[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Subject Index][Author Index]
Re: Cladistics question
At 08:10 AM 1/24/98 +1100, Dann Pigdon wrote:
>When I see a cladogram
>(or whatever else they are called) that consists of neatly branching
>lines I begin to suspect that a great generalization has taken place.
In fact, as Wagner has correctly pointed out in a different post, it
represents the most parsimonious distribution of derived character states
for a given data base. It is the phylogenetic equivalent of a line on a
bivariate plot. Through in the consistency and retention indicies, Bremer
Support and Bootstrap Values for the nodes, etc., and it is like the line
and the dots on a bivariate plot.
>Given the nature of the fossil record, and the scale of geological
>time, this is probably to be expected. But it does beg the question:
>given their incomplete and greatly generalized nature, are they really
>worth arguing so fiercely over?
You should watch the paleoanthropologists go at it. They argue over much
>Surely one person's "mostly-guess" is as good as another's.
No, and that was one of the main advances of Hennig's system over most
previous systems. Now there is a quantifiable set of attributes by which to
judge and select particular trees. From the vast universe of possible
trees, one small subset will explain the data to a better degree than the rest.
(Note that "small" can be much greater than 1: The tree on Figure 9 (p. 688)
of Chris Brochu's paper in the latest Journal of Vert Paleo is the consensus
of 11,340 equally most parsimonious trees, but it has a helluva lot of
structure to it. 11,340 is a lot, but it is a tiny, tiny fraction of the
godzillions of possible trees for a matrix with 62 ingroup taxa).
The days of "do whatever makes you feel good" systematics are vanishing,
although some still cling to them.
Thomas R. Holtz, Jr.
Vertebrate Paleontologist Webpage: http://www.geol.umd.edu
Dept. of Geology Email:email@example.com
University of Maryland Phone:301-405-4084
College Park, MD 20742 Fax: 301-314-9661