[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Subject Index][Author Index]

Re: Mesozoic roots of parrots and passerine birds



> That is true, but FGB-int7 did not support the
> falcon-parrot-passerine
> link in the Hackett et al. analysis -- the second intron of
> the
> eukaryotic elongation factor and somatotropin introns did.
> In fact,
> FGB-int7 links passeriforms to picocoracians (which is
> close to their
> traditional position in morphological analyses) and falcons
> to
> mousebirds.

Interesting. The former looks like several loci adding up (they tend to be 
"somewhere in the vicinity" according to most loci), while the latter is more 
odd. Mousebirds are notoriously "jumpy" (and IONO how much to trust their 
purported relationship to *parrots*).

So it looks like this is not an aggregate signal, but more likely one locus' 
particularly pronounced signal. Well I hope I'll find this out some time in 
'12/'13. I'll let you know.

I might compare signal distribution supporting particular "clades" ("loud and 
clear" singleton vs weaker but aggregate signal) in various loci while I'm at 
it. Has this ever been done, comparing multiple loci and seeing  where they 
propose a phylogeny that diverges from e.g. majority-rule consensus? And then 
e.g. comparing against the pertinent nodes' support and temporal distri.

I think the solution to these problems - IF there is a molecular solution 
(likely but not certain) - is if anything in analysing the signals, not adding 
more data. We should have enough data to yield an obvious solution, but it's 
far from being as obvious as in cases where this approach is proven to work 
("Sylviidae" s.l. for example, or Galloanseres). Something might have happened 
that violate the analyses' underlying basic assumptions, so we have to get 
"under the hood" of the data. The take home message of "Metaves" is apparently 
that there are strong "signals" that are false, and that these cannot be 
foltered out by the usual analysis protocol. 

The problem is exacerbated by the dating issue. "Eufalconimorphae" are not 
entirely unreasonable, but Cretaceous? What about Halcyornithidae 
(http://www.tandfonline.com/doi/abs/10.1080/14772019.2010.505252 
http://www.tandfonline.com/doi/abs/10.1080/14772019.2010.513703)? Close enough 
to a "missing link" somewhere near the base of Eufalconimorphae, but alive and 
kicking after 50 Ma including a mega-extinction and near-complete ecological 
turnover (and carnivorous to boot?) and still very much *looking* like the 
misssing link you'd expect? It works too well in some aspects, and it 
completely sucks in others. Those parts of the tree we *have* resolved for good 
fit together more or less awkwardly, but overall in a satisfying way.

Ah well, I'll try and find out more. I'll only be a matter of years... O_o

Regards,

Eike