FWD>certainty/un

[Leonard Krishtalka]

> And Warren responds:
>
>    It is not uncertainty in phylogeny reconstruction methods that I object to.
>  It is the inability to test the methods empirically against real data.
> ... Extinction has a tremendous effect on how well we are
> able to reconstruct phylogenetic trees, and I'll bet it has occurred in almost
> all but the very most recent plant groups.  This source of difficulty can
> never be removed from the picture.

        I don't know about "testing the methods empirically" but one can
(and does) test the RESULTS of, say, PAUP generated trees against results
from trees generated using different data sets.  For example, PRs of
Recent mammals generated from sequence data are tested against those
generated from morphology of the skeleton, soft anatomy, etc., and, in
turn, these are tested against PRs that include fossil mammals. The
degree of consensus of PRs from different data sets implies their
degree of probability.  Geologists and paleontologists have used the same
principle to reconstruct the age of strata, an activity in which no
single method (or approach) is testable without reference to a second:
e.g., radiometric dates of rocks (e.g., K-Ar based) are tested against those
determined using different radioactive elements (e.g., U-Pb), against the
pattern of magnetic polarities, against stratigraphic correlations
determined from fossil faunal composition, etc.

        Without such independent confirmation, the creationists would have
a field day with notions of an earth (or cosmos) older than 10,000 years
(or with the principle of descent with modification).
Also, I wouldn't worry about extinctions affecting the outcome of PRs of
Recent plants.  The paleobotanical record grows daily, and is an
excellent source of independent tests against the Recent.

        Krishtalka