The Western Pyrgulopsis: Testing the 147-planet hypothesis

Science is the construction of testable hypotheses about the natural world.  Have I mentioned that before?  No matter. It bears repeating.

Well-thumbed [1]

And in the history of our subdiscipline, from its 18th century origin even unto the present day, only one definition of the word “species” has yielded an hypothesis that was testable, verifiable, falsifiable, or in any sense of the word scientific.  That is the biological species concept as it arose from the Modern Synthesis, usually attributed to Ernst Mayr: 

A species is a population or group of populations reproductively isolated from all others.

 I do not mean to criticize the morphological species concept of the eighteenth and nineteenth centuries.  In creating the binomial system of nomenclature, Karl von Linne gave us a systematic method of cataloguing the biological world.  That method was entirely subjective, not scientific in the least.  But the morphological species concept led Charles Darwin to construct the first testable hypothesis about that portion of the natural world we call ours.  That species evolve by means of natural selection.

 

The year 1900 saw the birth of the science of genetics, and following shortly thereafter the Modern Synthesis of Darwin + Mendel, and from that synthesis the biological species concept.  And at the date of that dawning, and forever afterward, we evolutionary scientists have no excuse for using anything else [2].

 

Yet the air is full of excuses, almost entirely of the form, “It’s too hard!”  Common sense tells us, for example, that lions and tigers are different species.  But the big cats evolved in allopatry, and if you put them together in zoos they hybridize, and any experiment one might design to rigorously test for reproductive isolation between them is entirely impractical [3].  Well, nobody promised you science would be easy.  So quit your bitchin.

 

The challenge that allopatry presents to the rigorous application of the biological species concept is especially acute in the freshwater gastropods.  And of all the subdisciplines and sub-subdisciplines of evolutionary scientists today, we malacologists may be the laziest.  We hypothesize a unique species in every pond and puddle, and throw up our hands in despair that any actual test might be possible for our hypotheses whatsoever.

 

As Exhibit A, I offer you the 100+ nominal species of Pyrgulopsis described from the American West.  We posted two essays about that misbegotten malacofauna in 2022, starting with a [6July22] retrospective on the life and work of our friend, Dr. Bob Hershler.  And on [7Sept22] we demonstrated that the biological species concept could be applied even in such extreme situations of allopatry as exist among isolated springs in the arid west.  We synonymized one pair of Pyrgulopsis species – gibba and vinyardi – at that time, using character phase disequilibrium as a tool to test for reproductive isolation [4].

 

Pyrgulopsis sulcata [5]

Linkage disequilibrium, gametic-phase disequilibrium, or (more broadly) character phase disequilibrium were concepts that every graduate student of my generation took for granted in the 1980s, but which seem to have long since disappeared from the curriculum.  I offered a fresh review of the disequilibrium concept, beginning from first principles, in my blog post of [4Jan22].

 

So last month [22Oct26] we reviewed the recent paper by Mike Young and colleagues [6], nibbling the number of nominal species of Pyrgulopsis in the American West down from 151 to 147 on the basis of mitochondrial COI sequence.  And his most amazing result – so strange and unexpected that I still have not gotten over it – was that at approximately 100 of the 1,101 sites he studied, he found a mixture of more than one COI haplotype.  Apparently, populations of Pyrgulopsis are not as isolated as has been thought.  And with Mike Young’s splendid 5,632-line database documenting DNA sequence variation in the snails inhabiting those 1,101 sites unfolded on my desktop, I promised you all that we would seize the opportunity to apply the biological species concept more broadly and more deeply than we did in [7Sept22], potentially across the entire fauna.

 

First, a simple example.  Mike sampled eight individual springsnails from Grass Springs, in White Pine County, Nevada, sequenced them for the COI mitochondrial gene, and on that basis assigned each a specific nomen.  Five were identified as P. landyei, and three as P. sulcata.  All eight were also sequenced for the nuclear ITS2 gene, which is (of course) not physically linked to COI.

 

And let us be formal.  The null hypothesis (H0) is that mating is random among these eight individual snails.  The hypothesis we are testing (H1) is that there is some barrier to mating between the five individuals identified as landyei and the three identified as sulcata.

 

Pyrgulopsis sterilis [5]

Referring to GenBank, we confirm that the maximum COI difference among the five landyei is not quite 1%, and that the maximum COI sequence difference among the three sulcata is also not quite 1%, but that the COI sequence difference between the two groups ranges from 3.1% up to 4.3%.  Fine.  Mike Young’s two COI bins, which he labeled landyei and sulcata, are confirmed.

Next examining the ITS2 sequences for the set of eight, we find much less variation.  The maximum ITS2 sequence divergence is just three nucleotides, very near the error rate of the technique.  A set of five sequences are identical, a pair of two sequences are identical but 2 nucleotides different, and there’s a third oddball, 3 nucleotides from everywhere.  Let’s call those three bins ITS2-L, ITS2-S, and ITS2-O.

 

Then the joint distribution of the two COI mtDNA sequences and the three ITS2 nuclear DNA sequences is:

Setting aside the ITS2-O oddball, the Fisher exact probability of the null hypothesis here is 0.048.  Lumping the oddball with sulcata, that probability reduces to 0.018.  Thus, despite the genuinely tiny ITS2 genetic distances between these two sets of tiny springsnails, our results confirm that reproductive isolation does exist between the two nominal species, landyei and sulcata.

