Re: Archosaur Origins...was:MESENOSAURUS ERRATA.

[Dinogeorge@aol.com]

In a message dated 8/23/01 6:52:40 AM EST, larryf@capital.net writes:

<< Interesting George,..and though I agree with you that prolacertiformes 
could
 be ancestral to archosaurs, I don`t believe Dave Peters sees it that way. I
 get the impression that his claim is that Pterosaurs are not archosaurs, and
 don`t belong to the group Ornithodira. Characteristics such as preorbital
 fenestra, and AM ankle joints are just convergences.
 
 I`m curious,...are you implying that since birds have a prolacertilian
 ancesty, they might be considered close sister group to pterosaurs? >>

This has quite a bit to do with taxonomy, so let me get into that a bit. Back 
in the old days, before fossils were incorporated into taxonomy, the Linnaean 
system was essentially a crown-group classification. Evolution had not yet 
been conceived of, and there was no plethora of fossil taxa that needed to be 
classified. Crown groups are pretty easy to define by morphology and 
typology, because the stem organisms intermediate between crown groups--the 
organisms that don't have all of the type features of the crown groups--are 
extinct (that's what being a crown group is mainly about), and that's how 
taxonomy was done then.

Once evolution became the paradigm of taxonomy, we should have simply 
expanded the definitions of the crown groups into all-inclusive stem groups. 
For example, we had Aves as the crown group that comprises all modern birds. 
We >should< have expanded Aves to be the stem group that includes all animals 
more closely related to modern birds than to their next-nearest stem group, 
namely, Crocodylia. For historical reasons (e.g., the fossil record didn't 
come to us in any pre-organized way), however, this wasn't done; ideas of 
classification take time to percolate. Now that we have a good enough handle 
on evolution, taxonomy, the fossil record, and so forth, we can with no 
difficulty expand all the high-level crown groups into stem groups. This is 
the most natural way to extend the old Linnaean-style high-level crown-group 
taxa to include fossil taxa.

Specifically applying this method to the tetrapods, we get (for example):

Aves: all animals more closely related to modern birds than to modern 
crocodiles
Crocodylia: all animals more closely related to modern crocodiles than to 
modern birds
Archosauria: all animals more closely related to Aves and Crocodylia than to 
modern lepidosaurs
Lepidosauria: all animals more closely related to modern lizards and snakes 
than to Archosauria
Diapsida: all animals more closely related to Archosauria and Lepidosauria 
than to modern turtles (if turtles really are "anapsids" and not reversed 
diapsids/archosaurs)
Chelonia: all animals more closely related to modern turtles than to Diapsida 
(unless as noted above they're actually >in< Diapsida)
Reptilia: all animals more closely related to Diapsida and Chelonia than to 
modern mammals
Mammalia: all animals more closely related to modern mammals than to Reptilia
Amniota: all animals more closely related to Mammalia and Reptilia than to 
modern amphibians
Amphibia: all animals more closely related to modern amphibians than to 
Amniota
Tetrapoda: all animals more closely related to Amphibia and Amniota than to 
modern bony fishes
Pisces: all animals more closely related to modern bony fishes than to 
Tetrapoda

and so on

What goes on taxonomically >within< these high-level groups at a lower level 
is, of course, another story.

This method partitions the entire Tree of Life with no gaps. All fossils 
automatically have a group to which they most naturally belong. In 
particular, all dinosaurs are classified as birds (a position I've held for a 
long time now), since they're closer to modern birds than they are to 
crocodiles. Crocs, rauisuchians, phytosaurs, aetosaurs, and many of those 
other wacky "thecodontian" groups would be classified as Crocodylia. I 
believe prolacertiforms and pterosaurs fall into the archosaur subgroup of 
Reptilia in this system, whereas fossil marine reptiles such as plesiosaurs 
and ichthyosaurs fall into basal Diapsida somewhere below archosaurs and 
lepidosaurs. Mosasaurs are, of course, well nested in Lepidosauria. 
"Pelycosaurs" and other therapsids become very basal mammals. Lungfishes wind 
up as basal tetrapods (so what? I'm ok with that).

One good thing about this system is that it dispenses with all those 
contrived "Archosauriformes," "Dinosauriformes," "Eudinosauria," "Sauropida" 
(not Sauropsida, because, unless I am mistaken, it derives from the root 
-ops) and "Theropida" (not Theropsida) kinds of groups required if we fix 
nodes for the high-level groups somewhere within their inclusive stem groups. 
We might find some use for these kinds of groups at lower levels in taxonomy, 
but the high-level stem groups certainly should retain the straightforward, 
more widely known, and least nit-picky names.