: Real species and ideology

[Richard Pyle]

Ron Gatrelle wrote:

> IF there are
> subs  (plateaus) within one rank (plateau) - there are subs at all ranks -
> evolutional plateaus (big and small).

And, of course, there are potential subs upon subs upon subs upon subs, ad
nauseum, well beyond the scope of ranks and intra-ranks defined in Linnaean
nomenclature.  Pretty soon one realizes that the whole notion of taxonomic
ranks is what condemns the taxa described within the ranked framework to be
necessarily "artificial".  It's somewhat amazing (and ironic) that some of
the staunchest opponents to PhyloCode end up making the case for it better
than anyone else!
:-)

> Aus wus gus is A taxon.  Aus wus wus is A taxon.
> Aus (Bus) wus gus is A taxon.

No matter what side of this argument you're on, "Aus wus gus" and "Aus (Bus)
wus gus" are nomenclaturally identical.  Ignoring the whole "disparity
between names and concepts" issue (yet another massive thread all on its
own), these are the "same" taxon.

> Don't bother trying to argue subspecies are
> not real and then turn around and recognize even just one - because the
> exception proves the rank.  Same goes for species.   If any of us
> acknowledge _anything_ as a species, then they are real - by what ever
> definition it was accepted.

I think the point being made by the non-subspecies camp is that if you have
enough evidence to recognize any consistent distinction between two
populations, then it meets the threshold of what should be thought of as
"good species".  In other words, their contention is about the convention of
maintaining "species" as the smallest identifiable evolutionary unit.

This perspective is what I imagine lies behind Barry Roth's comment:

> I agree with proponents of phylogenetic species concepts (e.g., Mishler &
> Theriot, 2000; Wheeler & Platnick, 2000) that the least inclusive taxon in
> formal taxonomy should be the species, and that where they exist,
> well-defined, diagnosable "subspecies" should simply be called species.

...which is fine by me -- he's choosing to define the "reality" of a species
in a particular way.  It's not the same particular way that I think is most
useful for nomenclaturally-based communication among biologists; but at
least he's making it clear what he means when he says "species".

On the other hand, it was this comment of his that I found a bit more
objectionable:

> It was astonishing to me to read that in year 2000 Mayr still
> recommended subspecies as a waffle "in ambiguous situations."

This, to me, is *exactly* what the subspecies rank should be used for.

Following Barry's species threshold criterion:

>...a species comprises all the individual organisms predicted on the basis
> of available evidence to prove to compose a least inclusive monophyletic
> unit recognized in a formal phylogenetic analysis.

If I were to conduct a formal phylogenetic analysis of all the people I
identify as my living kin, and all of the people that my wife identifies as
her living kin, and all the people that Barry identifies as his living kin;
then I would very likely end up with three unambiguous monophyletic clades.

If we included my two children, then we'd find a morphocline (or would it be
introgressive hybridization?) between the lineage of my wife's kin and the
lineage of my kin...but let's not confound the issue.

Except for my kids, would my formal phylogenetic analysis not recognize
three least-inclusive phylogenetic units? Wouldn't, then, I be compelled to
recognize three distinct species?

For purposes of nomenclatural communication, it seems to me a more useful
threshold for the rank of "species" is to draw lines between those
populations whose complete set of descendants (until the end of time) are
highly unlikely to experience more than trivial gene exchange ("trivial" in
this case meaning lateral gene flow or rare hybridization). In other words,
Humans, Chimps, and Orangs can be confidently considered (and labeled) as
different "species".

In cases where current or future significant gene flow is plausible (e.g.,
the various populations of humans around the world, each with their own set
of apomorphic characters that could establish least-inclusive monophyletic
units), then it makes communicative sense to treat the entire scope of
organisms as a single "species".

Of course, we will find numerous cases of "tough calls"; that is, separate
populations have clear and consistent differences, but for various reasons
show evidence of plausibility of current or future gene flow (morphoclines,
hybrid swarms, etc.).  These are the cases where I think the rank of
"subspecies" makes sense from the standpoint of nomenclatural communication.
In other words, I would recommend subspecies as a waffle in ambiguous
situations.

Having said that, I think such decisions of rank assignment should be
tempered by consideration for nomenclatural stability -- at least when
Linnaean names are involved.  For example, if someone discovers two distinct
sub-populations within a broader population that has been traditionally
referred to as a single species, and it's not clear to what extent gene
exchange is currently happening or will plausibly resume between the two
sub-populations in the future, then establishing two separate subspecific
epithets does not disrupt stability of the species-level nomenclature.
Conversely, if two very consistently distinct parapatric populations have
traditionally been referred to by separate species epithets, but a discovery
of hybrid swarms in regions of sympatry is made later, I would be a bit more
reluctant to sink one of the long-standing names into a subspecific status
within the other (note: "bit more reluctant" is NOT the same as "wouldn't").

Purely objective? Nope.  Scientific?  Not really (but sort of, in the sense
that science is really about trying to understand the natural world, and to
do so requires a formal means of communication).  Practical?  I think so.

So, in summary:  When thinking about "species", "subspecies" and such from a
nomenclatural perspective, decisions should be made in the context of what
best serves the communicative needs of biologists.  When thinking about
"least inclusive monophyletic units", "evolutionary lineages" and such,
decisions about boundaries between units should be made in the context of
phylogenetic analyses. If both camps do their jobs well, the differences
should be minimal. But when differences do exist, each camp should respect
the decisions made by the other.

The problem, of course, is that both camps are fighting over the same
Linnaean nomenclatural system, which brings me back to .... oh, hell -- it's
after 10pm, and I've got other stuff to do tonight.  Perhaps I should end it
here.

Aloha,
Rich

=======================================================
Richard L. Pyle, PhD
Ichthyology, Bishop Museum
1525 Bernice St., Honolulu, HI 96817
Ph: (808)848-4115, Fax: (808)847-8252
email: deepreef at bishopmuseum.org
http://www.bishopmuseum.org/bishop/HBS/pylerichard.html