 

Let me stop now, and be absolutely clear.  This is science.  I have tested an hypothesis about the natural world.  To be precise, and being quite formal in this particular example, I stated a null hypothesis and an hypothesis to be tested, that a set of eight hydrobiid snails sampled from Grass Springs in White Pine County, Nevada, constitute two different species.  And the likelihood of the null hypothesis is p = 0.048. 

This is the first rigorous demonstration of reproductive isolation between any pair of nominal species of Pyrgulopsis in the entire history of the benighted fauna.  Real evidence now confirms that the number of valid, biological species of Pyrgulopsis in the American West is greater than one.

Let us next take a more complicated situation.  A total of 17 individual Pyrgulopsis were sampled from a set of five closely-neighboring springs in the Mud Meadow Creek area of Humboldt County, Nevada.  All springs are connected by water, at least under some conditions of rainfall, maximum distance apart no more than a couple kilometers.

Mud Meadow

Those 17 individual snails were identified by Mike Young and colleagues as four different species on the basis of their COI sequence: notidicola (7), umbilicata (6), varneri (2), and one species (“CS10”) apparently undescribed (2).  What is the evidence that these four nomina do in fact mark valid, reproductively-isolated, biological species?

 

Referring to GenBank, we find that Young identified three different ITS2 sequences from those 17 snails, which I have labeled ITS2-V, ITS2-U and ITS2-N in the table below.  And the joint distribution of the four COI sequences and the three ITS2 sequences was as follows:

Then the probability of the null hypothesis in the varneri-umbilicata case is p = 0.035, that probability in the varneri-notidicola case is p = 0.028, and that probability in the umbilicata-notidicola case is p = 0.0006.  Thus the hypothesis that notidicola, umbilicata, and varneri might be valid, biological species is supported by the non-random association of two genetic markers that should have assorted independently, COI and ITS2.

 

And similarly.  We cannot reject the null hypothesis in the case of the two individuals initially identified (on the basis of COI) as “CS10.”  The distribution of the (nuclear) ITS2 gene yields no evidence of reproductive isolation between the CS10 subsample and the subsample labeled notidicola.

 

Two final points, before summing up.  It must certainly have occurred to my alert and erudite readership that the apparent COI/ITS2 disequilibrium in the 17 snails Mike Young sampled from five springs in the Mud Meadow Creek area might be a function of geographic isolation, rather than reproductive.  Well yes, there certainly is significant geographic structuring in Mud Meadow, but that’s not the entire story.  The six umbilicata were entirely restricted to Springs B, C, D, although at Spring D, Young also identified one varneri. The sample from Spring E contained the other varneri, plus four notidicola.  And the sample from Spring A contained two notidicola and the two CS10.  So, gene flow is apparently not precluded among all four nominal species sampled across all five sites.

 

And second, with regard to what geneticists call “binning.”  This sounds like a technicality, but it is important.  I followed the lead of Mike Young with my COI binning, which based on his baroquely arbitrary technique, ultimately seems to have yielded less than 1% variation within, and greater than 3% between.  For the record, the COI sequence divergence between P. notidicola and Mike’s hypothetical “CS10” was 4.8 – 5.0%.

For my ITS2 binning, I simply inspected the data and used my biological intuition.  When I sorted the N = 17 ITS2 sequences from Mud Meadow Creek into the three bins I recognized, I allowed up to 1% error.  But I made an exception for the two CS10 sequences, which differed by 1.4% from each other.  Each of those two CS10 sequences did, however, demonstrate a 100% match to (different) notidicola sequences.  Between bins, genetic divergence at the ITS2 locus was typically in the 3 – 4% range.

 The Mud Meadow Creek situation stands in stark contrast with that at Grass Springs, where the maximum divergence among all eight of the ITS2 sequences recovered, from reproductively-isolated populations of two apparently good biological species, was just 0.6%.  At Grass Springs, single nucleotide substitutions apparently yield an evolutionarily-significant signal.  While at Mud Meadow Creek, a 5% COI sequence divergence between notidicola and CS10 did not.

And so we land, with a thud, upon the bottom line of Mike Young’s wonderful database, cringeworthy analysis, and amazing result.  When I began the present re-analysis of Mike’s 5,632-line spreadsheet, I joined him in the hope that DNA sequence data might ultimately prove more useful than they have turned out to be.  Although there is way, way more sympatry in the Pyrgulopsis fauna of the Arid West than I ever imagined, endemism is still more the rule than the exception. I confess that I was hoping to confirm that observed levels of DNA divergence between species in sympatry might be useful to calibrate a ruler for estimation in allopatry.  That forlorn hope has faded.

 

Then to summarize, over both this month’s essay and last month’s as well.  Although Mike Young’s 147-planet hypothesis is not a significant improvement over the 151-planet hypothesis that we started with, deep within his 5,632-line database we have discovered a useful tool.

