[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Subject Index][Author Index]
Re: Syntarsus [still longish]
Since Mickey has more than adequately responded to much of
Dinogeorge's commentary on cladistics, I will try not to respond to the
points he touched on, except where I have something new to say.
One thing I think it would be useful to remember is that a cladogram
is a _phylogenetic HYPOTHESIS_, which must be tested with new charcters and
new approaches. Gauthier 1985 (85?) was tested, and refined, now we have
Holtz 1994 and Seus (1995?) and various other phylogenetic hypothesis which
will be tested and transformed until well beyond our times. In fact, this
is what I find so compelling about systematics etc, that it is a search for
a "truth" we cannot ind, and therefore we can have endless fun trying.
At 11:50 AM 9/11/96 -0500, Dinogeorge writes:
">>"s are my writing
>is inherently subjective; and (3) no reliable method exists to weigh
>morphological characters against one another (for example, how many skull
>characters is a postcranial character worth? one? one-half? ten?)...
You have an excellent point, and sums up very well why I
*don't* weigh characters. I have yet to be convinced that there is a
basis for weighting characters at all. My logic may be flawed on
this, but it seems to me that parsimony is parsimony, and enough
binary, equal-weighted characters will produce a more scientific study
than fewer, more "intelligently" weighted characters. (Don't get me
wrong, I have the utmost respect for the judgement of those
paleontologists who weight their characters).
>If at most one cladogram for any group is the correct one, virtually all the
>cladograms so generated >must< be wrong.
If I had a nickel for every scientific hypothesis (or theory even)
which has later been demonstrated to be wrong, I would never have to
"...consume mass quantities of grant money producing endless streams of
cladograms with [my] computer..."
>> preaption .... Basically, the idea is that an
>> organism can have the potential to develop a character without
>> actually expressing it, which said potential can be passed to
>> daughter species, which then may or may not express that character(
>> BOY does it sound hokey when it's written out like that).
>This is a tautology. How will you know when the potential to express a
>character is present, except if the character is expressed?? You DARE to
>criticize paleontologists at the same time that you try to pass THIS off as
(Was *that* a tauNtology?) It is a hypothesis for explaining very
close parallel evolution, specifically, why these two animals can have an
allegedly very similar structure which is not represented in their putative
sister taxa. I understand that this is a complex subject, which requires
looking much more deeply into evolutionary theory than I am probably
qualified to speak on, but I provide what I hope is a sufficient explanation
below. Suffice it to say that I am well aware of how these statements
sound, and I had hoped that they would inspire reflection and not knee-jerk
As for the nature of science, science is hypothesis, experiment,
interpretation, theory, test, theory, test... ad infinitum. I do not know
of a theory or experiment refuting my basic hypothesis, which is based on
my, admittedly incomplete, knowledge of evolutionary theory.
I do not criticise paleontologists (are we now gods among men,
immune from reproach? I like it!), I examine their theories. I will "DARE"
any day of the week to discuss a topic intelligently with another person,
and if my science is wrong (or more likely poorly worded), I will do what
the rest of us proles do and look silly.
>> In the example we are using here, the ur-coelophysoid may have had
>> everything it needed to develop crests (in fact, probably did),...
>Yeah--like a SNOUT!
See my "clarification" posting...
>> conclusion). This is a gross oversimplification, but you can see
>> the danger in picking one trait to define your phylogeny on.
>Sorry, but I have no idea what you're talking about.
That's what I've gathered from your phylogenies. It comes
back to weighting. Mr. Williams was basing his phylogeny on extreme
weighting of the crests of Dilo. and Syntarsus k. My point was that
this could have been extremely short-lag parallel evolution, where the
common ancestor was headed that way, but the coelophysid common
ancestor speciate crestless, while Dilophosaurus (I'll skip
Liliensternus for now) excercised it's clade-given right to sprout a
double-bone-mohawk. Then, as the Coelophysids radiated, one (or more)
of their species also climbed the morphological hill to crest-land.
It really is just delayed parallel evolution, with a little story
about why it might have happened attatched.
