In article <43ecm1$l89 at studium.student.umu.se> Agneta.Guillemot at historia.umu.se (Ludvig Mortberg) writes:
>Of course, and it has been done, but most of the time the molecular
>people just use similarity as a criterion for relatedness.
I'm not sure what you mean here. Most molecular phylogeneticists
use some kind of choice criterion. There are a few of them and they
are not all apart from choice criteria that are used in morphology.
>>I think that sequence information should only be used if nothing else
>works. If you can't identify any reliable morphological traits try
But what about the fact that any data set of any sort could be
misleading in some way?
That is, morphology can be misleading due to the frailties of the
experimentor's idea of character homology.
And molecular data can be so as well due to such things as alignment
parameters ro GC content or branch lengths.
I think as many data as can be had from as many data sets as possible
is the only reasonable way around this. That is, where they agree
and mutually corroborate each other, is bound to be due to history.
>sequencing a couple of genes with broad outgroups. What to sequence is
>hard to decide. Go for something linked to morphology or metabolism.
Again, things linked to something else are not going to stand as
independent estimators of that something else.
Mark
--
Mark E. Siddall "I don't mind a parasite...
mes at vims.edu I object to a cut-rate one"
Virginia Inst. Marine Sci. - Rick
Gloucester Point, VA, 23062
