Re: Differences between *Vancleavea* and thalattosaurs

[David Marjanovic <david.marjanovic@gmx.at>]

>  Okay, my cards are on the table. Characters below shared with
>  Askeptosaurus OR Miodentosaurus or both (quite a variety between
>  them!) and not shared with proterochampsids AND erythrosuchids (the
>  bounding taxa for Vancleavea, sensu Nesbitt et al.)
>
>  [...]
>
>  2. Nasals separated by premaxilla ascending process (that's the
>  "neomorph" in Vancleavea and the related: Premaxilla contacts
>  frontals

You identified two correlated characters. :-) But what makes you so sure 
the neomorph is separated from the pmx only by a break? Don't you think 
that's the first thing Nesbitt et al. would have thought of? If you have 
the fossil in front of you, it's possible to distinguish sutures from 
breaks...

Apart from the fact that the pmx touch it, the neomorph reminds me of 
the interfrontonasal found in *Eryops* and the "microsaur" *Crinodon* 
(if that's not a dorsal exposure of the sphenethmoid, a regular 
occurrence in frogs).

>  3. Frontals longer than nasals

Just make sure this isn't counted already (as a ratio of snout length to 
postorbital skull length or something).

>  5. Upper temporal fenestra closed or slit-like and the related:
>  Postorbital-parietal contact long (extensive)

Again, you demonstrate you're fully capable of identifying correlated 
characters. :-)

>  6. Parietal skull table broad

Is it broad, or is it short? Broad relative to what?

>  8. Mandibular fenestra absent

Might be correlated to a strong bite, see tyrannosaurs. That would mean 
that presence/absence of the mandibular fenestra is only applicable to 
taxa with weaker bites.

>  9. Cervicals decrease in size anteriorly

Are you sure that's not a consequence of the relatively small skull?

>  10. Metacarpal II is the longest metacarpal

In us, too. No idea why.

>  11. M4.3+m4.4 fused

Impossible to find out without the fossil in front of you, and probably 
impossible even then.

Also, why did you (offlist) reconstruct the hand with a phalangeal 
formula of 2-3-4-4-4 instead of 2-3-4-5-3? The latter has the same 
number of bones and is way, way, way less freaky.

>  13. Metatarsals II-IV not shorter than half the tibia

Probably correlated to an aquatic lifestyle.

>  14. Metatarsals I and V are wider than II, III and IV

Assuming you identified them all correctly.

>  15. Pedal digit IV not narrower than III

That's plesiomorphic.

>  Granted, other than these characters, between Erythrosuchids and
>  Proterochampsids, one or the other includes one character or another
>  of Vancleavea, which in itself is amazing, but are we sure we don't
>  know any intervening taxa between the erythros and proterochampsids?

The Early Triassic archosaur record isn't as good as I'd like it to be. 
Except maybe in Russia which is crawling with supposed proterosuchids, 
erythrosuchids, and even rauisuchians that almost nobody has ever looked at.

>  Finally, I reduced my taxon list to match Nesbitt, et al. 2009. +
>  Askeptosaurus
>
>  Doing nothing else the tree resulted in 499 steps. [...]
>
>  moving the Proterochampsids to the Nesbitt order results in 510 MPTs
>  moving Sphenosuchus to the Nesbitt order results in 518 MPTs moving
>  Vancleavea to the Nesbitt order results in 566 MPTs removing
>  Askeptosaurus from inclusion so the set matched the Nesbitt set
>  results in 505 MPTs
>
>  moving Vancleavea to the outgroup or just inside Mesosuchus, where it
>  nested originally resulted in 492 trees and moving Vancleavea between
>  Proterosuchus and Erythrosuchus resulted in 501 trees.

Why do you bother giving us the number of trees? I don't see what your 
point is.

Surely you don't mean it as some kind of support measure...?

Also, I notice you didn't address my question as to whether you knew the 
difference between ordered and irreversible characters. That's important 
because, as long as you don't know this, you basically can't do 
phylogenetics yourself or understand how other people do it.