[Taxacom] Evolutionary info from non-monophyly

Bernard.Pfeil at csiro.au Bernard.Pfeil at csiro.au
Sun Nov 23 17:28:23 CST 2008

With regard to paragraph 4, I would argue that "new taxa of higher rank" don't arise de novo within a paraphyletic group as a defensible objective truth (taking as given the inference of relationships).  It is our application of rank that makes the more inclusive group paraphyletic (if we give both groups the same rank).  This application of rank is subjective and often comes with the burden of history, say by the new discovery that family A is actually nested within family B.  Keep them as families and one has the problem of the hierarchical content of the ranked name being made meaningless; change A to a lower rank (or B to a higher one) and you may not.

Of course, if one isn't concerned about the relationships and in communicating that information using ranked names (at least a summary of this information), then using ranks to communicate something about character differences with other ranked names is the obvious thing to do with ranks.

The difficulty comes about because we (some of us at least) have this idea that a certain *amount* or a certain *kind* of difference in characters among groups signifies a certain rank.  I would suggest that characters, like beauty, are in the eye of the beholder and therefore have no special significance in the decision regarding rank.  They do, of course, have a very real role to play in the discovery of relationships, but this is not the main issue as raised by Richard Zander.  I see ranks as a convenience to communicate our current understanding of relationships, but recognise that they are entirely human constructs.

We, as a scientific community, need to finally decide which way we will go with ranked names - to use them to convey relationship information, or character information.  But I don't see that we can in all cases use them to convey both things unambiguously.  If the relationships among taxa have an existence independent of human thought and are potentially discoverable (and provide predictive possibilities), then this is a useful thing to represent and communicate.  If decisions regarding character differences (especially the eye-catching and pretty kind such as flowers) are instead thought to be largely subjective, then I don't see what we gain in trying to communicate this in a ranked nomenclature.  Sure, talk about characters, compare them, test their adaptive significance.  But keep this out of the ranked nomenclature and instead think about a different way of summarising that information.

my 2c

Bernard Pfeil

-----Original Message-----
From: Richard Zander [mailto:Richard.Zander at mobot.org]
Sent: Saturday, 22 November 2008 7:12 AM
To: taxacom at mailman.nhm.ku.edu
Subject: [Taxacom] Evolutionary info from non-monophyly

As an alternative view on monophyly as a classification method in
molecular phylogenetics, I submit that there are considerable
evolutionary data to be had from the apparent splitting of morphological
taxa on molecular trees.

Basically, phenomena like "massive homoplasy" and cryptic species (and
genera and families) can be explained by morphological stasis of
isolated populations of one taxon, with continued DNA changes in
non-coding sequences, and survival of at least two of such populations.
These imply a shared ancestor of the same name deep in the evolutionary
tree, a "virtual fossil." This requires the at least tentative
acceptance that at least some taxa are affected by punctuated
equilibrium, and that splits in molecular cladograms may represent
isolation events but not necessarily speciation events that affect both
products of the split.

The alternative explanation of parallelism or convergence is less
plausible because at least well-known traditional taxa based on
expressed traits have been diagnosed only after examination of many,
sometimes thousands of specimens over 250 years by multiple taxonomists
to try to establish that total parallelism or total convergence is fully
minimized in these taxa.

The application of phylogenetic monophyly as a classification tool when
dealing with molecular cladograms may organize relationships of genetic
lines of descent and isolation but, by ignoring any other evolutionary
features, such as new taxa of higher rank (signaling new evolutionary
developments) arising from paraphyletic taxa, limits greatly our aim at
developing a classification with the largest evolutionary component
possible. It also requires acceptance that only the pure form of
gradualist evolution and the biological species concept are operant in

A copy of this newly published paper is at


while a vigorous exchange of personal views by bryological
phylogeneticists and evolutionary taxonomists on paraphyly and other
controversial subjects has been archived in a blog:


You'll have to (if interested) winkle out the relevant posts as they
have been scattered among several topic threads, like classification,
monophyly, morphology, paraphyly and species concepts. What is nice is
that a debate like this is long overdue.

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

More information about the Taxacom mailing list