Re: Nomina Dubia Part II: Rapator

[Tim Williams <tijawi@yahoo.com>]

David Marjanovic <david.marjanovic@gmx.at>  wrote:


> I'm not sure if that's what you mean, but... just to make
> sure: there is no such thing as a character that is
> inherently diagnostic at a particular level.


I'm afraid you've lost me here David.  A specimen can be classified as 
diagnostic at different taxonomic levels.  For example, _Rapator_ is classified 
as "Megaraptora indet."  whereas _Walgettosuchus_ is classified as "Theropoda 
indet.".  This is because _Rapator_ (based on a metacarpal I) has characters 
that allow it to be assigned to Megaraptora, whereas _Walgettosuchus_ (based on 
a tailbone) can only be classified at the level of Theropoda.


> "A and B are different, but are they different enough to be
> classified as separate genera?" is not a scientific
> question.


Sure it is.  It may be subjective, but it is certainly scientific.  


> * I'm not saying this has been done; I lack an opinion on
> whether *Ceratops* should be considered a nomen dubium.


I don't.  I think it's a nomen dubium.  And I don't think I need a 
specimen-level statistical analysis to tell me so.  So let's bite the bullet, 
and junk the family Ceratopsidae.


> Concerning the last sentence, a nomen dubium is one that
> pertains to a specimen which cannnot be reliably
> distinguished from _at least two_ other taxa.


Yes, exactly.  What are we arguing about?


> Again,
> "reliably" does not mean at any particular level, because
> the levels don't exist.


OK, now you've lost me again.  These "levels" are hierarchial (increasingly 
inclusive) clades.  _Allosaurus_ is inside Allosauridae, which is inside 
Carnosauria, which is inside Theropoda, and so on...  


> *Australovenator* is the one that could be a junior
> synonym; *Rapator* is the one that could be a nomen dubium.
> *Rapator* is a nomen dubium if it cannot be distinguished
> from _both_ *Australovenator* _and_ *Megaraptor* by any
> characters other than those which can be shown to be due to
> individual (etc.) variation in... close relatives, the less
> close, the le
nd_ if furthermore
> *Australovenator* and *Megaraptor* _can_ be distinguished
> from each other by such characters. (If they can't be, all
> three must be lumped into *Rapator ornitholestoides*, unless
> you allow the geographical location of *Megaraptor
> namunhuaiquii* as a diagnostic character.)


I know what you're saying.  You and Mickey are taking the view that any 
character that is mentioned in a paper as being "different" between genera is, 
by implication, a diagnostic character.  Thus, you would argue that we need a 
separate study to demonstrate that said character is prone to intraspecific or 
ontogenetic variation in order to refute its validity as a diagnostic 
character.  


I say this is fruitless in most cases, because we don't have enough data either 
way - and probably never will.  So I err on the side of caution, and think we 
should avoid erecting new genera and species based on fragmentary material that 
have diagnostic characters that are a little shady (like the shape or 
orientation of a process on the metacarpal, or the length and orientation of a 
horn core).  This is current practice, BTW.  So what I'm proposing is not 
controversial.   If discovered today, _Rapator_ and _Ceratops_ would probably 
not be named. 

 

> > Because _Rapator_ is known only from an isolated metacarpal I, it is a 
> > nomen dubium.
>
> This alone is not correct.


I know.  The context in which this sentence appeared made it very clear that I 
think it's fine that we name genera or species based on a fragmentary specimen. 
 But such a specimen *must* have unambiguous autapomorphies or a unique 
combination of characters.  


Cheers

Tim