• We must return to those other 100 - 2 sample sites where Mike Young documented mixtures of COI haplotypes, sequence a nuclear marker such as ITS2, and test for reproductive isolation in those 98 sites as I have outlined for the first two sites above. Dr. Kristy Pilgrim, Mike’s colleague at the USDA Genomics Center in Missoula, has advised me that they have retained DNA samples from all 4,270 of their newly-sequenced individuals [7].  It seems possible to me that, working in twos and threes, extending by transitive property, a large fraction of this mess might be cleaned up across the American West.
• And while we’re at it.  Let's prospect for morphological markers as well as genetic markers, using the technique I outlined in my essay of [7Sept22].  First do some rigorous shell morphometrics and take some good observations on soft-part anatomy at those sites where mixtures of COI haplotypes have been confirmed, then sequence, then look for disequilibrium between genetic markers and external morphology.  This would be helpful for field biologists who don’t have PCR machines mounted on the tailgates of their pickup trucks, if nothing else.
• Where populations of especially-interesting endemic species are known only in allopatry, I would prescribe a rigorous course of laboratory mating experiments.  Pyrgulopsis are tiny, and probably have quick generation times, and their environment should not be difficult to reproduce in the laboratory.  They ain’t lions and tigers, so let’s pull up our big-boy pants and do it.
• And while we’re at it.  Note the adverb “probably” above.  Laboratory culture is the second-best way to learn the biology of an organism [8].  We are utterly ignorant of fundamental details about the life history of Pyrgulopsis: fecundity, egg laying habit, growth, age and size at maturity, mating habit.  We don’t have a clue on sex ratios, much less sex determination. How many millions of dollars have been spent on these little-tiny speck snails, and we don’t know if they’re annuals, perennials, semelparous or iteroparous [9]?

And in conclusion.  I owe an apology to Jacob Burkhart and his colleagues at the US-FWS Midwest Regional Office, who initially contacted me in late September for help with their report.  Sorry, I have not.  In the end, science cannot help you with your “Species Status Assessment” any more than science can help with anything a natural resources manager might ever be called upon to do.

 

But speaking now not as a scientist; speaking simply as a private citizen of the United States.  Early in that report you sent me for review in September, I read this: 

“Section 3 of the Act (16 U.S.C 1531 et seq.) defines a “species” as including any subspecies of fish or wildlife or plants, and any distinct population segment of any species of vertebrate fish or wildlife which interbreeds when mature.”

If that is indeed the letter of the law, then the number of confirmed species of Pyrgulopsis in the American West is not N = 151 and it is not N = 147.  As of November 5, 2026, that number is N = 5.

Until we malacologists find enough guts, grit, and professional pride to address the four bullet-point suggestions I have offered above, my advice to all you state and federal natural resource managers out there would be to file your Pyrgulopsis reports deep into the bottom drawers of your back cabinets and turn your attention over to condors or ferrets or any other element of the American biota from sea to shining sea where anybody might actually have a frigging clue.

 

Notes:

 

[1] Mayr, E. (1965) Animal Species and Evolution.  Belknap: Harvard.  797 pp.

 

[2] Well, one excuse – and it’s a good one – is in the case of populations or groups of populations reproducing asexually.  My practice has been to fall back on the morphological species concept in those cases.  See, for example:

• Fun with Campeloma! [7May21]

[3] And there’s no point in it, either.

 

[4] Liu, H-P, and R. T. Dillon, Jr. (2021) Resolving the species status of Surprise Valley Pyrg (Pyrgulopsis gibba) and Vineyard Pyrg (Pyrgulopsis vinyardi).  Report to Stantec Environmental Consulting. FWGNA Circular 6: 1 – 5. [pdf]

 

[5] Hershler, R. (1998) A systematic review of the hydrobiid snails (Gastropoda: Rissooidea) of the Great Basin, western United States. Part I. Genus Pyrgulopsis. The Veliger 41: 1-132.

 

[6] Young, Michael K.; Pilgrim, Kristine L.; Miskow, Eric S.; Lundskog, Chante L.; Perez, Kathryn E.; and Schwartz, Michael K. (2026) "Molecular refinement of the taxonomy of Pyrgulopsis springsnails in western North America," Western North American Naturalist: Vol. 86: No. 2, Article 10.  Available at: https://scholarsarchive.byu.edu/wnan/vol86/iss2/10

 

[7] Alas, Kristy has also advised me that they have exhausted their entire budget for Pyrgulopsis research.  I take some comfort, however, from a very wise observation once offered to me by physician who is also a close family friend.  He said, “If money will solve your problem, you don’t have a problem.”

 

[8] The best way to learn the biology of an organism is by in situ experimentation.  No, whole-genome sequencing isn’t even in the top ten.

 

[9] For a review of the classification of life history strategy in the freshwater gastropods, see Chapter 4 in:

• Dillon, R.T., Jr.  (2000) The Ecology of Freshwater Molluscs. Cambridge University Press.  509 + xii pages, 27 tables, 99 figures.  ISBN: 0 521 35210 X.

The Western Pyrgulopsis: 151 planets in the sky?

I love retirement for many reasons, not the least of which is that it is a good excuse.  I very rarely submit grant proposals or papers to scientific journals anymore, and so very rarely volunteer to review anybody else’s.  I do not attend meetings because – you know – fixed income!  And as for serving on committees, boards, and panels, I have paid my dues.

Pyrgulopsis neritella (A,B) and landyei (C-F). From [1]

But on 28Sept26 I received a courteous email from a Dr. Jacob Burkhart from the US Fish & Wildlife Service, Midwest Regional Office, out in Minnesota somewhere, I think.  He explained that he was working on a team that is completing a “Species Status Assessment for several Pyrgulopsis springsnail species found in the Steptoe Valley of Nevada,” and asked me if I “would be willing to provide a peer review” of their “proposed treatment of springsnail taxonomy.”  And my rusty old armor cracked open just a bit.  His situation was just so pathetic.

