy, OTOH, provides guidelines for a precise representation of the
branching structure of life caused by desent with modification. To my way
of thinking, trying to improve the Linnaean system to keep up with modern
thought is about as viable as breeding a new strain of horse to keep up
with modern traffic.

_____________________________________________________________________________
T. MICHAEL KEESEY
 Home Page               <http://dinosauricon.com/keesey>
  The Dinosauricon        <http://dinosauricon.com>
   personal                <keesey@bigfoot.com> --> <tmk@dinosauricon.com>
    Dinosauricon-related    <dinosaur@dinosauricon.com>
     AOL Instant Messenger   <Ric Blayze>
      ICQ                     <77314901>
       Yahoo! Messenger        <Mighty Odinn>