> I'm not a big fan of majority rule consensus trees, each MPT being just
> as likely as any other, even if 99% of them agree in some way the other
> 1% doesn't. Doesn't mean the odd 1% of trees is less likely.
Yes. But when the strict consensus is phylogenetic grass, then that
doesn't mean that every possible tree is equally likely, just that all
MPTs are equally likely. The majority-rule consensus appears to give me
some more idea of what the MPTs look like.
I tried to make an Adams consensus tree, too. Said "Out of memory"
and refused. Understandable after a night of branch-swapping.
> Majority consensus isn't even a robusticity measure like
> bootstrap (which is annoying itself in the exclusion of data).
With enough replicates, all the data is included, no? -- I didn't try
bootstrapping, because the support for any node can't be high when the
consensus is grass.
> > +--allzero outgroup
> > `--+--*Caudipteryx* (all species lumped)
> > `--+--*Archaeopteryx* (both species lumped)
> > `--+--basal Troodontidae (*Sinovenator* + *-ornithoides* lumped)
> > `--+--*Microraptor* (both species lumped)
> > `--+--Scansoriopterygidae (both species lumped)
> > `--+--+--*Rahonavis*
> > | `--*Shenzhouraptor*
> Nice to see your outgroup expanding.
All part of the ingroup, except for the "allzero outgroup". (Judging from
the PDW-like position in which Archie came out, that may not have been a
bad idea.)
BTW, there appears to be a bug in the program. It always tells me
"ancestor 'standard' included in analysis" even though the settings are
that there's no ancestor. Now I looked what that ancestor looks like...
"?" everywhere. No wonder it clusters with the practically unknown
*Alexornis* instead of with the outgroup!
> I don't think the Patagopteryx position is as odd as it seems.
> It has oddly primitive coracoid morphology.
And it lacks all the enantiornithe features that even *Jibeinia* appears
to have.
> What keeps it outside Ornithothoraces in your trees?
Didn't check that yet (the lists would be difficult to interpret when the
root is in the wrong place).
> I can't tell from the horrible photocopy quality in Clarke's thesis, and
> Marsh's drawing is inaccurate.
At least. :-) Many thanks.
--
+++ GMX - Mail, Messaging & more http://www.gmx.net +++
Bitte lächeln! Fotogalerie online mit GMX ohne eigene Homepage!