[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Subject Index][Author Index]
Clade nomenclature Re: Ceratosauria vs. Neotheropoda?
Combined answer to stuff written by several people, whom I can't cite
easily because I can't find any such setting in Thunderbird.
(Hmmmm. Are there native speakers who would use "stuff" and "whom" in
the same sentence?)
I think you might be overstating the importance of "original meaning"
in framing definitions and converting time-honored groups to clades.
There's no rule that says you *must* adhere to the original meaning
of a name when converting it into a clade name.
Indeed not. Instead, there's a rule (Art. 10.1) that requires
"minimiz[ing] disruption of current and/or historical usage":
The original meaning is not always the same as the one that has been
most common. As an extremely simple example, Lissamphibia Haeckel 1866
explicitly excluded the caecilians (part of the paraphyletic
Phractamphibia Haeckel 1866) in the original publication, but the next
paper to use that name (a few years later by someone else) included
them, and so have all publications ever since. Art. 10.1 clearly argues
against restricting Lissamphibia to frogs + salamanders (a clade
nowadays called Batrachia... a name with a very complex history behind
it) and for using it for the crown-group of extant amphibians, if these
form a clade to the exclusion of us amniotes.
Personally, I would prefer that Ornithosuchia be defined such that it
includes _Ornithosuchus_ - for the sole reason that the name was
erected specifically to include _Ornithosuchus_.
Mike Keesey has already explained Art. 11.7:
Similarly, Pseudosuchia was erected specifically to *exclude*
crocodiles (hence the name). Yet some definitions of Pseudosuchia
allow the crocodilian-containing clade Suchia to be a subset of
Pseudosuchia, which strikes me as ridiculous.
There are several more such cases, like Paracrocodylomorpha which is
meant to include Crocodylomorpha, and the fortunately forgotten
Eotetrapoda which was meant to include all limbed vertebrates (Tetrapoda
being limited to the crown-group in that scheme). Rec. 11G (more
precisely Note 11G.1) is against that:
http://www.ohio.edu/phylocode/art11.html (bottom of the page)
Auctor, actually. Connected to "augment" and "Augustus". The overzealous
Renaissance etymologists who meddled with the mediaeval spelling _autor_
(which is preserved in many languages today) were right that it wasn't
classical, but, well, wrong about everything else, even the language
it's from (Latin, not Greek as the _th_ requires).
However, sometimes compromises have to be made: Dinosauria now
includes birds (Aves), even though this would make Richard Owen spin
in his grave. But the nesting of birds inside Dinosauria is a
phylogenetic reality, and so overrides historical usage. Ditto for
nesting tetrapods inside Osteichthyes, or putting mammals inside
However, the PhyloCode nonetheless recommends against making it so by
definition, and even uses Dinosauria as Example 1 (of Rec. 11F) on
selecting internal specifiers from among the originally included taxa.
Science, not nomenclature, should be how we know that birds are dinosaurs.
(Also, if it hasn't become too boring, Owen is continually rotating
anyway because nobody listens to his peculiar brand of creationism
anymore. But I digress.)
I do dislike the name Osteichthyes, but the alternatives proposed so far
-- Euteleostomi, Neoteleostomi, Osteognathostomata -- all have drawbacks
of their own...
In situations like this, the only options are to expand the
understanding of the taxon (as with Dinosauria, now
all-but-universally understood to include birds), or to draw back and
not assign a clade definition of the name at all (which, happily,
sees to be the way we're headed with the paraphyletic group
I fear it's not a done deal, though. I'm not quite sure what for
instance Robert R. Reisz, whose middle R is said to stand for "Reptile",
is contributing to the Companion Volume...
Now, in the case we had some original meaning of Ceratosauria
including Ceratosaurus and ornithomimids, but excluding Allosaurus
(and if we want to follow original meaning instead of common usage)
we may do what Mike said and make a [node-based] definition with
internal specifiers Ceratosaurus and Ornithomimus and external
specifier Allosaurus, and then make the name get destructed at
But we don't want that. We want to keep the name Ceratosauria for the
clade for which it has been constantly used in the last few decades. For
that, a branch-based definition like "everything closer to *Ceratosaurus
nasicornis* than to *Allosaurus fragilis*" would probably be best.
Note that you *can*, however, define "Archosauria" without
Archosaurus rossicus, because Archosauria is not named after
Archosaurus. (Archosauria was named much earlier.)
Apparently *Archosaurus* was named after Archosauria in both senses of
"after" -- it is the oldest known archosaur and was probably thought to
be a direct ancestor of all others, thus, in a sense, "the" archosaur
(such as the only archosaur of its Permian time).
> Similarly, Pseudosuchia was erected specifically to *exclude*
> crocodiles (hence the name). Yet some definitions of
> Pseudosuchia allow the crocodilian-containing clade Suchia to be
> a subset of Pseudosuchia, which strikes me as ridiculous.
Me too. Under the PhyloCode this will likely be "Pan-Crocodylia" (or
at least, informally, "pan-Crocodylia").
What about Crurotarsi?