Dr. Rob Dillon, Coordinator

Tuesday, February 4, 2014

What Is A Subspecies?

Editor’s Note – This essay was subsequently published as: Dillon, R.T., Jr. (2019c) What is a subspecies?  Pp 77 - 82 in The Freshwater Gastropods of North America Volume 3, Essays on the Prosobranchs.  FWGNA Press, Charleston.

Subspecies are populations of the same species in different geographic locations, with one or more distinguishing traits.  This is the modern textbook definition [1], due primarily to the 1942 work of Ernst Mayr [2].  The concept is so simple and basic to the training that most of us probably received as undergraduates that it is difficult to imagine how anybody could become confused about it.  But they do.

No, not any of you reading this essay!  I’m sure you all get it.  But (hard though it may be to believe) there are some evolutionary biologists active in our field today who imagine that Mayr’s definition of the noun “subspecies” says more than it does.  So indulge me while I make three (obvious) points. 

Point number 1 is that no definition of the word “subspecies,” ancient or modern, has ever specified anything about heritability.  Yes certainly, the morphological differences that distinguish subspecies A and B might indeed have a genetic basis.  In fact, subspecies are sometimes referred to as “insipient species.”  But even if the “one or more distinguishing traits” are entirely ecophenotypic in their origin, arising as plastic responses to differences in environment A and environment B, the validity of the subspecific status of populations A and B would not be compromised.

Point number 2 follows from point number 1.  If the trait(s) by which a subspecies is distinguished need have no heritable basis, then populations united under the same subspecific trinomen need share no evolutionary history.  Similar ecophenotypic responses can be elicited multiple times independently.

And point number 3 is that the modern definition of the subspecies also says nothing about the existence of morphological intermediates.  Yes, it is certainly possible that subspecies A and B might be universally distinguishable.  But typically just the opposite is the case.  To quote a recent review [3], “Intergradation at the boundary between two geographic replacement forms is the touchstone of trinominalism.”  So if, for example, a third population C exists, in an intermediate region or an intermediate environment between population A and population B, the demonstration of a perfectly intermediate C phenotype would not compromise the subspecific status of A and B.  Subspecies only need be distinguishable somewhere, not everywhere.

So to take a real example.  Many rivers of the Piedmont and upper Coastal Plain from Virginia to Georgia are inhabited by populations of (typical) Pleurocera catenaria, bearing shells with prominent costae, dissected by spiral cords.  These populations reach maximum abundance in rocky shoals in medium to large rivers.   Smaller and sandier tributaries, entering the larger rivers farther east in the coastal plain, are sometimes inhabited by populations with much plainer shells, lacking spiral cords and often demonstrating reduced costation as well.  Goodrich [4] referred these populations to a subspecies, Pleurocera ("Goniobasis") catenaria dislocata.

I strongly suspect that the shell differences between dislocata populations and typical catenaria populations have little heritable basis [5].  And in fact, a population genetic survey I published in 2002 [6] suggested that the dislocata phenotype seemed to have evolved multiple times independently.  The matrix below shows that the genetic identify between a Savannah dislocata population (“Srp”) and a Savannah catenaria population (“McC”) was 0.86, and that between a Broad/Santee dislocata (“Sant”) and Broad/Santee catenaria (“Cola”) was 0.89, but that the genetic identity between the two dislocata populations was only 0.81.

And although dislocata populations are distinctive in South Carolina, they intergrade with typical catenaria throughout the Tar River drainage of North Carolina.  The figure at the top of this essay shows three shells from Red Bud Creek in Nash County, a dislocata individual on the left without spiral cords, a typical catenaria on the right with strong cords, and an intermediate individual in the middle.

But none of these considerations – the questionable heritability of the distinguishing traits, their multiple independent origins, or the existence of intermediate forms – compromises the validity of Goodrich’s subspecific designation Pleurocera catenaria dislocata.

Many of my colleagues tell me that they “don’t like subspecies.”  And without question, the description of subspecies has fallen out of favor in modern evolutionary biology.  It is an old-school concept, dependent on the taxonomist’s subjective understanding of the adjective, “distinguishing.”  And modern classifications are supposed to reflect the evolutionary history of the populations being classified.  It seems misleading to lump a subset of Pleurocera catenaria populations under the subspecific trinomen “dislocata” that have, for reasons we admittedly do not understand, come to look similar independently.

On the other hand, some of us (on rare occasions, perhaps) find the composition of freshwater gastropod communities provides helpful information about the environment from which they were sampled [7].  So if populations of Pleurocera catenaria dislocata are indeed associated with shallower, sandier creeks than typical P. catenaria catenaria, the trinomial distinction would seem to serve a valuable function.

Moreover, the Linnean system of nomenclature was instituted neither to recapitulate the evolutionary history of the organisms being classified, nor to facilitate ecological generalizations.  Carl von Linne conceived his system of taxonomy for information retrieval, like a Dewey-decimal filing system for critters [8].  And very simply, a trinomial carries more information than a binomial.

So in the final analysis, I find the ecological and “information retrieval” arguments for subspecies more compelling than the evolutionary arguments against them.  There does exist a small (but not negligible) scientific literature associated with the pleurocerid nomen “dislocata,” which would be lost (or become much more difficult to Google, anyway) if that nomen were to disappear today.

