[Taxacom] Apples and oranges (was: burn out)

Kenneth Kinman kennethkinman at webtv.net
Tue Apr 14 11:56:23 CDT 2009

Curtis and Richard, 
     Unfortunately the base of the Aves clade (Archaeopteryx +
recent birds) is in a very unstable area of phylogenetic trees. So let's
take a more stable phylogenetic tree (Rosopsida clade) as an example. A
great deal of evolution happened between it's base (near the origin of
tricolpate pollen) and the base of the "Core Eudicots".   And it
happened relatively quickly (within 10 million years?), so we are very,
very fortunate that several side branches in that transition have
survived (the Orders within my Subclass Ranunculidae). 
      The floral morphology then changed extremely quickly
between Gunnerales and "Core Dicots" (the base of my Subclass
Dilleniidae), probably due to gene duplications. However, there were
still many unknown species and genera (even families?) along the stem
leading to crown-group Gunnerales and the stem leading to crown-group
"Core Dicots" (as well as the stem between them leading back towards
earlier clades of Ranunculidae). Therefore it is futile to argue what
species or genus is the ancestor of "Core Dicots". It's actually a
paraphyletic bunch of species and genera (and families?) that have left
no trace (at least none that could be pegged to those stems, because
even a fortunate fossil find doesn't contain pertinent information like
      Given these gaps, we can safely ignore the paraphyly
and just call Gunnerales and "Core Dicots" sister groups. It's an
extremely useful "illusion", just as it is extremely useful to classify
things as solids or liquids (and just when cooling magma stops being a
liquid and becomes "solid" would be a futile exercise).   You can
cladistically define away the paraphyly, but it is still just a very
useful illusion (however, doing this to 100% or paraphyletic groups
causes more harm than good).  
      Therefore, in groups like angiosperms where we have a
relatively poor fossil record, we can get by with relatively few
paraphyly cuts in the classification (especially at Class and Order
level). However, at species level you have much more information, and
you have paraphyly staring you in the face all over the place. That's
where traditionalists and strict cladists will continue to butt heads,
even after some kind of consensus classification at higher levels is
finally reached. So comparing paraphyly at species level with paraphyly
at Class level is in a way like comparing apples and oranges (because
Classes split a long, long time ago, while many species are still in the
process of splitting).  The elimination of paraphyly at species level
thus eliminates more evolutionary information than the elimination of
paraphyly at higher taxonomic levels.  Brown bears and polar bears are
just one obvious example of this.  At Class level, there are bigger gaps
and fewer taxa, but the failure to make those few needed paraphyletic
cuts affects a lot more people.
         ------Ken Kinman 
Curtis Clark wrote: 
On 2009-04-13 06:50, Richard Zander wrote: 
The ancestor of Aves is the closest paraphyletic taxon. 
What's your metric for "closest"? For all your criticisms of the
weaknesses of parsimony analysis, this seems rather vague. Do you mean
number of generations, phenetic similarity, or what? I assume Aves had
to descend from it. Do you mean at the same rank (whatever that means
analytically), or what? 
I can think of analytical approaches to grade taxa (I have mentioned
them, but no one seemed interested), and I don't find grade taxa
valuable. It seems to me that their proponents should either be able to
do as well or else explain why analysis is unnecessary and what
substitutes for it. 

More information about the Taxacom mailing list