So here is my reply, verbatim:

“Well, OK.  One of my favorite quotes in the history of the world comes from Dr. Seuss’ Cat in the Hat: ‘This mess is so big, and so deep and so tall, We cannot pick it up. There is no way at all.’ But I suppose it is my duty to help you try.”

So on 2Oct26 my new friend Jacob sent me a 6-page document entitled “NV Springsnails Taxonomy Summary For Peer Review."  In this document he and two FWS colleagues briefly reviewed their motivation, which turned out to be rooted in that “Megapetition” from the Center for Biological Diversity we reviewed way back in [14July09], see also [18May11].  Jacob and colleagues then soldiered bravely through thirty years of scattered, disorganized, ambiguous, confusing and contradictory gobbledygook, ultimately resolving to “evaluate the following entities for the purposes of this SSA report: P. landyei, P. neritella, P. sulcata, P. sterilis, and P. CS28.”  Bless their hearts.

 

And along with his little document for review, Jacob attached a huge bolus of supporting materials, including an important paper by Mike Young and five colleagues [2] that I will confess to you right now, I had been cravenly avoiding.

 

In early August I had received a comment from Mike at the bottom of my [7Sept22] post to this blog, “Just 125 of Pyrgulopsis in the American West.”  See note [3] about my species count.  I had prophesied that “One day, I feel sure, somebody will come behind (Bob Hershler) and test the 125 (Pyrgulopsis species) that remain.”  And I had concluded “I cannot imagine when, or by whom.  Not it.”

 

In August, Mike Young had answered, “We did it. And it appeared yesterday in Western North American Naturalist.”  To that comment I responded, “Great news! Could you send me a pdf?”  Alas, his answer turned out to be no.  And my effort to download a copy for myself was defeated by a paywall.

 

One of the very few things I do not love about retirement is that I have lost my access to most of the scientific literature being published today.  Sure, I could impose on colleagues in higher education to download a pdf for me, and occasionally I do.  Or, I could kick the Pyrgulopsis can down the road, and pray that the Son of Man might Return in all His Glory before I ever saw it again.

P. neritella (1.3 mm), P. landyei (1.4 mm). From [1]

Mike Young had been most cooperative when I reviewed his 2021 paper on Physa [5] back in 2024, and I knew him to be a hard worker, with excellent technical skills.  I had been impressed both by the quality and the quantity of the data he had gathered for that research effort, especially by the rigor with which he documented every nucleotide in every sequence, and every point to the right of the decimal in every sample site.  He was breathtakingly naïve regarding the biology of his study organisms, and pristinely ignorant of the published literature.  Consequently, he was utterly unable to state an hypothesis, much less test one.  But I was.  And in Mike Young's gigantic trash-snail dataset I discovered a light sprinkling of rare cases bearing directly on the point of important hypotheses.

 

It was a lot of work to mine the diamonds from a mountain of physinine slag, however, and I just was not sure I heard the call to do it again.  The Pyrgulopsis mess is today an order of magnitude worse than the Physa mess was at the dawning of my career, and going in the wrong direction.  Burch’s 1980 Bible [6] listed 16 species of Pyrgulopsis/Fontelicella in North America, and 65 species and subspecies of Physa.  Today we [7] count seven good, biological species of Physa [8], and 151 species of Pyrgulopsis.

 

Don’t get me wrong.  Science is the construction of testable hypotheses about the natural world.  To the extent that our current hypothesis about the western Pyrgulopsis is testable, it is science.  It is the intellectual equivalent of looking into the heavens and describing 151 planets, wandering randomly across the night sky [9].

 

Now here was Mike Young’s Pyrgulopsis paper laid out before me, with all 13 of his supplements in a zipped folder beside, and no sound of the trump to call me home, and I was clean out of excuses.  So, I dove in.  And I have never, in a professional career spanning over 50 years, ever laid eyes upon a more splendid dataset, a more cringeworthy analysis, or a more amazing result in the long and glorious history of freshwater gastropod population genetics [10].

 

First, the splendid dataset.  Mike and his colleagues sampled 4,270 individual Pyrgulopsis from 644 locations.  All were sequenced for the mitochondrial COI gene.  And “from each COI clade exhibiting a marked level of divergence and represented by more than one individual” Mike Smith sequenced one or more of three additional genes, mitochondrial ND1, nuclear ITS2 and nuclear 28S. 

 

To these 4,270 lines of data Mike and colleagues added 1,362 Pyrgulopsis sequences previously published, mostly of COI but a smattering of others.  And his entire 5,632-line [11], 22-column data set he made available as an excel spreadsheet in his Supplementary Material 2.  Splendid.

The 644 sample sites of Young et al. [2]

Regarding Mike Young’s analysis, it was, in two words, baroquely arbitrary.  Young characterized his species concept as “phylogenetic… requiring reciprocal monophyly” while specifying “an additional requirement for species-level rather than population-level divergence,” which is so obviously circular one wonders how such reasoning could be entertained in Western thought.  The application of this concept turned out to require such mental gymnastics such as: 

“Requiring reciprocal monophyly for the P. pilsbryana species complex would have resulted in recognizing all specimens as members of the type species, despite many populations exhibiting levels of divergence that were characteristic of valid species among other members of this genus.”

