Squares & sticks

tivol at tethys.ph.albany.edu tivol at tethys.ph.albany.edu
Tue Jun 21 13:49:44 EST 1994


Dear James,
	In your most recent post you use the notation (A,B),D, etc.  Does this
mean that taxa A and B share trait D?  If A and B share D to the exclusion of
*any* other taxa, then by definition B and C do not share D; in fact, C does
not share D.  If I understand correctly, it would be much more informative to
consider characteristics shared by N taxa and to draw, for example, curves of
number of instances vs N for each character.  Taxa which are too close toge-
ther, or characters which are too essential, will not give a large RAS score,
but will give a lot of information vis-a-vis N-tuple scores.  An example of
an informative N-tuple would be if a character occurs in clusters in taxa,
such as a trait which appears in three species in one genus, four in another
and three in a third, with other species in each of the genera not having the
character.
	The dependence of characters would be a *big* clue.  That is if a spe-
cies develops the ability to use a new substrate through a mutation, and if
the new substrate is much more plentiful than the one previously used, the
species would presumably lose other gene products which were important for
metabolic steps involving the old substrate as it spread through the niche
created by its ability to use the new substrate.  Not all dependent charac-
ters will be obvious, nor will they always exhibit a connection like the ex-
ample, but the dependence itself may shed light on a hitherto unknown connec-
tion.

					Yours,

					Bill Tivol



More information about the Bioforum mailing list