Re: Reptilia; dispelling "delusions"

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

>       In vertebrates, osteological characters obviously give us the best
way
> to classify overall.

Trivial in fossils.

> That is why scientists have long characterized
> Mammalia using the transformation of certain jaw bones into the three ear
> ossicles (rather than using hair or endothermy, or anything else the
fossil
> record rarely preserves).  I would like to see that same thing happen to
> Aves---based on osteology (not a crown group or phylogenetically
anchored).

Unlikely. Such "key characters" are prone to split when more fossils are
discovered. This has happened to your Mammalia example, which lumps several
events into one character. Apart from the fact that the stapes has "always"
been involved in hearing and the quadrate + quadratojugal (malleus, anvil)
has "always" contacted the stapes, the lower jaw stuff that makes up the
incus (hammer) in crown-group mammals has been involved in hearing, and
probably supported an eardrum, since some early therapsids. The clade
(Tritheledontidae + Mammalia) has AFAIK* evolved the dentary-squamosal jaw
joint but originally retained said "lower jaw stuff"** connected to the
lower jaw. Either the crown group or (crown group + *Hadrocodium*) broke the
connection between splenial and prearticular, respectively between the lower
jaw and the middle ear.
Non-crown-group Mammalia is diagnosed on braincase characters anyway,
otherwise *Adelobasileus* (a Late Triassic braincase) could never ever be
accepted as the basalmost mammal by non-crowngroupers (e. g. in Benton's
book). _You_ suggested*** to define Mammalia as a node containing
*Adelobasileus* rather than having a crown group! :-)

* Judging from the name *Diarthrognathus* (two-joint-jaw) which refers to a
tritheledontid; I have found suggestive but no explicit wordings in the
literature. Tritheledontidae = Ictidosauria has traditionally been regarded
as the just-not-yet mammals.
** The "postdentary bones" -- articular, prearticular, angular and
surangular.
*** Maybe someone has published such a definition, I don't know.

> Reptilia have a common ancestor (first amniote), so
> although excluding two exgroups is admittedly arbitrary, it is still
> natural, useful, and has a long tradition as well.

I can see how something arbitrary can (sometimes) be useful. I can't see how
it can be natural. Please explain.

>       If only we had such an osteological character complex [...]. In
> my opinion that would be the most effective way to battle Feduccia's BAND
> crusade.

Why?