>I have wings, but the trouble is they are >unexpressed< and look and function
I thought your arms were secondarily re-adapted from wings (BCF, not
OIDD :). They've been trying to steer us away from that concept for years,
but in a way it is true. I don't think it's such a stretch to imagine that,
from a certain genotypical, phenotypical, or behavioral aspect, certain
structures and behaviors are both easier and more likely to produce the same
result. This has been commented on before, it is a subset of the discussion
of parallel evolution, but could just as easily be called "parallel exaptation".
Most of the structures in organisms, or, more appropriately, the
genetic code for them, are, in a way, overbuilt, ready to be modified for
some other purpose, should the need arise (gross exaggeration and
oversimplification). We used to marvel at how nature had shaped each
organism "precisely for purpose", or however that went. In truth, evolution
is a packrat, always hedging it's bets. Why do we have an appendix? I
don't know the current thinking, but it sounds to me like the risk of
appendicitis is less than the possible advantage our genome might gain
eventually by retaining it.
Is an organism whose ontogeny recapitulates its phylogeny more
likely to be "evolutionarily fit" than one that isn't? You bet! How many
evolutionary innovations are linked to paedomorphosis (birds)/ retention of
embryological characteristics (the Anhinga's claw). Obviously, nature never
thought "oh, I oughtta save a little DNA for a rainy day", but, over the 3
1/2+ billion years of evolution, I'd be surprised if conservatism has not
been selected for strongly.
All of this leads to a great many special evolutionary effects that
seem impossible. You have written to me that you do not believe that
Therizinosaur ancestors could regrow all of the soft tissue in support of
re-development of their fourth toe. Yet they could, simply because a
geneome doesn't like to throw out "useless" sections, especially ones that
were recently quite useful, it files them away against a future need (once
again, a gross oversimplification).
>> I still say that if you're gonna base your phylogeny on a "by hand"
>> analysis of characters, try not to bas it on characters that have a
>> *direct* effect on reproduction, feeding strategy, movement, or
>But--if you do your phylogeny by computer, you can, of course, include all
>these kinds of characters and many more besides into the mix.
Exactly my point. Yes, there is a danger here (Gauthier's
Carnosauria yet again), but with enough characters, this should filter out.
Remember, it's not how many characters A and B have in common, it's how many
more they have in common than either does with C.
>Since cladistic analysis of dinosaurs is a hopeless, inbred mess, with
>everyone using everyone else's character matrices and so forth, you might
>just be better off doing Tarot readings or something.
It's funny you should put it that way, because last I checked,
nearly everybody (Gauthier, Padian(?), Sereno, Currie(?), Seus, Holtz, Xiao,
Lehman, Rowe...) uses cladistic methodology. No wonder Bakker's split
academia, they're all using hopeless, inbred methods!
BTW: I tried dowsing once. I know I'm not good, but you've never
seen a pendulum that still!
>> Anyway, the only way you're going to convince me of this is to set
>> up a data matrix...
>Why would this be any more convincing than any other reasonably organized
Because it would be carried out in a well-documented, reproduceable
manner, with some degree of control. If I recall correctly, those are at
least some of the requirements of a scientific experiment, which is how we
test theories. While I do not think sittin' around head-cladein' is bad, I
would not be convinced by a cladogram that was not done in at least a
recognizably scientific manner.
>>If you use at least 40
>> characters and get Dilophosaurus and S. kaentakae as a clade, I
>So your particular convincement-cutoff is 40 characters. Why? Why not 39 or
Why not 3? That seems to be how non-cladists do it.
Ok, maybe 40s a bit much, but that's not your point. It's the
thought that counts. I am not convinced by a cladeogram with eight taxa and
fifteen characters, especially if it refutes a well established phylogenetic
hypothesis. Granted, I think it'd take me a good couple years or so to come
up with forty characters for such a study, but when I did, if it came out
with that result, I'd run straight to the press with it.
At 01:04 PM 9/11/96 -0500, Stan Friesen wrote:
>This was, indeed, my reason for saying that a display structure like
>a crest, is of limited value in phylogenetic inference. They often
>change rapidly in closely related forms for just this reason.
But, on the other hand, a character is a character.
| ********************PLEASE NOTE NEW E-MAIL ADDRESS****************** |
| Jonathan R. Wagner "You can clade if you want to, |
| Department of Geosciences You can leave your friends behind |
| Texas Tech University Because your friends don't clade |
| Lubbock, TX 79409 and if they don't clade, |
| email@example.com Then they're no friends of mine." |