And how, one might wonder, did Mike Young determine that "characteristic" level of species divergence, which seems to have been violated in what he decided a priori must be a P. pilsbryana “complex?”

The best way to summarize Mike Young’s baroquely-arbitrary analysis is to look back from a sample of his results.  The figure below is a tiny slice off the top of Young’s Figure 2, which is intended to relay the results for his Pyrgulopsis pilsbryana “species complex.”  At left is the maximum-likelihood COI tree showing six clusters.  The seven rows of boxes at right relay the results of his “identification of candidate species.”  This little subset of seven nominal species includes four previously-described (bedfordensis, blainica, ignota, and fresti), the daggers by ignota and fresti indicating that they are “not considered part of the pilsbryana complex,” apparently interlopers from some other “complex” unspecified.  There are also two candidate species as yet undescribed (CS21 and CS22), and one “informal group” sampled from Brush Creek “not assigned to a candidate species.”

I do not know how the six COI clusters at left map onto the seven candidate species at right.  Has the “informal group” from Brush Creek been subtracted from the gene tree?  We’ll let that go and soldier on.

Young et al. [2] Figure 2, detail.

Each of the seven rows has 11 columns, each column corresponding to one “species delimitation method.”  For a key to the 11 such methods, see Young’s complete Figure 2 caption at footnote [12].  Orange boxes indicate that this particular COI clade was delimited as a species, gray means no, white means no data, and an asterisk indicates “more than one taxon within the clade.”

 

And the main result I would like to highlight is this.  Six of the seven “candidate species” shown in this tiny slice off the top of Young’s results were in fact subsequently listed in his Table 1 as “valid,” all but that "informal group" from Brush Creek.  This despite the fact that no data were available for anything other than COI sequence for bedfordensis, blainica, fresti, or the as-yet-unnamed CS21 whatsoever.

 

Mike Young’s baroquely-arbitrary classification system ultimately devolved to ad hoc binning on the basis of one, single character.  On COI sequence alone [13], he and his colleagues ultimately listed 21 Pyrgulopsis species as “candidates for synonymy” and identified 17 “previously unrecognized clades” that “were delimited as candidate species,” for a net decrement of 4 species.  Our 151-planet hypothesis is now a 147-planet hypothesis.

 

I belong to an elite generation of evolutionary biologists who are old enough to have been influenced by Sneath & Sokal's numerical taxonomy [15], but not so old as to have forgotten it.  And vividly I remember how systematic biologists of that age used to rage against “single character classifications.”  Even given a genetic basis for the character being analyzed, it was universally understood fifty years ago that populations are the unit of evolution, and universally appreciated that populations demonstrate genetic variation both within and among, and with but a single character any interspecific variance that might be attributable to reproductive isolation cannot be recognized.

 

Note that I said above, “To the extent that our current hypothesis about the western Pyrgulopsis is testable, it is science.”  The bottom line is that Mike Smith’s cringeworthy analysis does not constitute a test.

But stepping back from the species hypothesis entirely, and simply understanding a COI sequence as the single genetic marker that it is.   We have now arrived at Mike Young’s amazing result, and the grand finale of the present essay.  Quoting him verbatim: 

“And given the fairly frequent co-occurrence of taxa—at 100 of the 1101 sites in this study, which is a minimum estimate—more thorough collecting and analysis of specimens from single sites is also warranted.”

Holy crap, could that be true?  The model of the Western Pyrgulopsis I have always borne in my mind has been one of complete allopatry.  One single nominal species per spring, never any more, right?  But returning to Mike’s big Supplement #2 spreadsheet, sorting by location and running an eager finger down the first 1,000 rows or so, I counted six sample sites at which more than a single COI haplotype was discovered.  Wow.  That is quite simply, and quite profoundly, amazing.

 

Mike labeled each COI haplotype as a different “delimited species.”  And in so doing, he unwittingly committed an act of science.  He constructed an hypothesis about the natural world that can actually be tested.  Next month, we test it.

 

Notes:

 

[1] Hershler, R. (1998) A systematic review of the hydrobiid snails (Gastropoda: Rissooidea) of the Great Basin, western United States. Part I. Genus Pyrgulopsis. The Veliger 41: 1-132.

 

[2] Young, Michael K.; Pilgrim, Kristine L.; Miskow, Eric S.; Lundskog, Chante L.; Perez, Kathryn E.; and Schwartz, Michael K. (2026) "Molecular refinement of the taxonomy of Pyrgulopsis springsnails in western North America," Western North American Naturalist: Vol. 86: No. 2, Article 10.  Available at: https://scholarsarchive.byu.edu/wnan/vol86/iss2/10

 

[3] In my Bob Hershler essay of [6July22] I counted N = 126 species of Pyrgulopsis in the American West by paging through his (2017) “Annotated Checklist” [4].  Apparently molluscabase.org lists N = 151.  I have no insight to offer regarding the origin of this discrepancy.

 

[4] Hershler & Liu (2017) Annotated Checklist of Freshwater Truncatelloidean Gastropods of the Western United States, with an Illustrated Key to the Genera.  US Bureau of Land Management Technical Note 449: 1 – 142.

