Cnidarian species concepts

Harvey E. Ballard, Jr. hballard at STUDENTS.WISC.EDU
Tue May 28 08:14:02 CDT 1996

Dr. Donald Potts wrote:
"One of my graduate students has several examples of cnidarian taxa that
have long been distinquished by discrete morphological and ecological
characters. We now have genetic (allozymes and ITS sequencing) data that
show so little genetic variation that they can't be interpreted: there are
few polymorphic loci (<10%), and no fixed differences if they are
polymorphic;  and genetic distances lie well within the range commonly
attributed to intra- or inter-population variation.

"Taken alone, the genetic data suggest all samples belong to a single
species; but this is belied by morphological and ecological patterns that
are consistent throughout their ranges of >1000 km. So another possibility
is that they are very recently differentiated taxa.
"We are interested in examples of how other people have approached analogous
situations. In particular, are there examples of compelling reasons to
justify retention of morphological taxonomic distinctions that are not
supported genetically? The literature seems full of two other situations:
        1. genetic separation of taxa not previously recognized morphologically;
        2. synonymizing taxa that cannot be distinguished genetically."


I am finishing a comprehensive phylogenetic study of infrageneric groups
and more intensive investigations (with complete sampling of taxa) of
certain complexes in the genus Viola, the true violets.  Admittedly an
angiosperm group but perhaps still relevant to your query for other
examples.  My study involves a reevaluation of macromorphological data,
hundreds of published chromosome numbers for virtually every group, and
nuclear ribosomal DNA  data in the form of ITS sequences for 70 species and

Almost all groups that are well differentiated morphologically and
cytogenetically--a few turned out to be based on dubious features at
best--also show considerable to extreme genetic divergence among groups and
a lesser degree of differentiation among species within groups in the ITS
sequence data.  In fact, the divergence is sufficient to allow me to
resolve population-level relationships within species of most groups.  One
group in particular, the "stemless blue violets" of subsection
Boreali-Americanae (infamous for their predominantly unrestricted and
lascivious hybridization) are strongly differentiated as a group based on
ITS sequence data but are very closely related genetically, differing by
only a handful of base substitutions among the most morphologically
divergenet species.  Their "terminal" position in one of the most highly
derived clades and the low sequence divergence speak themselves of a
geologically very recent origin.

However, the morphologically distinguishable populations occupy modally
different ecological conditions on the landscape.  They do indeed
hybridize, hybrids are subfertile from pollen stainability tests, and where
"intermediate" ecological conditions between those of the parents exist in
sufficient areal extent, hybrids will be produced that fill the
morphological and ecological gap between the two parents.  All species that
come into contact hybridize where conditions are appropriate to permit
establishment of hybrid progeny.  While localized introgression may in fact
occur (although no rigorous testing of this long-proposed hypothesis has
ever been published), hybrid populations are essentially always confined to
the zone of contact and generally do not "swamp out" the parental
populations except in very isolated and extreme circumstances.

At least as broadly defined, ecologically isolated taxa, the phenotypically
very divergent sets of populations constitute "species" in their own right,
according to my own viewpoint.  Each also varies in different directions in
particular morphological details and each has its own distinctive
geographic range, suggesting that the different taxa have indeed evolved
under different conditions and are on different evolutionary trajectories
(to get abstract).  They don't behave "nicely" as many systematists are
wont to hope but, again, from the plants' evolutionary standpoint, they
behave in accord with the spirit if not the letter of the biological
species concept--gene flow is limited by ecological isolation among
morphologically and ecologically delimited taxa but not (presumably) within
them.  Moreover, they accord (largely) with my own personal pragmatic bias
concerning the recognition of taxa as species: I'm more inclined to give
credence to a taxon as a distinct species if I can indeed detect it with my
eyeball or a handlens, too; I distrust recognition of cytotypes or other
biochemically distinguished taxa at the species level if I can't identify
it positively without relying on efforts short of full-throttle laboratory

I've interpreted the group to be a recently derived one, perhaps during one
of the earlier interglacial stages of the Pleistocene.  The group has not
yet acquired additional isolation mechanisms beyond the ecological.
Nevertheless, morphologically and ecologically distinct taxa can reliably
be distinguished and these hold discrete geographic ranges and show their
own unique morphological variation patterns.  I recognize them as species
regardless of their apparently young age and incomplete genetic
differentiation based on ITS sequence data.  I would expect that another
molecular data set such as RAPDs would distinguish them just fine, but
genetic differentiation alone shouldn't drive my decision to recognize them
as distinct species.

Hope this example helps.

Harvey Ballard

Harvey E. Ballard, Jr.
Department of Botany, University of Wisconsin-Madison
132 Birge, 430 Lincoln Drive
Madison, WI 53706-1381
phone: (608) 262-2792 (Rm. 161, Herbarium); fax: (608) 262-7509
e-mail: hballard at

More information about the Taxacom mailing list