paraphyly

[John Alroy <jack@homebrew.uchicago.edu>]

Stan Friesen presents a fine summary of the "common sense" argument  
against strictly applying the criterion of monophyly to the naming of  
fossil groups like the Ornithomimosauria. However, I think there is a  
good alternative to this argument (that naming all and only  
monophyletic groups leads to a proliferation of artificially highly  
ranked fossil groups like the "Class" Ornithomimosauria). It seems to  
me that instead of preserving all sorts of paraphyletic higher  
groupings like "Reptilia," we should simply get rid of the entire  
ranking system itself.

Here is my argument (if it isn't obvious from Stan's argument): if we  
REALLY went ahead and gave a formal name and Linnean rank to every  
clade we knew about, we would literally have to have scores and  
possibly even hundreds of separate ranks! Just look at the semantic  
mess that Malcolm McKenna got bogged down in when he tried to give a  
rank to every known higher-order grouping of mammals back in 1975  
(e.g., "Parvorder," "Mirorder," "Grandorder," "Superorder,"  
"Infraorder," etc., etc., etc.). Ranks are an artificial, misleading,  
and essentially meaningless carry-over from the pre-evolutionary days  
of Linnaeus, and as far as I can see they serve only to 1) give nerds  
like me something to argue about, and 2) confuse the heck out of  
undergraduates taking intro bio or intro evolutionary bio courses,  
which I think is a real shame.

Actually, I do agree that there is a place for paraphyletic groups in  
paleontological classifications. However, I think they should be  
avoided when this is reasonable, and perhaps restricted to the genus  
level, where I really find it hard to see how we can avoid them. For  
example, if every horse genus had to be monophyletic, we would have  
to double or triple the number of genera because we have fossils of  
practically every ancestor-descendant pair of horse species going all  
the way back to the start of the Eocene! Only the genus Equus itself  
would be preserved. The genus Dinohippus would have to be split into  
at least two genera; so would, say, Mesohippus or Hyracotherium  
(there is a big push to split Hyracotherium these days anyway). If  
Pliohippus turns out to fall on the "main line" after all (as is  
becoming more and more apparent), it would have to be split into at  
least two and maybe three or even four genera! What an awful penalty  
to pay just because you happen to leave some descendants...