genetic and morphological systematics

[John Grehan]

> From: Taxacom Discussion List [mailto:TAXACOM at LISTSERV.NHM.KU.EDU] On
> Behalf Of Richard.Zander at MOBOT.ORG
> Well, those quotations, John, are not particularly impressive, since
they
> are merely cautionary, which is of course always acceptable in
science,
> but
> without giving particular reasons for the caution then the argument is
> vacant and not instructive.

Impressive or not, the point here is that a geneticist practicing
molecular systematics was willing to make these statements (which
negates suggestions to the contrary by some who do not like the
orangutan related evidence)

If one is interested in the reasons they can read the article.

> Molecular studies are supposed to be better than morphological studies
in
> that the data is sufficiently abundant that it can be used in a
> statistical
> context. 

Yes - which I find to be a largely phenetic argument of overall
similarity being more significant that uniquely shared similarity.

Unfortunately the requirement that the data be (at least mostly)
> independent and randomly distributed is doubtful. Certainly gene data
> (used
> to get deep branch information because the gene sites mutate slower)
is
> subject to selection, even synonymous codons are subject (sometimes)
to
> codon bias and this is often not tested for. Non-coding sites commonly
> give
> the same results as coding sites, so what does this suggest about non-
> coding
> sites?

Marks draws attentions to problems of this nature, although not the
problem of how base comparisons are different to morphology in terms of
cladistic assignment of polarity.

John Grehan