Fwd: Re: Fwd: Re: Different codes - in support of the phylode

[Philip Cantino]

>
>At 01:34 PM 10/19/00 -0400, you wrote:
>>The eudicot clade is one of the most strongly supported major clades
>>of angiosperms.  Why shouldn't it have a formal scientific name
>>rather than being designated with a symbol (Ken's preference) or an
>>informal English name (Tom Lammers' preference)?
>
>I'll grant you, this is one clade that does not seem in any danger
>of collapse in subsequent analyses.  I'd have no problem with naming
>this one via the ICBN.  Especially if Campanula L. were the type
>genus!  : - o
>
>>I suspect that the only reason why the eudicot clade has never been
>>formally named is that to do so under the current system would
>>require giving it a rank, and it isn't clear what rank it should
>>have.
>
>Good point.  We like "monocot" and "dicot" to be classes, and
>Cronquist, Takhtajan, Thorne, and Dahlgren have used subclass and
>superorder to recognize diversity within these two.  It's hard to
>come up with a rank between class and subclass ("supersubclass"????)

The problem is even worse if angiosperms are treated as a class, as
they often are.



>Part of the problem is what to do with the grade at the base.
>Essentially, what we have in the angiosperms are two major groups,
>monocots and eudicots, plus a lump of "miscellaneous" at the base.
>Cladistic thinking won't let us give names of the same rank to
>non-sister taxa, so really, what rank do we assign to *any* of them?
>The monocots are not sister to eudicots, so it really isn't right to
>call either of them each a subclass, from a cladistic perspective.

The problem is not created by a requirement that taxa at the same
rank be sister taxa (this Hennigian ideal is not always followed,
even by people who would call themselves cladists).  It is simply a
function of there not being enough standard ranks to accommodate all
the well supported clades that could justifiably be named in a large
and diverse group like the angiosperms.



>And this, I guess, is why cladists favor a rankless nomenclature system.
>IF you accept that cladistics is the only way to classify, then I
>guess you must accept rankless nomenclature as the only way to go.
>

The issue that the "eudicots" example raises is not whether
cladistics is the only way to classify but whether the current system
allows one to give formal scientific names to all of the well
supported clades that warrant naming.  If the ranks of order,
superorder, and subclass are already occupied by taxa that are either
more or less inclusive than the eudicot clade, we must invent another
category (e.g., supersubclass) if we want to name this clade.  But it
gets worse:  If the usual convention is followed, once a
supersubclass is named, every subclass of angiosperms must belong to
some supersubclass.  (This is what Kevin de Queiroz calls the
Principle of Exhaustive Subsidiary Taxa, which happens to have the
memorable acronym, PEST.)  The PEST convention would require that we either
name at least one other (probably nonmonophyletic) supersubclass to
accommodate all of the subclasses that don't belong to the eudicot
supersubclass or that we assign each non-eudicot subclass to its own
supersubclass.  Whew!  It's no wonder no one wants to name the
eudicot clade.

With phylogenetic nomenclature, you would simply make up a name
(since this clade has never been given a formal name under the ICBN)
and give it a definition that would explicitly link the name to that
clade.  You wouldn't have to worry about classificatory issues when
naming it, as you do with rank-based nomenclature.

Phil

Philip D. Cantino
Professor and Chair
Department of Environmental and Plant Biology
Ohio University
Athens, OH 45701-2979
U.S.A.

Phone: (740) 593-1128; 593-1126
Fax: (740) 593-1130
e-mail: cantino at ohio.edu