 

[5] Young, M.K., R. Smith K.L. Pilgrim, and M.K. Schwartz (2021) Molecular species delimitation refines the taxonomy of native and nonnative physinine snails in North America.  Scientific Reports 11: 21739. https://doi.org/10.1038/s41598-021-01197-3.  For my review, see:

• The SRALP and the SRNLP: A new hope [14May24]
• The SRALP and the SRNLP: Physa acuta were found [11June24]
• The twelve phascinating Physa of Bliss [2July24]
• Cytoplasmic male sterility in the Snake River Physa [7Aug24]

[6] This is a difficult work to cite. J.B. Burch’s North American Freshwater Snails was published in three different ways.  It was initially commissioned as an identification manual by the US EPA and published by the agency in 1982.  It was also serially published in the journal Walkerana (1980, 1982, 1988) and finally as a stand-alone volume in 1989 (Malacological Publications, Hamburg, MI).

 

[7] The clarity brought to the Physa of North America is in large part due to the many years of laboratory and field effort logged by myself and my students, the likes of which I could never hope to match with Pyrgulopsis in what remains of my sojourn in this World of Woe.  It is true that a smattering of samples have passed across my lab bench in recent years.  But I must confess, I have never seen a Pyrgulopsis on the hoof.  So, let’s give that jumbo 28-oz can of Bush’s Baked Pyrgulopsis another swift kick, okay? No?

 

[8] The confirmed species of Physa in North America are: acuta, carolinae, gyrina, jennessi, pomilia, vernalis, and zionis.  One might also count Aplexa elongata as Number 8. The jury is still out on ancillaria/parkeri and an "Owyhee Wet Rock Physa."

 

[9] Without the entry of a single new datum into evidence, Galileo and every other astronomer peering into the night sky of the seventeenth century immediately discarded Ptolomy’s geocentric model of the heavens and embraced the heliocentric model of Copernicus.  Two reasons.  First, the heliocentric theory is powerful, yielding strong predictions about the behavior of planets as yet undiscovered and capable of rejecting planets that cannot exist.  And second, the heliocentric theory is elegant.  I do not understand what has happened to our profession in these latter generations, but I do know this.  Most of the faculty in the university science departments, research institutes and government agencies of today would not know a theory – elegant, ugly or otherwise – if it bit them on the bottom.

 

[10] Mike Young’s discovery of significant sympatry in the western Pyrgulopis fauna scored 94.1 international amazingness units, edging out the 93.2 iau scored by Bianchi and colleagues back in 1994.  For a complete ranking of the top five, see:

• The third-most amazing research results ever published for the genetics of a freshwater gastropod population [1].  And the fourth-most amazing, too. [3Mar22]

[11] The excel spreadsheet downloadable as Supplementary Material 2 does indeed contain 5,632 lines, but only 5,631 are data, line #1 being the column labels.  Apparently the gene sequence(s) from one individual Pyrgulopsis have been lost, for an error rate of 0.018%

 

[12] The complete caption of Figure 2, from [2]:  "Identification of candidate species for most members of the Pyrgulopsis pilsbryana species complex, structured following highly supported clades (branches with red dots and all collapsed nodes have bootstrap support >85) from the maximum-likelihood tree for the COI gene fragment. Taxa in bold are hypothesized candidate species; those marked with a dagger are not considered part of this complex (see text). Each column depicts the results of 1 species delimitation method (A, ASAP; P, Poisson tree processes; S, statistical parsimony networks) for mitochondrial data (COI, ND1, or the concatenated sequences [CON]) or diagnosability (D) and substantial molecular support (B, bootstrap support ≥85) for nuclear data (ITS2 or 28S). Cell fill denotes whether a COI clade was delimited as a species (orange, yes; gray, no; white, no data). A cell with an asterisk denotes that more than 1 taxon within the clade was delimited. Vertically oriented labels indicate informal group membership for specimens not assigned to a candidate species."

 

[13] Well, to be completely accurate.  One of the species Mike proposed for synonymy was P. vinyardi, under P. gibba.  That confirms the result Hsiu-Ping Liu and I published in the gray literature back in 2021 [14], finding no evidence of disequilibrium between COI and penial morphology.  So in the single particular case of P. vinyardi, the evidence of synonymy is more than just mtDNA sequence.

 

[14] Liu, H-P, and R. T. Dillon, Jr. (2021) Resolving the species status of Surprise Valley Pyrg (Pyrgulopsis gibba) and Vineyard Pyrg (Pyrgulopsis vinyardi).  Report to Stantec Environmental Consulting. FWGNA Circular 6: 1 – 5. [pdf]  For a review, see:

• Just 125 species of Pyrgulopsis in the American West [7Sept22]

[15] Sneath, P.H.A, and R.R. Sokal (1973) Numerical Taxonomy: The Principles and Practice of Numerical Classification.  W. H. Freeman. 573 pp.  

The peculiar pleurocerids of the Interior Highlands III: Pleurocera canaliculata lawrencii

Subspecies are populations of the same species in different geographic locations, with one or more distinguishing traits [1].  They are races given a Latin name.  And those “distinguishing traits” are subjective – they reside in the eye of the beholder.  They might, or might not, have a genetic basis.

My readership will be familiar, I trust, with the subspecies of Pleurocera canaliculata as recognized by the FWGNA project.  In 2013 we published the discovery that the populations of pleurocerid snails bearing gracile, acutely-conical shells inhabiting waters of the United States from New York to Kansas previously identified as “Pleurocera acuta” are conspecific with populations bearing much heavier, more robust shells identified as Pleurocera canaliculata, with not one but two announcements on this blog in June of that year [2].

