subspecies are real

Doug Yanega dyanega at UCR.EDU
Sat Apr 17 14:18:30 CDT 2004

Ron wrote:

>But if evolution is
>real, then the sisters are merely unknown (past and future) or undiscovered
>(present).  We call the past sister "mother" and the future sister
>"daughter" but there are only sisters all.  I contend that before any
>species reached that "rank /state" it was a sister (subspecies ) with a
>conger first - unless we embrace its creation as is.

This is a fallacy, based on the assumption that speciation only
occurs through vicariance. If one is dealing with sympatric
speciation, there is no point in the process during which one has
regional isolates, so there would be no "subspecies" at any point;
the segregates would be considered "forms" up until whatever moment
the gene flow drops below whatever arbitrary threshold is necessary
for something to be considered an "independent" lineage. There are
numerous such cases in nature (in fact, in the last month, I've seen
two seminars on exactly these sorts of systems, one dealing with
Lycaenid butterflies, the other with Timemas). You start with one
species on one host plant, and a segregate within the population
switches to another host plant, eventually splitting off
*ecologically* from the population on the original host, with no
vicariance involved. If the concept of a subspecies is "something on
its way to becoming a species but not yet sufficiently reproductively
isolated", then one could only accommodate this sort of sympatric
thing (which clearly fits the concept) if one broadened the
definition of subspecies by removing the geographic component, but
that would eliminate essentially the ONLY practical, objective
criterion used in the standard definition. In other words, if one can
only define a subspecies on the basis of gene flow, only geneticists
will be able to do so, and those of us using purely specimen-based
taxonomy will be hard-pressed to supply the proper type of evidence
(along the lines of what Denis Brothers pointed out). Back in the
good old days, it didn't much matter if the general species concept -
as a *concept* - included reproductive isolation, because almost no
one could measure it, and no one seriously objected that the way
taxonomists defined a species was not in strict accordance with our
species concept. That situation has changed.

If genetics enters into how one defines a subspecies, it's open to
debate whether we could or should utilize a *formal* system of
trinomials to accommodate them. As time goes on, and the body of
evidence accumulates as to how trivially easy it is to find genetic
differences between populations, I am increasingly concerned that -
should people decide that every genetically variant population needs
its own trinomial - we may find ourselves in trouble. Not necessarily
the proliferation of names (which would be burdensome but probably
ultimately tolerable), but the basic, unavoidable problem that most
historical names are based on specimens for which neither its *exact*
geographic provenance, nor its genome, are retrievable bits of
information. There are a lot of Linnaean, Fabrician, and other
ancient taxa for which there will never be any hope of confirming
which, of the dozens of "subspecies", is the nominate subspecies -
and likewise for most of the many synonyms, of which the majority
could, by this definition, be perfectly valid subspecies! Basically,
trying to adopt a genetic criterion for taxon definition IN THE
CONTEXT of the present taxonomic system is very much ramming the
proverbial round peg into a square hole.

This, without question, is why folks like Hebert are proclaiming
taxonomy to be a dead, obsolete science, to be discarded in favor of
purely gene-based classification systems such as he has proposed.
That the traditional system has been able to accommodate *any* sort
of a subspecies concept over the last century is a testimony to how
forgiving the taxonomic community is regarding its own internal
inconsistencies, especially after the advent of genetics (something
which was not part of the picture for most of the history of
taxonomy). But we're reaching the limit, I think, of how far we can
bend without breaking. Rigid adherence to tradition is going to
create increasing tension, of the sort that breeds proposals like
Hebert's - to scrap the entire nomenclatural tradition and start over
from scratch so those people who are focused on the genetic aspects
of biodiversity and evolution are not hobbled (as they see it, with
justification) by practices which are incompatible with theirs.

What I'm leading up to here is that I think some sort of compromise
is ultimately going to have to be worked out if taxonomy as we know
it is to survive, and of the two options that immediately come to
mind, I don't see either of them as leaving room for trinomials. In
the first option, we can formally draw the nomenclatural line at
species, and cut everything below that level loose from ICxN
restrictions; this drastically reduces the conflict between the
gene-based workers and the specimen-based workers simply by reducing
the overlap between them (without actually resolving the basic
conflict itself). The second option is to keep the system largely
intact, but make one major, heretical concession: to allow, under
certain (and fairly restricted) circumstances, the designation of
neotypes for taxa which already have valid type specimens,
specifically to permit the objective association of a genome (and
regional population) with a given name. Under such a system, however,
I believe nearly every taxon will be designated as a full species;
subspecies-level taxon names will nearly vanish simply because so few
entities meet the full set of criteria (as Denis pointed out). Either
way, I don't see a bright future for trinomials, which is a change
for me, since the group I'm most interested in (bees) has a long
history of trinomial usage, and I once contemplated creating a few
myself. Old habits die hard, but they *can* die.


Doug Yanega        Dept. of Entomology         Entomology Research Museum
Univ. of California - Riverside, Riverside, CA 92521
phone: (909) 787-4315 (standard disclaimer: opinions are mine, not UCR's)
   "There are some enterprises in which a careful disorderliness
         is the true method" - Herman Melville, Moby Dick, Chap. 82

More information about the Taxacom mailing list