combined answer Re: Are diadectomorphs really reptiles?

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

> Casineria has a sister taxon from younger strata that does have a skull.
> Even without that Latter Day cousin, you can still put Casineria into a
> cladistic analysis and arrive at another conclusion.

I will put it into my thesis, and then I'll see. Couldn't you publish in the 
meantime? You know, you make lots of statements of fact that we simply cannot 
judge because you haven't published your reasoning...

> Solenodonsaurus and Diadectomorpha as amniote cousins? That's tradition.

No, it's Laurin & Reisz 1999 (and Vallin & Laurin 2004). L & R 1999 redescribed 
the previously much-neglected *S.* at length and then added it to their data 
matrix.

> And even you will have to admit, they stand out like two cats at a dog
> convention.

Not at all, no. If anything, they stand out like *Borophagus* at a dog 
convention. Do you have the monograph on *Tseajaia* (Moss 1972, U. of Cal. 
Publications in Geological Sciences 98)?

> Go back to the chalkboard and you'll find a spectral continuum of
> morphological change in sister taxa.

Under the indefensible assumption that the fossil record is good enough.

> Diadectes never had gills!

1) We don't know that. Remember that *Siren* is heavily herbivorous, it's even 
a hindgut fermenter.

2) So what? Nobody said the amniote egg cannot have evolved before the origin 
of Amniota -- Amniota has a node-based, not an apomorphy-based, definition. All 
we can tell is that lepospondyls plesiomorphically had gilled larvae (gills are 
occasionally preserved in *Microbrachis*) and that amniotes don't; the ontogeny 
of *Solenodonsaurus* and Diadectomorpha are anyone's guess at present.

> <There was, in most cases (the absence of lepospondyls being the most glaring 
> exception), no opportunity to provide closer relatives.>
>
> Look again. An entire lineage of closer relatives from an unlikely source.

Why don't you either just tell us or publish?

> <Enough are known to provide most of the topology of the tree, yes. But not 
> enough are known that we could watch the evolution of every character.>
>
> Look again. You're not using the best inclusion set of taxa.

I'm not using anything at present, because I haven't run any analysis yet.

> re: scales and synapsids.
> > > Given that ichthyostegids were scaled like other osteolepids,
> > > scales may have been present on the earliest amniotes, which go
> > > back almost as far. Not the same sort of scales, but then, what
> > > the hey.
>
> <Nope. These scales are bone plates in the dermis (and present not just in 
> *Ichthyostega* -- there are probably no other "ichthyostegids" --, but also 
> present all over the body in almost all tetrapods except lissamphibians, 
> diadectomorphs and amniotes; I'm not sure about seymouriamorphs); amniotes 
> plesiomorphically retain the ventral scales -- that's what the gastralia are. 
> Sauropsid scales are thickenings of the epidermis, consisting of nothing but 
> keratin. Calling both "scales" is highly misleading.>
>
> Evolution changed one form of scale to another. The loss of scales
> is another branch on the tree. 

You can't simply assert that bone plates in the dermis are homologous to 
thickenings of the horn layer of the epidermis. You have to demonstrate it.

> Westlothiana is in the ingroup. Nests with Paleothyris which has
> similar supratemporals and tabulars and lots of other synapomorphies.

Then why didn't it in Vallin & Laurin 2004 or Ruta & Coates 2007? And why is it 
covered in dermal scales all over? That would be an interesting reversal... as 
would be its lack of an astragalus...

> re: Diadectomorphs as stem-reptiles.
> [...]
>
> Just line up the "usual suspects" and you'll see immediately that some
> prior placements just don't make sense. There are better, more gradual
> sequences that include more synapomorphies. Spoiler: Diadectormopha is
> part of a continuum with living ancestors. Not an evolutionary dead
> end.

Everything that isn't an ancestor is an evolutionary dead end, so of course the 
diadectomorphs are a dead end, no matter if they are amniotes or not. Are you 
saying they are paraphyletic (like the bizarre result of Ruta & Coates 2007)?

> I have great respect for this particular hypothesis of Carroll's. It's
> the only one that makes sense and is supported by cladistic analysis.

Which hypothesis of Carroll's? Carroll has never considered the diadectomorphs 
amniotes or paraphyletic.

> re: Spinoaequalis
> Excellent! Now you're getting somewhere!  Find more like this one and
> you've got your lineage. Credit where credit is due.
>
> <What lineage?>
>
> The original question was all about Petrolacosaurus. Remember?

Well, *S.* is still a branch of its own, with its own autapomorphies. (And the 
phylogenetic position in the description appears to be a bit off -- W4tP.)

> <We can arrange all these taxa in a tree. It just so happens that some
> of the side branches have diversified on their own, forming a
> suprageneric taxon. They are there, dude. You can't deny them.>
>
> The problem with nearly all prior cladistic analyses is there use of
> suprageneric taxa. Suprageneric taxa, whether you realize it or not,
> are made up of individual taxa with body parts. Just pick from the
> genus category and the confusion of suprageneric chimaeras will
> dissipate and you'll see clearly. It's actually less work because you
> no longer get to cherry-pick characters from a number of specimens to
> suit your cladogram. No more decisions to make. "Is you is or is you
> ain't."

You are talking about two different things here. The one is using suprageeric 
OTUs (which I agree should be avoided). The other is _finding_ clades composed 
of several genera as each other's sister-groups. I got the impression you 
denied that the latter could even happen.

> What I'm looking for is this sequence: Euparkeria ?  ?  ?  ?
> Herrerrasaurus. Then we'll compare notes.

For the last time, it's not a sequence, it's a tree.

It is not

*Euparkeria*--?--?--?--?--*Herrerasaurus*

, it is

  ,--------------*Euparkeria*
--|--------------*Osmolskina*
  `--+-----------?
     `--+--------?
        `--+-----?
           `--+--?
              `--*Herrerasaurus*

, with time running from left to right and NOTHING running from top to bottom 
or vice versa, with each "+" indicating a branching point, with

,
|
`

indicating an unresolved trichotomy, and with the whole thing being a mobile -- 
you can rotate each branch around the long axis of its supporting internode, 
and exchange the three branches of the trichotomy at will, and nothing changes.

And for that tree, I'll refer you to the literature, though it is unfortunate 
that no work on archosauromorph phylogeny has been done for about 10 years, the 
polytomy of the *Qianosuchus* paper excepted.
-- 
249 Spiele für nur 1 Preis. Die GMX Spieleflatrate schon ab 9,90 Euro.
Neu: Asterix bei den Olympischen Spielen: http://flat.games.gmx.de