The subspecies of Pleurocera canaliculata

My colleagues and I coined the term “cryptic phenotypic plasticity” to describe this phenomenon, because the evidence did not suggest to us that the basis for the acuta/canaliculata distinction was heritable.  But we saved Rafinesque’s (1824) nomen “acuta” as a subspecies under Thomas Say’s (1821) canaliculata because the gracile, acutely-conical shell form and the heavy, more-robust shell form are, both in our judgment and in the judgment of many generations of malacologists who came before us, “distinguishing traits.”

The geographic range of Pleurocera canaliculata, considering all subspecies together, is huge.  Quoting chapter and verse out of the Burch Bible [3], the range of just what was historically identified as Pleurocera acuta is “Ohio River streams and tributaries; Great Lakes and tributaries; Mississippi River and westward to Nebraska and Kansas; through the Erie Canal into the basin of the Hudson River; Cumberland and Duck rivers, Tennessee.”  And that’s not even including the other subspecies – the heavily shelled typical form and the extremely slender and gracile pyrenellum of middle Tennessee and North Alabama.

Focusing in 1939 on just “the Mississippi drainage exclusive of The Ohio,” which really isn’t much of a focus at all, Goodrich [4] gave the range of P. acuta as “lakes and streams of Minnesota to streams of Louisiana on the western side of the Mississippi, and in the St. Croix River on the east side to the Illinois River.”  In that vast region he synonymized (without comment) four previously-described species under acuta (Raf 1831): elevata Say 1821(!), alexandrensis Lea 1845, haleiana Lea 1845, and “lawrencei” Lea 1869.  Goodrich also recognized two subspecies in the region: lewisii Lea 1862 and his own hinkleyi Goodrich 1921.

OK, jumping forward from the ancient history to the modern.  As regular readers of this blog might recall, in October of 2021 I spent a productive week working in Shi-Kui Wu’s beautifully-curated collections at the University of Colorado in Boulder, preparing for our recently-debuted Freshwater Gastropods of Missouri web resource.  And it was there that I first laid eyes on the shells borne by populations that Goodrich and all subsequent authors have identified as Pleurocera acuta from anywhere west of the Mississippi River.

Shells of adults from the Meramec River.

And they simply looked different.  Their shell morphology is unusually diverse, in spire height, in relative body whorl height, and especially in whorl angulation.  Typical Pleurocera canaliculata shells of all three subspecies previously known to me demonstrate a strong angulation – sometimes described as a keel – low on the whorl, at the posterior end of the aperture.  Indeed, that strong angulation manifests itself in an aperture outline classically described as “auger-shaped” or “trapezoidal.”  A trapezoidal aperture was the key character used by every authority from Isaac Lea [5] to Jack Burch to distinguish the entire genus Pleurocera from the genus Goniobasis or “Elimia,” a distinction we now understand to have been entirely artificial [6].  More evidence of which anon.

Some of the shells borne by a minority of individuals Goodrich and subsequent authors identified as “Pleurocera acuta” inhabiting streams in the Missouri Ozarks do indeed bear strong angulations low on their body whorls, especially when young.  See (A) in the sample from the Meramec River (Jefferson Co, MO 0925-02) above and below.  But most do not, especially in adulthood.  Rather, the shells born by most of the nominal “Pleurocera acuta” of the Ozarks demonstrate entirely rounded whorls, with little or no angulation or keel in evidence.

Meramec River YOY juveniles

Another unique feature of the shell morphology demonstrated by Ozark populations traditionally identified as “Pleurocera acuta” is the occasional presence of spiral cords.  The seven shells figured below were specially selected from a sample of about 40 from the Gasconade River (Maries Co, MO 0925-05) to demonstrate such sculpture.  In some shells a central cord develops so strongly that it might be described as a mid-whorl keel, unique in populations of P. canaliculata of any subspecies, in my long experience.  No malacologist of the classical school would ever identify shells such as those figured below as belonging to a gastropod of the genus Pleurocera.  From 1862 to 1980, classically-minded malacologists would have called those “Goniobasis.”  After 1980, most would have identified them as “Elimia.”

Pleurocera shells from the Gasconade R.

So in 1869 our old buddy Isaac Lea published a brief Latinate description of “Goniobasis lawrencii” from “Washita River, near Hot Springs, Arkansas, Dr. Lawrence” [7].  This he followed with a more complete English description and figure of Goniobasis “lawrencei” (note respelling) in 1874 [8].

One might feel a twinge of sympathy [9] for the good Dr. Lea, in his valiant effort to describe a shell so variable in its featurelessness: smooth, subcylindrical, rather thick, spire raised, whorls flattened.  In his remarks he focused on what “not one of the ten specimens” sent to him had, which was an apex.  But he speculated that “The upper whorls may in perfect specimens be found to be folded or carinate.”

George Tryon [10] omitted Goniobasis lawrencii entirely from his 1873 monograph of the North American “Strepomatidae,” probably because Lea’s full description and figure had not as of that date found publication. Calvin Goodrich caught it in 1927, however, respelled it “lawrenci,” and dispatched it with a single line, “The types are Pleurocera acuta Raf” [11].  He mentioned the taxon again in his 1937 catalog of Pleurocera acuta synonyms exclusive of the Ohio, now spelling it “lawrencei,” as we noted five paragraphs above [4].  But the nomen disappeared from the published literature thereafter.

