Place in taxonomy

[Robert Mesibov]

In on- and off-list responses to my posts there are two recurring points:

"You shouldn't weigh biogeographical evidence more heavily than character
evidence when you build a phylogeny." Of course not, and I never suggested
you should.

"You shouldn't trust biogeographical evidence in building a phylogeny,
because location isn't heritable, and a phylogeny (a history of descent) is
necessarily all about heritable characters."

The latter statement is flawed because it confuses two different things:

(a) a true phylogeny, which is a real, historical network of relationships
by descent, and

(b) a phylogenetic hypothesis, which is a guess about (a).

We would naturally like (b) to match (a) as closely as possible. A
currently well-respected method of coming up with (b) is to start with a
taxon-by-character matrix and use cladistic methodology to derive a network
of relationships between character-states. Character-states are heritable,
so they seem "safe" as evidence to use in hypothesising patterns of
descent. ("Safe" but far from ideal, for reasons even cladists accept.)
Within the framework of this method, spatial locations aren't "safe"
because they aren't heritable.

But this isn't the only way to come up with (b). You're allowed to use  any
and all relevant evidence, including biogeographical data. If your
favourite method doesn't allow you to use non-heritable, but relevant,
properties of taxa, maybe your method is a bit too restrictive? Maybe you
should be a bit more Kinman-eclectic?

This isn't the place to get into an argument about how you know that (b) is
a match for (a), or more accurately, why you believe that (b) fairly
represents (a). I shake my old head in wonderment, though, when someone
says "It's the best possible (a)-(b) match because it's the best possible
method and I used it properly." "Best currently available" isn't the same
as "best possible," and a method that ignores relevant evidence isn't
likely to be current for very long.

Dr Robert Mesibov
Honorary Research Associate
Queen Victoria Museum and Art Gallery
Home contact: PO Box 101, Penguin, Tasmania, Australia 7316
(03) 64371195; 61 3 64371195