[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Subject Index][Author Index]

Re: Cladistic definitions of Dinosauria, Saurischia, Sauropodomorpha

On Mon, 19 Sep 2005 13:58:04 -0700 "T. Michael Keesey" <keesey@gmail.com>
> On 9/19/05, Phil Bigelow <bigelowp@juno.com> wrote:

> > Congruency of an idea over time has its good points, no doubt 
> about that.
> >  But should that be one of the prime factors in defining a clade?
> > Shouldn't the communication of new knowledge also be a factor?

> No! Then we'd have to change our definitions every time new data was 
> discovered.

I don't see why that would have to be the case.  Provided that there was
enough empirical evidence used in defining the clade the first time, then
there would be no need to worry about any new discoveries messing things
up later.  Certainly you would agree that the dinosaurian characters in
birds are voluminous.

And to be clear:  I *loath* the idea of "redefining" a clade, and I hope
that Phylocode comes up with rules that roughly parallel the rules for
gen. sp. naming in the ICZN.  Once a clade has been defined in the
literature, it remains that way forever (unless it has broken some other
naming rule, such as priority, preoccupation, etc., etc.).

Example: If Padian and May's (1993) definition is ever formerly adopted,
and if birds are later found to be outside of Dinosauria, then I would
advocate just leaving Padian and May's clade name in place, and then
creating a new clade that excludes birds.  Call it Eudinosauria, or
Owendinosauria, or something similar.

> _Cetacea_ has been recently found by some studies to lie within
> _Artiodactyla_. If this becomes an established hypothesis, would you
> ever advocate using _Balaena_ as a specifier for _Artiodactyla_?

Ahh, interesting.  Well, I wouldn't dismiss it outright.
I'm not opposed to the idea of using "extreme" examples of members of a
clade to define the clade.
But I would only advocate such a thing provided that all of the following
criteria are rigorously met:

1) Artiodactyla is indeed monophyletic.
2) Artiodactyla hasn't already been cladistically defined (priority of
publication date).
3) The empirical evidence for cetaceans being members of Ariodactyla must
be *overwhelming* (such as is the case with birds being dinosaurs).

Since #3 isn't rigorously met, I would not advocate using a cetacean
specifier for the Artiodactyla.  And until Phylocode is formalized, #2 is
still up in the air.  But if the evidence was overwhelming, and if #1 and
#2 have been addressed to my satisfaction, then I might give the idea
some support.

But don't worry.  I promise not to write any papers on the phylogeny of
Artiodactyla until the Code is formalized!  ;-)

> Definitions are supposed to be stable (at least eventually)

I agree completely.

But mistakes can happen, and science is forced to live with them.  Such
naming problems crop up regularly in the ICZN world, yet life goes on. 
But I don't believe that birds will ever be proven to be non-dinosaurs,
so Padian and May's definition will never be shown to be unwise.

Just for the record, I suspect  that the final version of Phylocode will
require that a clade definition that uses a Linnaean name *must* use as
specifiers only traditional members of that Linnaean group (you appear to
believe this too).  So this discussion will probably be moot in a few(?)
more years, and Padian and May will be SOL (unless their definition gets
"grandfathered in" as being valid).