RE: Philosophies for Character Ordering

[Mike Milbocker <mmilbocker@psdllc.com>]

In reference to the below discussion concerning ordered and unordered
characters, it should be noted that a single tree has been obtained in the
ordered case because an additional constraint, other than parsimony, has
been added - namely the notion of ordered characters. The ordering scheme
itself presupposes something about the evolution. If you're going to go down
this path, you may as well do your associations based purely on the ordering
concept, and dispense with cladistics. When you say that a certain number of
caudal veterbrae are more primitive than another, you are making an unproven
statement about evolution.

Regards,

Mike Milbocker



Mickey Mortimer wrote:

>>Now, if these are all ordered, we get one tree.
(Tyranno(Ornithom(Confuc,Protopt,Sinornis))).
The fact Confuciusornis and Protopteryx have an intermediate state between
Ornithomimus and Sinornis for characters 1-3 DOESN'T make them form their
own clade to the exclusion of Sinornis.  Just like the fact Ornithomimus,
Confuciusornis and Protopteryx have an intermediate state between
Tyrannosaurus and Sinornis doesn't make these three form their own clade to
the exclusion of Sinornis.  This is how almost everyone codes characters.

Using Wilson's philosophy, leaving the characters unordered, gives us four
equally possible trees.
(Tyranno(Sinornis(Ornithom(Confuc,Protopt))))
(Tyranno(Sinornis,Ornithom(Confuc,Protopt))))
(Tyranno(Ornithom(Sinornis(Confuc,Protopt))))
(Tyranno(Ornithom(Sinornis,Confuc,Protopt))))
Notice how in three of these trees, Confuciusornis and Protopteryx clade
together to the exclusion of Sinornis because of their intermediate state?
And how it can't tell Sinornis is closer to Confuciusornis and Protopteryx
than Ornithomimus is, despite the fact the birds are more similar to each
other in characters 1-3?  I have a problem with this, and I'm sure others do
too.<<<














-----Original Message-----
From: owner-dinosaur@usc.edu [mailto:owner-dinosaur@usc.edu]On Behalf Of
Mickey Mortimer
Sent: Thursday, May 20, 2004 10:20 PM
To: dinosaur@usc.edu
Subject: Re: Philosophies for Character Ordering


Jaime Headden wrote-

>   Intermediate states are already their own state. Given a matrix where
> all characters are of equal weight, and previous states in a
> transformation sequence don't tally-total up the numbers for a particular
> character, as in an 8-sacralled animal having a "better" number if it
> gained the two extra vertebrae over a 6-sacralled one sequentially (went
> through a 6, then 7-sacral state), than if a second animal gained it's two
> extra sacrals in one event. A matrix should already be able to treat
> *Protopteryx* and Confuciusornithidae as sistergroups given enough
> synapomorphies in an unweighted system. Just as the avian looking basal
> dromaeosaurids would seem to make a clade based on shared features that,
> rather, could easily be step-wise sequence transformation towards the
> larger, more terrestrial dromaeosaurids, yet their similarities would be
> prone to group them together exclusive of other dromaeosaurids, in many
> observations (as in, say, Paul [2001]);

Intermediate states are of course technically their own state, but they
don't cause the intermediate taxa to clade to the exclusion of derived taxa
if the states are ordered.  I made my own test matrix to show this.  It's
very simple-
Tyranno   00000
Ornithom 00011
Confuc     11111
Protopt    11111
Sinornis   22222
Imagine these are characters like "caudal vertebrae - number - 45 (0); 20
(1); 10 (2)" or "manual digit III phalanges - number - four (0); two (1);
one (0)", although these exact characters aren't correct for these exact
taxa.

Now, if these are all ordered, we get one tree.
(Tyranno(Ornithom(Confuc,Protopt,Sinornis))).
The fact Confuciusornis and Protopteryx have an intermediate state between
Ornithomimus and Sinornis for characters 1-3 DOESN'T make them form their
own clade to the exclusion of Sinornis.  Just like the fact Ornithomimus,
Confuciusornis and Protopteryx have an intermediate state between
Tyrannosaurus and Sinornis doesn't make these three form their own clade to
the exclusion of Sinornis.  This is how almost everyone codes characters.

Using Wilson's philosophy, leaving the characters unordered, gives us four
equally possible trees.
(Tyranno(Sinornis(Ornithom(Confuc,Protopt))))
(Tyranno(Sinornis,Ornithom(Confuc,Protopt))))
(Tyranno(Ornithom(Sinornis(Confuc,Protopt))))
(Tyranno(Ornithom(Sinornis,Confuc,Protopt))))
Notice how in three of these trees, Confuciusornis and Protopteryx clade
together to the exclusion of Sinornis because of their intermediate state?
And how it can't tell Sinornis is closer to Confuciusornis and Protopteryx
than Ornithomimus is, despite the fact the birds are more similar to each
other in characters 1-3?  I have a problem with this, and I'm sure others do
too.

> <This makes more sense, but I'm confused by an aspect of it.  Say the
> character is "antorbital fenestra size".  Wilson would say, the pneumatic
> diverticulum that occupies the fenestra could halt its growth during
> development, so it should be treated as an easy loss character.  But
> structures usually grow in concert with another structure shrinking. When
> the diverticulum enlarges, the surrounding bone shrinks.  So why not treat
> the bone growth as easy loss instead of the diverticulum growth?  Instead,
> the bone growth is effectively "easy gain", which no developmental
> philosophy supports.>
>
>   Easy to do if one includes the clause that the two be the same relative
> condition. However, even as diverticulum growth increses, bone should not
> immediately be seen to "shrink," only it's relative size decrease. The
> bone, if it stops growing, and the antorbital fossa increases in area,
> would still be the same size, as in the aperture of the antorbital
> fenestra. Or diplodocoid vertebral pleurocoels, in which case the
> pneumatic diverticula would be present early in life, and the bone grows
> around the diverticula, creating smooth-surfaced chambers and rounded rims
> to any aperture or lamina dividing diverticula.

The bone volume doesn't "shrink" in a developmental sense, but it does
(relative to the rest of the animal) in an evolutionary sense.  I meant you
could change the character to "maxillary bone extensive" or something, to
reflect the fact more bone around the antorbital fenestra has to grow in
order to make the fenestra relatively smaller.  And if this bone's growth is
halted early in development, the antorbital fenestra would be larger.  You
can think of most size-change characters in two ways, each which results in
weighting a different way, if you use a directional ordering assumption like
"easy loss".

Mickey Mortimer