[Taxacom] Position Announcement

[John Grehan]

> -----Original Message-----
> From: Richard Pyle [mailto:deepreef at bishopmuseum.org]

> Thanks, John -- fair enough.  Some minor points of disagreement (e.g.,
I
> still feel that both lines of evidence are better than one -- so long
as
> they are interpreted carefully), 

That's the big problem I think. How does one interpret 'carefully' the
two lines of evidence? In morphology I don't think systematists expend
effort in carrying out both a phenetic and a cladistic analysis and then
carefully interpret each against the other. If (and I know it's an if)
molecular systematics is really phenetics in cladistic clothing, then
there would seem to be no better foundation for bothering with molecular
phylogenies either.

The reality of any careful interpretation rests against empirical
examples, and here is where I run into the brick wall. Morphology
clearly points to the orangutan over any other sistergroup with humans.
This works for both living and fossil taxa. So how is one to 'carefully
interpret' the molecular contradiction? At present the only answer seems
to be a phenetic resort to the 'law of large numbers' - the idea that if
many molecular bases give a particular answer then it must be true. The
possibility that they just reflect the preponderance of primitive
retention is ruled out of the question. For example, see the assertion
from a reviewer below:

"Later in the manuscript I responded in another way to this issue as the
authors conclude "the molecular similarities between humans and African
apes are more likely to reflect an overwhelming preponderance of
primitive retentions than derived novelties": It seems to me that this
argument would have been easier to make in the days when analyses were
based on, say, a single mtDNA gene that one could suggest was far from
non-neutral in its evolutionary history. But, your challenge here is to
explain why selection for derived nucleotide base pairs would work
across thousands of nuclear DNA bases and many, many presumably often
unlinked nuclear genes as well as non-coding and intragenic regions in
the same direction and only in the orangutan lineage (e.g., most recent
studies like Steipen and Young 2006; Schrago et al. 2003). This
challenge will only grow larger as we enter an era of whole genome
genomics-based phylogenetic analyses."

John Grehan

but now that I've woken up a bit, I'm not
> quite cantankerous enough to argue about it! :-)
> Aloha,
> Rich

But it's a good argument that must be argued rather than ignored if the
place and scientific status of morphology is to be retained in the
future of systematic theory and practice.

John Grehan