Re: Livezey and Zusi's big bird morph analysis [...]

[David Marjanovic <david.marjanovic@gmx.at>]

> > But damn....
> > this is THE morphological analysis of living birds.

For years to come, I think, Neornithes phylogenetics will mostly consist of 
adding taxa (especially fossils) to that matrix.

> - given the sheer amount of chromosomal rearrangements
> present in Accipitridae, one could actually ponder
> whether it is wise to include them at all -

That doesn't matter for the mitochondrial genome.

> As regards their phylogeny:
> (Isn't their use of Neoaves fairly unusual?)

Livezey & Zusi use it just like everyone else: it's the sister-group of 
Gallanseres.

> Group B is generally well-supported, though the timing
> of the galliform/anseriform split is debatable,

Do you mean Fig. 10B = Fig. 13 (which is paraphyletic)? Where does it say 
anything about timing? Do you mean Palaeognathae?

> as is ratite phylogeny...

Certainly.

> the ratite morphotype, however
> similar it may look, is certainly a (partly) neotenic
> homoplasy, with moa probably becoming flightless first
> and ostriches last.

I agree, but this doesn't tell us anything about ratite monophyly (which 
both morphology and molecules consistently support -- Livezey & Zusi cite 10 
references).

What I don't like about this paper is its fear of missing data, which led to 
the exclusion of large numbers of interesting fossils or their merger into 
OTUs like Enantiornithes. Even Dinornithiformes (plenty of complete 
skeletons, even some soft tissue available) was not included in the analysis 
but instead inserted into the tree a posteriori. Apart from the fact that 
calculation time increases drastically with the number of taxa, I can't see 
a reason for this practice. The more taxa, the better. -- At least 
*Aptornis* is in (sister-group of *Rhynochetos*, both together sister-group 
of *Eurypyga*), as are the dodo and the solitaire (sister-groups, together 
the sister-group of *Goura* and then *Didunculus*).

I wonder if that's why *Lithornis* was found to be the sister-group of 
Neornithes. I also wonder if adding some palaelodids and *Juncitarsus* would 
separate Gaviidae and Podicipedidae. Interestingly, Zusi & Storer (1969) are 
cited for myological evidence for a relationship between Podicipedidae, 
*Rhynochetos*, and *Eurypyga* -- all of them Metaves...

What might happen if someone adds the plotopterids...? Livezey and Zusi 
outright say (p. 26) that they "did not merit analysis herein" because their 
skulls are "woefully incomplete". They don't seem to have tried if their 
inclusion produces phylogenetic grass.

Among the living, the Acanthisittidae were not available for study (p. 88) 
and are not included.

Finally, I don't understand how using the ICZN, acting as if it extended to 
all ranks, acting as if the principle of exhaustive subsidiary taxa were 
real, and not using phylogenetic definitions, let alone erecting 
paraphyletic taxa (!!!), constitutes avoiding the "controversy" between the 
Linnaean codes and the PhyloCode (p. 87)... At least there is now a taxon 
called Raptores.

Incidentally, what is most impressive is Mayr's productivity. Cited papers 
range from 2001a to 2001g, from 2004a to 2004f, and from 2005a all the way 
through 2005i!