And even more the shame for such better-known pleurocerid nomina as acuta, pyrenellum, and unciale/uncialis from the American interior.  The populations described by these names are indeed geographically separate (usually) and morphologically distinctive (usually) to the point that they have been recognized as valid species until quite recently [9].  Why not save the nomina as subspecies?   As long as we are clear that there may be no heritable basis for the distinction between Pleurocera canaliculata acuta and Pleurocera canaliculata canaliculata, and that the acuta phenotype seems to have arisen many times independently, and that intermediate populations exist between the typical, robust canaliculata form and the more slender acuta form [10], it seems to me that the information indexed to the name “acuta” is worth saving.

So at present, the FWGNA project recognizes four sets of subspecies, all in the family Pleuroceridae:  the catenaria/dislocata pair, the clavaeformis/unciale pair, the floridensis/timida pair [11], and the canaliculata/pyrenellum/acuta triplet.  Although none of these subspecific distinctions likely has any evolutionary basis, all these nomina most certainly do serve important indexing functions, and hence warrant preservation.

But everything I have written in my essay this month is predicated upon a firm understanding of the word “subspecies,” as defined in the first sentence of this essay.  What mischief might be visited upon evolutionary science by our colleagues should any of them become confused?  Tune in next month!


[1] Quoted verbatim from the glossary of Price, P. W. (1996) Biological Evolution. Saunders College Publishing.

[2] Mayr, E. (1942) Systematics and the Origin of Species. Harvard University Press, Cambridge, MA.

[3]  Mallet, J. (2007) Subspecies, semispecies, superspecies.  In Encyclopedia of Biodiversity [pdf]

[4] Goodrich, C. (1942) The Pleuroceridae of the Atlantic Coastal Plain.  Occas. Pprs. Mus. Zool. Univ. Mich. 456: 1-6.

[5] Urabe, M. (2000) Phenotypic modulation by the substratum of shell sculpture in Semisulcospira reiniana (Prosobranchia: Pleuroceridae). J. Moll. Stud. 66: 53-59.  See:
  • Semisulcospira Research: A Message from The East.  [6Jan08]
[6] Dillon, R.T. & A.J. Reed (2002)  A survey of genetic variation at allozyme loci among Goniobasis populations inhabiting Atlantic drainages of the Carolinas.  Malacologia 44: 23-31. [PDF]

[7] See “Models of Species Distribution” (pp 391 – 407) in my book, The Ecology of Freshwater Mollusks (Cambridge University Press, 2000). [html]

[8] The information-retrieval function of zoological nomenclature is not incompatible with its scientific function, but not compatible, either.  See:
  • When Worlds Collide: Lumpers and splitters.  [4Sept12]
[9] Dillon, R. T., Jr. (2011)  Robust shell phenotype is a local response to stream size in the genus Pleurocera.  Malacologia 53: 265-277.  [pdf]  See:
  • Mobile Basin III: Pleurocera puzzles [12Oct09]
  • Goodbye Goniobasis, Farewell Elimia [23Mar11]
[10] Dillon, R.T.,Jr., S. J. Jacquemin & M. Pyron (2013)  Cryptic phenotypic plasticity in populations of the freshwater prosobranch snail, Pleurocera canaliculata.  Hydrobiologia 709: 117-127 [html] [pdf]  See: 
  • Pleurocera acuta is Pleurocera canaliculata [3June13]
  • Pleurocera canaliculata and the Process of Scientific Discovery [18June13]
[11] Although web pages are not (at present) available for the typical Pleurocera floridensis floridensis.  Coming one day, I promise.


  1. This post is really weird, Rob. I know you're a free thinker, but you're posting something that 1) you must know the huge majority of your readers disagree with, and 2) you're calling those who disagree with you stupid (excuse me, "confused."). That's what we on the internet generally call "trolling." I guess most trolls would also 3) post under a pseudonym, which you didn't, but...this one is weird.

    1. So Chris, what is your preferred definition of the word, "subspecies?" Do you have a literature citation you could share with us?

    2. Hm. just seeing your question a month later when looking at the more recent installment. I'm going off the top of my head, and I have no literature citation. I would say subspecies are populations or groups of populations within a species, each of which, on average, differs morphologically from other such populations or groups. I think of subspecies as having discrete geographic ranges, and while one subspecies may grade into the adjacent one, most (75% of?) individuals in a given range can be diagnosed morphologically. Though I didn't state it explicitly so far, it's implicit that genetic variation underlies the subspecies differences. If poor diet produced stunted individuals, that would not be a basis for describing them as a subspecies. We can no longer ask Mayr what he intended, but...count me as thinking he, and nearly all, evolutionary biologists would consider heritability essential to the concept. OK, now I'll go read the latest installment.

    3. 'When I use a word,' Humpty Dumpty said, in rather a scornful tone, 'it means just what I choose it to mean — neither more nor less.'
      'The question is,' said Alice, 'whether you can make words mean so many different things.'
      'The question is,' said Humpty Dumpty, 'which is to be master — that's all.'