Felix Breden's suggestions of using either a distance measure based on band-
sharing or the presence or absence of the bands strikes as "cladistic"
characters strike me as sound. However (1) the distance measure, if nonlinear
with branch length, should be transformed to approximate linearity, as
otherwise there will be distortions in the tree, and (2) the presences and
absences are not necessarily for independent characters, and any technique
(such as bootstrapping) that relies on having independently evolving
characters would then be suspect.
[Let me take this occasion to repeat my usual refrain: "cladistic" and
"phenetic" are properly used to describe approaches to classification. Any
attempt to divide methods of inferring phylogenies into these two categories
is likely to be misleading.]
-----
Joe Felsenstein, Dept. of Genetics, Univ. of Washington, Seattle, WA 98195
Internet: joe at genetics.washington.edu (IP No. 128.95.12.41)
Bitnet/EARN: felsenst at uwavm
UUCP: ... uw-beaver!evolution.genetics!joe
