Dinosauria Defined

From: "King, Norm" [nking.ucs@smtp.usi.edu]

I believe I read somewhere (I think it was here) that we should define dinosaurs as all the descendants of the most recent common ancestor of Megalosaurus and Iguanodon.

From: Thomas_R_HOLTZ@umail.umd.edu (th81)

A brief history.

Gauthier (1986) formally defined "Saurischia" as birds and all taxa closer to birds than to Ornithischia.

Padian and May (1993) formally defined "Ornithischia" as Triceratops and and all taxa closer to Triceratops than to Saurischia.

Dinogeorge@aol.com

Since Saurischia is the same thing as Dinosauria, because ornithischians descended from certain primitive saurischians, this definition is vacant. There are no taxa closer to Triceratops than to Saurischia, once you realize that Triceratops descended from/is nested within Saurischia/Dinosauria. Other people may have different opinions, of course; but I've examined all the characters I can find that supposedly unite and distinguish Saurischia from Ornithischia, and they're all questionable: either plesiomorphies, or apomorphies shared with some prosauropod taxa, or just plain ill-defined.

From: Thomas_R_HOLTZ@umail.umd.edu (th81)

At the 1994 meeting of the SVP, while talking to Kevin Padian, I mentioned the fact that I regreted the above authors not using Megalosaurus and Iguanodon instead of birds and Triceratops. This would have preserved two of the original three (the last is Hylaeosaurus) of Owen's 1842 taxon "Dinosauria". It would also reflect an important historical point: birds are NOT central to our original understanding of dinosaurs, and are considered dinosaurs only because they fall within that monophyletic clade.

Padian agreed it would have been a better idea, but it is too late for that. I use Meg and Ig as the "type" dinosaur taxa in my class informally.

From: "King, Norm" [nking.ucs@smtp.usi.edu]

I can provide that definition to my students (but I will have to follow up with an explanation as to how they can recognize a dinosaur when they see one). I predict that these questions will be asked about the cladistic definition:

Why are those two genera singled out? Isn't it true that after all these years we still do not have even one complete composite skeleton of Megalosaurus? Why not choose Allosaurus, instead?

From: Thomas_R_HOLTZ@umail.umd.edu (th81)

These taxa were singled out because they were two thirds of the original concept of Dinosauria. Megalosaurus is more complete than most think (work by Freidrich von Heune and grad student Laura Canning have shown/are showing that), and although not as complete as Allosaurus, is clearly a theropod and not some non-dinosaurian creature.

From: "King, Norm" [nking.ucs@smtp.usi.edu]

Do we really know when ornithischians branched off (separated?, whatever--what would the correct cladistic term be?) from saurischians to know that we're not leaving behind something like Eoraptor or a herrerasaurian? Let me rephrase that so no one here will jump on me for misunderstanding _Eoraptor_ and herrerasaurs, which has nothing to do with the point. Might we be leaving something behind that many people would label as a dinosaur based upon its over-all anatomy (hips, sacrum, limbs, feet, shoulder, neck, etc.)?

From: Thomas_R_HOLTZ@umail.umd.edu (th81)

As for the last: yes, yes, yes. We (i.e., phylogenetic taxonomists) WANT to leave behind grade-based thinking of taxa. Taxa in this system are NOT defined by characteristics; they are soley defined by ancestry. An animal is a dinosaur if and only if it descends from the common ancestor of the two taxa picked.

Or, to go to the paper by Holtz and Padian at last year's SVP meeting, _Eoraptor_ and Herrerasauridae are dinosaurs only if they are shown (by the most parsimonious distribution of derived characters) to fall within the clade joining Ornithischia and Saurischia (or _Triceratops_ and birds, or _Iguanodon_ and _Megalosaurus_ informally). In some of our cladograms, they did; in others, they did not. Kevin presented the most divergent cladogram (i.e., the one in which both _Eoraptor_ and Herrerasauridae are excluded from Dinosauria) to illustrate the extreme condition.

Of course, very close sister groups will be morphologically very similar to the most primitive members of a clade (because of descent with modification, and all that stuff).

From: "King, Norm" [nking.ucs@smtp.usi.edu]

What is the closest thing we have discovered to that most recent common ancestor of Megalosaurus and Iguanodon?

From: Thomas_R_HOLTZ@umail.umd.edu (th81)

Eoraptor. Whether it is within the clade or just outside of it, it clearly represents the closest thing we know to a common dinosaurian ancestor.

(Incidentally, this aspect of the Holtz and Padian talk at SVP seems to have gotten lost in all the discussions of the symposium (as has the fact that there WAS a second author to the paper, who presented it...)

From: "King, Norm" [nking.ucs@smtp.usi.edu]

As always, I (we) will appreciate any pointers.

From: Thomas_R_HOLTZ@umail.umd.edu (th81)

Jeff Poling wrote:

>I realized my article called "Dinosauria defined" does NOT
>have the definition of dinosauria.

The definition of Dinosauria is something you can get into public debates over... :-)

Under phylogenetic taxonomy, Dinosauria = all descendants of the most recent common ancestor of Triceratops and modern birds (Holtz & Padian, 1995, and in prep.), with Ornithischia = all taxa sharing a more recent common ancestor with Triceratops than with Saurischia (Padian & May, 1993) and Saurischia = all taxa sharing a more recent common ancestor with birds than with Ornithischia (Gauthier, 1986).

When I suggested how much nicer things would have been if they had used Iguanodon instead of Triceratops and Megalosaurus instead of modern birds, they thought it was a great idea. Unfortunately, under the principles of phylogenetic taxonomy, the first definition of a taxon under those rules has priority. Oh, well.