[Taxacom] Null Hypothesis in Taxonomy

Richard Zander Richard.Zander at mobot.org
Mon May 7 15:02:33 CDT 2007

I wonder what is the null hypothesis in taxonomy?
Consider contriving a big data set of, say 100 2-state traits, and 50
taxa. Assign the traits randomly. You should get clusters of taxa just
by chance. Now eliminate all traits that do not match well with other
traits (taxonomists ignore character states that don't match well). The
taxa should now be "well supported."
These taxa would be the null for special creation (morphology). The
alternative hypothesis is . . . what?
We can analyze the random data (with annoying too-random data removed)
with parsimony and find a nice well-supported cladogram, demonstrating,
in a way, that neutral mutations can be generated through time and end
up with the same well-supported clusters.
This tree would be the null for Darwinian evolution (morphology). We
have consilience in that molecular data would generally support the
random data if nature provided random morphological traits (which it
does at times, e.g. drift). The alternative hypothesis is . . . I
suggest some combination of morphological neutral evolution and trait
changes that are clearly associated with selection. 
Therefore, we taxonomists should try to include information on niche and
functional evolution in our taxonomy, not just be content to generate
inferred genealogies, which is the null. 
Richard H. Zander 
Voice: 314-577-0276
Missouri Botanical Garden
PO Box 299
St. Louis, MO 63166-0299 USA
richard.zander at mobot.org
Web sites: http://www.mobot.org/plantscience/resbot/
and http://www.mobot.org/plantscience/bfna/bfnamenu.htm
For FedEx and UPS use:
Missouri Botanical Garden
4344 Shaw Blvd.
St. Louis, MO 63110

More information about the Taxacom mailing list