[Taxacom] Usefulness vs convenicence (Protista)

Richard Zander Richard.Zander at mobot.org
Wed Dec 22 14:13:12 CST 2010

Phylogenetics is strictly synchronic (relationships at present time). It clusters presentday exemplars by similarity (nonultrametric, to get different and unique support for different nestings) or through coalescence methods we start with presentday exemplars and end with nestings of just those exemplars. Unless a phylogeneticist rejects his/her Structuralist roots, no theory of diachronic (relationships, largely theoretical, through time) progenitor/descendant relationships are entertained, and no consideration of data not in the data set matters. 
Examples of difference between phylogenetics and evolutionary systematics:
Phylog: consider three terminal taxa ((ab)c)... The synchronic relationships are thus given (with lots of assumptions and limitations). Phylogenetic classification is based on this nesting structure. It is generally the same as that of evolutionary systematics, BUT If a or b is of a taxonomic rank higher than a and c, phylogeneticists lower that rank. Why? It was once thought to be willful cussedness, but now we see that the synchronic structuralist pattern would be compromised by any consideration of macroevolution. The pattern is artificially preserved by strict phylogenetic monophyly.
Evo. Syst: consider the same three terminal taxa in cladogram ((ab)c)... This is acceptable as at least tentative evolutionary relationships given that all similarity based clustering is acceptable as information about evolution since Darwin's time. What can we decide about evolution? Maybe a shared ancestor generated a and b? More parsimoniously maybe a or b was the ancestor of b or a, or both b and c, or a and c? We go then to nondataset information, particularly about autapomorphies, and find, say, that a is a widespread taxon with known fossils occuring in geologically stable enviroments, has half the chromosomes of b or c, has lots of plesiomorphic traits given morphological analysis, AND b and c occur in recent, unusual environments, have lots of unique and highly adaptive traits, etc, etc. With this we can say a is theoretically the progenitor of b and c, and look for more data, and check out the rest of the cladogram, and so on in pursuit of an evolutionary classfication based on both cladistic (synchronic, sister-group) and caulistic (diachronic, ancestor-descendant) relationships, both pattern and theory-based, using both phylogenetic, structuralist deduction and evolutionary, empiricist induction. Evolutionary systematists do not nomenclaturally hide macroevolution, but consider it important in biodiversity and conservation work.
Another example:
Phylog: Consider 4 terminal exemplars (((ab)c)b) ... A phylogeneticist would see this as phylogenetic polyphyly and probably rename the second b as a different taxon. 
Evol. Syst.: An evolutionary systematist would generate a theory that the two b's are surviving elements of one ancestral taxon at whatever taxonomic level b had. If additional information about the two b's showed disparate evolutionary trajectories (remember trajectories?) from data not in the dataset (chromosome number, chemistry, geography, habitat, development, parasites, whatever), and such additional information could be used in a key, then one of the b's would probably rate a new name as an instance of evolutionary polyphyly. In this case, after and because of thoughtful reflection, evolutionary systematists would follow the phylogenetic classification. BUT if the two b's were essentially the same in respect to other data, then a theory would be supported that a and c are generated from a shared ancestor of the same taxonomic diagnosis as the two b's. The original classification would then be supported.
Another comparison:
Phylog.: Extinct taxa never matter. Except when fossils are added to a morphological cladogram, and then these are usually considered entirely synchronic.
Evol. Syst.: One of the ways to approach investigating which taxon gave rise to which (meaning which ancestors of some taxon gave rise to presentday exemplars of another taxon) is to consider an extinct lineage inserted in the cladogram below the clade being considered that makes the clade paraphyletic/apophyletic. What ancestral taxon would the node be for that extinct lineage? This forces one to consider data outside the dataset to create alternative theories to weigh against each other. Example: ((polarbear, brownbear1) brownbear2)... is a terminal lineage. If another (extinct) polar bear lineage were below these three on a cladogram, it would be incredible to consider the polar bear as progenitor to a brown bear lineage given the autapomorphies of the polar bear. Such an extinct lineage is then considered theoretically very improbable (Dollo Rule). Inserting an extinct brown bear lineage does not shock progenitor-descendant theory. (This is aside from epigenetic or hox gene anomalies, which should be shown relevant before being considered.)
Richard H. Zander
Missouri Botanical Garden
PO Box 299
St. Louis, MO 63166 U.S.A.
richard.zander at mobot.org

From: taxacom-bounces at mailman.nhm.ku.edu on behalf of Kenneth Kinman
Sent: Tue 12/21/2010 8:33 PM
To: taxacom at mailman.nhm.ku.edu
Subject: [Taxacom] Usefulness vs convenicence (Protista)

Hi Richard,
      Thanks for the answer to my first question, which clarified what
you meant.  I would also be interested in clarification on the other
point I raised (about whether or not paraphyletic groups are strictly
synchronic). Especially as it might relate to another posting in which
you said that all paraphyletic taxa are unnatural (which I found rather
                          Ken Kinman             

Richard Zander wrote:
     Okay, Ken: By mish-mash I meant that, because nobody involved in
the present discussion is a pure ideologue, we mix phylogeny-speak with
evolution-speak. In one sentence one of us may say "((ab)c)" and in the
next "b is a peripatric puppy". It is totally true that phylogenetic
clustering is valuable information, but the apple vs orange occurs when
the interpretation of the cluster nesting ENDS in classification versus
when the interpretation of a theory from the cluster (when one is
available, which it often is) ends in classification. We use the highly
supported cluster pattern when it is important to emphasize the high
support, rocket science, and sneak in theory when no phylogeneticist
higher on the social scale is looking. 
* * * * * * * * * * * *
Richard H. Zander
Missouri Botanical Garden, PO Box 299, St. Louis, MO 63166-0299 USA Web
sites: http://www.mobot.org/plantscience/resbot/ and
Modern Evolutionary Systematics Web site:

More information about the Taxacom mailing list