Re: questions about the Odontochelys study

[Tim Williams <tijawi@yahoo.com>]

Mike Keesey wrote:


> > And I don't recommend making such a name up just
> yet :-)
> 
> Why not, as long as the definition collapses should
> archosaurs prove
> to be descended from the final common ancestor of
> testudines and lepidosaurs?


I agree.  Including the right negative/external specifiers in each definition 
would ensure that a Testudinata-Archosauria clade and a 
Testudinata-Lepidosauria clade were mutually exclusive.  The respective 
definitions would ensure that only one (at most) could be used in any topology. 
 This allows a given name to be explicitly linked to a specific phylogenetic 
hypothesis (turtles closer to archosaurs than to lepidosaursm or vice versa).

For example, if a Testudinata-Archosauria clade (let's call it 'Chelarchia') 
was defined as the least inclusive clade that includes turtles and birds, but 
not lizards, and a Testudinata-Lepidosauria clade ('Chelolepida') was defined 
as the least inclusive clade to include turtles and lizards, but not birds, 
then if the former clade was recovered, then Chelolepida would collapse 
(self-destruct).  

If turtles (Testudinata) were recovered as outside the Archosauria+Lepidosauria 
clade (a third phylogenetic hypothesis), then neither name is viable. 


Cheers

Tim