The units of evolution, paraphyly, etc.

[Harvey E. Ballard, Jr.]

Genetic differentiation, involving genes, are ultimately responsible for=
 phenotypic changes that result in the production of different species. But=
 the degree of genetic differentiation differs dramatically among groups,=
 and even among lineages within the same genus.  Regarding the "unit of=
 evolution", the vast majority of theoretical literature has reiterated for=
 decades that the POPULATION is the unit of evolution.  Indeed, it takes no=
 great imagination to envision that individual populations within a larger=
 collective that we may call a single species are capable of changing=
 somewhat independently of other populations.  Even the ability--not the=
 theoretical capability but the actual physical phenomenon--of gene flow to=
 occur to any significant extent among populations of a particular,=
 widespread species has been called into question by some (I believe rightly=
 so).  More than a few studies have documented isolation--reproductive or=
 other--limiting or precluding gene flow among different sets of populations=
 that one could justifiably call species, but that show very little or NO=
 observable genetic differentiation according to isozymes or chloroplast DNA=
 data.  Should these be subsumed into a single species just because our=
 methods are too coarse or our sampling of genetic variation too cursory to=
 detect genetic differentiation?  I argue, no.  In fact, I would argue that=
 we should be taking the documented or interpreted presence of limitations=
 to gene flow among populations within many groups as the result of some=
 important genetic differentiation, whether or not we are capable of=
 detecting differentiation with our methods; if we can, so much the better;=
 if we can't, it doesn't change the existence of isolating barriers or=
 limitations to gene flow that we can see, test or infer (regardless of what=
 the populations do in the laboratory).  I'll even be radical and argue that=
 we should interpret genetic differentiation from the PLANT'S point of view,=
 and ignore the source of genetic variation: in the case of taxa of hybrid=
 origin, as with those of an origin by divergence, if the populations are=
 predominantly isolated reproductively or ecologically, reproduce themselves=
 effectively in nature, and exhibit a pattern of interpopulational and=
 intrapopulational variation that suggests a different evolutionary=
 trajectory that is independent of the parental taxa, I couldn't care less=
 that the populations originated as de novo hybrids once upon a time (nor do=
 the plants--it seems only systematists get their shorts in a bunch over the=
 issue of whence the initial genetic variation arose).

How does this relate to the acceptance or rejection of paraphyly?  Although=
 comparatively few studies have yet been done examining multiple populations=
 of particular species and several species in a particular complex (I'm=
 thinking of plants here, for what that's worth), it's not infrequent in the=
 literature to witness populations of a particular widespread species, using=
 molecular phylogenetic data, that associate more closely with populations=
 of other related species of more restricted distribution or somewhat=
 different ecological preference than with populations of the same species. =
 Indeed, one would expect to see evidence of this association of narrowly=
 restricted species "embedded within" a series of populations of a=
 widespread species IF different marginal populations of the widespread=
 ancestral taxon gave rise to different geographically restricted related=
 species (once geographical isolates of the ancestor).  I would suggest that=
 ANY progenitor-derivative event in which a smaller population with reduced=
 genetic variability in comparison to the total ancestral population will=
 result effectively in a monophyletic series of populations in the=
 derivative and a paraphyletic series in the ancestral taxon (although some=
 folks seem to circumvent this paradigm by saying that one cannot apply the=
 terms "monophyletic" and "paraphyletic" to species).  The further caveat=
 that proponents of the genealogical phylogenetic species concept would=
 make, if I understand the theoretical argument, is that the paraphyletic=
 ancestral taxon would eventually become monophyletic, given enough time for=
 gene flow to "tie together" the remnant populations assignable to the=
 ancestor (or what it's now become).  Extremists who accept only=
 monophyletic taxa recommend not recognizing the paraphyletic taxon until it=
 becomes demonstrably monophyletic.  I would like to see more studies with=
 multiple populations of individual species in species complexes of closely=
 related taxa to find out whether or not there is in fact evidence for=
 substantial gene flow uniting populations of widespread species.  In a=
 study of one of the more primitive species complexes in the genus Viola, in=
 which ITS variation is sufficient to resolve even population-level=
 variation, we have found that the widespread V. canadensis has several=
 regional endemics embedded within it and that marginal populations of V.=
 canadensis in the same region of each endemic generally associate with the=
 endemics rather than "interior" populations of V. canadensis.  If gene flow=
 has not continued to the degree that it has "united" all populations of=
 remaining V. canadensis following what is apparently a series of speciation=
 events from geographical isolates in this relatively old (late Miocene?)=
 species complex, how might we expect it to have proceeded in more recent=
 groups?  Doesn't this argue against the pragmatic applicability of a=
 genealogical species concept or any other that refuses to accept=
 paraphyletic species?

My problem with accepting only paraphyletic species involves three points.=
 (1) A number of papers have recently suggested that an enormous number of=
 paraphyletic species probably exist, in keeping with evolutionary=
 hypotheses of certain speciation models. (2) We biologists worldwide, and=
 individually, should be exerting much MORE effort to catalog and=
 characterizing biodiversity from an alpha taxonomic standpoint before it=
 slips from our grasps in the next few decades, and to leave taxa UNNAMED=
 means to leave them uncharacterized--a suggestion that I find=
 unconscionable and cowardly in the face of the biodiversity crisis=
 assailing us. (3) Considering that organic evolution proceeds in a=
 "non-monophyletic" fashion a goodly share of the time, systems of=
 classification that refuse to recognize paraphyletic species reject a=
 substantial proportion of the products of evolution and, in essence, create=
 an "unnatural" classification system.

As some folks have suggested, here and elsewhere, the "compromise" that I=
 argue for is separation of cladistic evaluations of relationships and=
 patterns of differentiation from the process of characterization and=
 categorization, at least at the species level where pragmatism and the=
 products of population-level evolution will simply often be incongruent=
 with the concept of monophyly.  Interpreting the processes and products of=
 evolution with hard data, and even "thinking like an evolving plant=
 population", rather than behaving as a theoretician imposing hypothetical=
 constructs onto the patterns of nature, would also help.

Boy, do I expect to be stepped on for that one!

Harvey Ballard


++++++++++++++++++++++++++++++++++++++++++++++++++
Harvey E. Ballard, Jr.
Postdoctoral Researcher, USDA-ARS, Potato Systematics Labs
1575 Horticulture, University of Wisconsin-Madison
Madison WI 53706
PHONE: (608) 262-0159; FAX: (608) 262-4743
and
Honorary Fellow, University of Wisconsin Herbarium
Botany Department
132 Birge, 430 Lincoln Drive
Madison WI 53706
PHONE: (608) 262-2792; FAX: (608) 262-7509

AFTER 31 JULY:
Assistant Professor
Department of Environmental and Plant Biology
Porter Hall
Ohio University
Athens OH 45701-2979
PHONE: (614) 593-1128; FAX: (614) 593-1130
E-MAIL [probably]: not yet known