No Subject

[John Alroy <jack@xxxxxxxxxxxxxxxxxxxxxxxx>]

Based on some recent posts I'm not sure if my previous off-the-cuff mention of
Sepkoski and Kendrick was noticed. Here is the full reference:

Sepkoski, J. J., Jr., and D. C. Kendrick.  1993.  Numerical experiments with
model monophyletic and paraphyletic taxa. Paleobiology 19(2):168-184.

The purpose of their study was to establish by simulation whether arbitrarily
defined, paraphyletic higher taxa would or would not record species-level
diversity patterns with fidelity, and ditto for strictly defined higher-level
clades. Their findings showed that paraphyletic groups and clades _both_
recover the basic diversity pattern, but that paraphyletic groups are a _more_
robust indicator when sampling is poor. However, paraphyletic groups tend to
dampen mass extinction signals, which means that discussions of mass extinction
patterns are if anything conservative because the species-level extinction rates
may be far higher than one might guess.

Let's keep in mind in our discussion of this issue that what we want to know
about is species (or "lineage") level diversity, and we are only using families
(or orders, or even genera) as proxies. Whether the higher-level taxa have any
biological reality on their own is a question for the philosophers, not for
students of diversity patterns. If strictly defined cladistic higher taxa show
different diversity histories than species, that's interesting but hardly to
the point. The fact that arbitrary higher level groups are good statistical
indicators of species diversity is all we need to know in this context.

Also, a point of clarification about my comments on Mike Benton. My points
concerned his recently published paper in Science, not the Fossil Record 2
database, which apparently is administered (?) by Mary Benton.