Warm bloodedness in archosaurs before dinosaurs

[Joshua Dyal <j-dyal@geocities.com>]

Having read Bakker's _Dinosaur Heresies_ and thinking that logically, if
not empirically, his idea of warm-bloodedness developing first among the
therapsids was a good one, I have come to a curious question.  

Part of the argument was also that the thecodonts, to use the informal
term, had to have developed endothermy of some degree to compete
successfully with the therapsids as we moved into the Triassic.  Here I
found what may be a little snag in the hypothesis.

I was looking at a (slightly) controversial cladogram of diapsids (found
in _the Dinosauria_) and wondering at what point some degree of
endothermy may have developed.  From Bakker's book, I had almost gotten
the impression that the thecodonts must have mostly been endothermic,
but to my surprise, I don't see how that can be.  The reason is that the
crocodilians are very high up on the cladogram, cutting off at least
half of the thecodonts.  I suppose it is possible, especially if
warm-bloodedness were "new" to the archosaurs, that crocodilians could
have dropped their metabolism back to ectothermic levels.  It is also
possible that endothermy developed in parallel, once right before the
_Ornithosucia_ node, and again at, maybe, the _Pseudosuchia_ node.  Does
anyone have any ideas about this?  At what points along the archosaur
cladogram are there possible stages where warm-bloodedness could have
developed?  Just to help you in helping me, I have attached a jpg image
that I threw together of the cladogram, with the crocodylotarsi branch
inferred from the text.  I ran out of room at the top, so one branch
actually turns downward, but just ignore that and pretend it goes the
direction it should  :)  Thanks everyone!

-- 
Joshua Dyal
j-dyal@geocities.com


P.S. Thanks to the group in general for the incredible patience with my
rather unlearned questions.