Re: ALVAREZSAUR SUBCLADES (AGAIN)

[Dinogeorge <Dinogeorge@aol.com>]

In a message dated 98-04-06 01:05:27 EDT, tkeese1@gl.umbc.edu writes:

<< A family within a subfamily? I'm all for the abandonment of supergeneric
 ranks, but are we really ready for this? >>

I now have a copy of the article, and the organization of the genera therein
is as follows:

Alvarezsauridae
  _Alvarezsaurus_
    _Patagonykus_
      Mononykinae new taxon
        _Parvicursor_
        _Mononykus_
        _Shuvuuia_

Inclusion of _Parvicursor_ in Mononykinae makes Mononykinae a junior
subjective synonym of Parvicursorinae, which was created earlier by Karhu &
Rautian as Parvicursoridae. This is because the suffix -inae is presently
reserved for subfamilies, and it makes Mononykinae a subfamily even though it
is only termed a "new taxon" in the paper.

Including a subfamily within the family Alvarezsauridae implicitly compels
dividing the family into at least two (more likely three) subfamilies,
although most cladists are becoming opposed to having monogeneric groups just
for such bookkeeping purposes. The subfamily that contains _Alvarezsaurus_
must be called Alvarezsaurinae, because that name has priority over any of the
other family-level names in this group. The ICZN-sanctioned subdivisions that
would most closely follow the spirit of the paper would be:

Alvarezsauridae
  Alvarezsaurinae (already exists because Alvarezsauridae does)
    _Alvarezsaurus_
  Patagonykinae (new taxon)
    _Patagonykus_
  Parvicursorinae (already exists because Parvicursoridae does; =Mononykinae)
    _Parvicursor_
    _Mononykus_
    _Shuvuuia_

To keep the subfamilies monophyletic, the first two would have to be defined
as stem-groups (there being no nodes between the common ancestor and the
genus): Alvarezsaurinae--all alvarezsaurids closer to _Alvarezsaurus_ than to
_Patagonykus_; and Patagonykinae--all alvarezsaurids closer to _Patagonykus_
than to (say) _Parvicursor_. Then Parvicursorinae could be node-based, as
defined in the paper: all descendants of the common ancestor of _Parvicursor_,
_Mononykus_, and _Shuvuuia_. This classification has no place for the common
ancestor of _Alvarezsaurus_ and _Patagonykus_, or for any stem genera that
might lie between _Patagonykus_ and the parvicursorines. The latter might be
included in a revised stem-based (rather than node-based) Parvicursorinae: all
alvarezsaurids closer to _Parvicursor_ than to _Patagonykus_. No big deal,
since common ancestors are never identified as such anyway.

Anyway, that's life among cladists who insist on monophyletic taxa. Maybe I'll
mail an edited copy of this post to _Nature_.