Three areas of systematics

[Harvey E. Ballard, Jr.]

For myself, I maintain taxonomy, phylogenetic classification and
evolutionary inference as interrelated but modally distinct realms under a
broader umbrella of "systematics".  All three have, or should have,
somewhat different goals and frequently require different types of data.  I
disagree strongly with the idea that it is always possible or even
necessary to have a classification which can, at least at the species
level, accomodate evolutionary modes and evolutionary responses of
populations or species through time.  Nice thought, but an examination of
empirical studies should be enough to discourage most folks in thinking we
can always strictly and fanatically apply human hypothetical concepts like
monophyly to species-level evolutionary scenarios to a satisfactory degree;
or that the rest of the naturalist and botanist community doing
"on-the-ground" field work can effectively utilize either phylogenetic or
evolutionary interpretations.  As far as my own belief, all three realms of
work are legitimate, and all three afford "correct" viewpoints, but one
should expect to arrive at different interpretations diversity in the same
group from these different vantage points a goodly share of the time--and
we DO, in practice.  In my mind folks stuck in the corner of one of the
three vantage points have a difficult time understanding why others aren't
flocking their vantage point; no wonder frivolous controversies
continue--many folks can't relate to others vantage points in this
tripartite area of science!

>From a strictly floristic or pragmatic taxonomic perspective, a phylogeny
is irrelevant; one needs thoughtful phenotypic characterization of species
presented in a useable format--a monograph or regional floristic treatment
with keys, descriptions, range maps and illustrations, etc.  Having an
overall estimate of the difference between closely related species by
eyeball is as useful as any knowledge of higher-order relationships.
Knowing that modern phylogenetic data place half the Violales with
Hamamelids and the rest with the higher Rosids is not pertinent to a
practicing floristician or a monographer; Holmes mentioned to Watson that
whether the earth revolved 'round the sun or vice-versa made not a wits bit
of difference to himself or his work.  The newest phylogenetic
reconstructions of relationships at any level may be interesting, but a
field botanist armed with the information is no better off.  Nor, I would
argue yet again--didn't we discuss this a few months ago?--does it have
regular or direct impact on landscape-level conservation, or other arenas
in which floristic or monographic or other classical taxonomic pursuits
really DO have critical impact.  Evolutionary scenarios are interesting,
too, but they have little impact on researchers attempting to make
first-stab attempts to characterize diversity on the landscape and get that
information to the botanist/agriculturist/naturalist community in a useable
format.

At this point in time, much of the phylogenetic work thus far published, at
least at higher taxonomic levels, are a posteriori evaluations of previous
classical systematic "classifications"--reexamining the circumscription of
higher-level taxa and the relationships of organisms making up those
higher-level taxa.  It seems ludicrous that nobody is willing to admit that
a morphological species concept is used initially in NEARLY ALL of
published phylogenetic studies.

How sophisticated it sounds to espouse phylogenetic or other species
concepts while conveniently overlooking one's implicit reliance on
classical systematic or strictly floristic/monographic circumscriptions of
species as the basis for making decisions on taxon sampling!  Such arrogant
neglect to admit the TRUE basis of one's taxon sampling--if one is willing
to admit that phylogenetic analysis in such a case is indeed
reevaluative--seems unfortunately to be rampant at the moment.  Some folks
seem to have lost track of who worked before them, and also of the great
debt we still owe to classical systematic studies and monographic
treatments.  After all, do those of us doing phylogenetic analysis really
run willy-nilly with our eyes closed and simply grab handfuls of leaves and
flowers to use for phylogenetic analysis, without relying on careful
examination of the diversity of our ingroup and outgroup taxa and reference
to literature--all of which has a strong basis in the morphological (or,
God Forbid it! the phenetic) species concept?

Evolutionary inferences, even those based on phylogenetic analysis, need
not--and at the species level I would argue OUGHT not, demand monophyly of
populations.  Others alot brighter than I and with alot more experience
have proposed that such a concept doesn't even belong in the purview of
studies at the species/population level.  A number of different modes of
speciation have accepted as plausible in the scientific community.  Only a
few instances would result in an ancestral population giving rise by strict
divergence to two approximately equal derivative sets of populations, such
that each was monophyletic.  The rest of the speciation models either
result in reticulation or greatly unequal derivative populations arising
from a paraphyletic (or "depleted") ancestor.  In the latter cases, some
have argued that gene flow would have to take place over a long period to
eventually produce a monophyletic progenitor/ancestral population.
Considering the growing evidence to indicate that gene flow is lower than
previously thought among populations of widespread species, and the
accumulating evidence of extensive geographic partitioning within species
based on molecular [phylogenetic!] studies, one would have to question
whether coalescence proceeds in the ancestral/progenitor population to any
appreciable extent in more than a few cases.  Again, too much wind and heat
on my part, and perhaps no light at all, but the criterion of monophyly
usually demanded at least in higher-level phylogenetic studies has no
place, I think, where we are attempting to make evolutionary inferences, at
least at the species/population boundary or within complexes of closely
related species.  Not unless one is willing to abandon all other
empirically obtained evidence to the contrary in order to champion the
recognition of only monophyletic groups of populations--in which case,
you'd better develop a new model of speciation, because most don't
accomodate it well.  [There hasn't been a cladistic model of speciation
yet, has there?]

Evolutionary inference also must rely--or should rely where possible--on
sound classical taxonomic data.  Learning that an apparent widespread
species is actually composed of different ploidy levels makes a big
difference--and I ran across this with Ken Sytsma a couple of years
ago!--if one is interpreting modes of speciation.  Even thoughtful
circumstantial evidence on ecological differentiation among closely related
species, based on field observations and herbarium material, will
invariably shed new light on the origination and maintenance of diversity
in a given group.  These data are obtained through classical taxonomic work.

Yet, again, all three pursuits are equally valid and all yield
"correct"--albeit often different--interpretations of patterns of diversity
with the same group of organisms; perhaps the greatest degree of conflict
is actually at the species/population level within species complexes.

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Harvey E. Ballard, Jr., Assistant Professor, Plant Evolution and Systematics
Department of Environmental and Plant Biology
Porter Hall, Ohio University
Athens, OH 45701
(740) 593-4659 (office & lab phone)
(740) 593-1130 (fax)
email: ballardh at oak.cats.ohiou.edu