From Lea [8]

And alas, continuing with variations on the theme of disappearance.  The “Washita River near Hot Springs,” spelled Ouachita River by modern cartographers, was completely impounded by a series of dams in the 20th century, beginning with the modest Remmel Dam in 1924, continuing with the impressive Carpenter Dam in 1932 and concluding with the mammoth Blakely Mountain Dam, closed in 1952. The only pleurocerid populations surviving in the Ouachita River today are the common Ozarkian P. potosiensis, hanging on in the shallow riffles above the Blakely Mountain Dam Impoundment.

 

Pleurocerid populations matching Lea’s figure nevertheless survive today scattered in tributaries of the Saline and Arkansas Rivers immediately north of the Ouachita, and become quite common and widespread in the White River and its tributaries draining the Ozark Highlands of Arkansas, and through Missouri tributaries such as the Gasconade and Meramec as figured above.

 

These populations bear shells sufficiently distinct from the other subspecies of P. canaliculata (acuta, pyrenellum, and the typical form s.s.) to warrant recognition at the subspecific level, Pleurocera canaliculata lawrencii (Lea 1869).

 

The FWGNA Project, coming into the present essay, recognized seven pairs, seven triplets, and one quartet of subspecies among the 127 valid biological species of freshwater gastropods in our 23-state study area.  Here we promote one of those seven triplets to become our second subspecific quartet. Most of the distinguishing traits among most of these 15 total sets, I feel most certain, are mostly ecophenotypic in their origin.  They are examples of “cryptic” phenotypic plasticity, because in identifying the populations bearing those traits as different species or even genera, the true origin of the variance was hidden by previous generations of Malacologists.

 

On the other hand.  One of the most influential papers I read during the course of my mostly-forgotten graduate education was published by Francisco Ayala and colleagues in 1974, “Genetic differentiation during the speciation process in Drosophila” [12].  Studying pairs of subspecies, semispecies, sibling species, non-sibling species and conspecific populations, Ayala was able to demonstrate a clear correlation between taxonomic divergence and genetic divergence, as estimated using the new technique of allozyme electrophoresis.  And in 1980, my buddy Steve Chambers was able to replicate Ayala’s findings in the enigmatic pleurocerid fauna of Florida and South Georgia [13].

 

It seems likely to me that the shell characters I have used in the present essay to resurrect Isaac Lea’s nomen “lawrencii” as a subspecies of P. canaliculata may indeed have a heritably genetic basis.  The regionalization of lawrencii populations to the Ozark Highlands, uplifted in the late Paleozoic and isolated from the center of pleurocerid diversity in the southern Appalachians by the late Cretaceous Mississippi embayment, suggests both the motive and the opportunity for genetic divergence.

 

And perhaps speciation?  Populations of Pleurocera canaliculata bearing both the typical and the acuta shell morphology are widespread in the Mississippi River and tributaries (such as the Black River) draining flatlands immediately to the east of the range of lawrencii.  Might those ranges touch?  Might the disparate forms of P. canaliculata hybridize, might they blend, might some reproductive isolation have evolved?  We’ll keep an eye peeled.

 

Notes:

 

[1] For further elaboration on the concept of the subspecies, see:

• What is a subspecies? [4Feb14]
• What subspecies are not [5Mar14]

[2] Dillon, R. T., S. J. Jacquemin & M. Pyron (2013) Cryptic phenotypic plasticity in populations of the freshwater prosobranch snail, Pleurocera canaliculata.  Hydrobiologia 709: 117-127 [pdf].  For a review, see:

• Pleurocera acuta is Pleurocera canaliculata [3June13]
• Pleurocera canaliculata and the process of scientific discovery [18June13]

[3] This is a difficult work to cite.  J. B. Burch's North American Freshwater Snails was published in three different ways.  It was initially commissioned as an identification manual by the US EPA and published by the agency in 1982.  It was also serially published in the journal Walkerana (1980, 1982, 1988) and finally as stand-alone volume in 1989 (Malacological Publications, Hamburg, MI).

 

[4] Goodrich, C. (1939) Pleuroceridae of the Mississippi River basin exclusive of the Ohio River system.  Occasional Papers of the Museum of Zoology, University of Michigan 406: 1 – 4.

 

[5] Aperture shape was the primary character Isaac Lea used in 1862 to distinguish Trypanostoma (“auger-shaped”) from Goniobasis (“subrhomboidal”).  Trypanostoma was a junior synonym of Pleurocera.  For more see:

• A House Divided [10May20]

[6] Dillon, R. T. (2011) Robust shell phenotype is a local response to stream size in the genus Pleurocera (Rafinesque 1818). Malacologia 53: 265-277 [pdf].  For more, see:

• Goodbye Goniobasis, Farewell Elimia [23Mar11]

[7] Lea, I. (1869) Descriptions of six new species of fresh water shells.  Proceedings of the Academy of Natural Sciences of Philadelphia 21: 124 – 125.

 

[8] Lea, I. (1874) Supplement to Isaac Lea’s paper on Unionidae. Journal of the Academy of Natural Sciences of Philadelphia (Second Series) 8: 55 – 69.

 

[9] The one who is authoring the present essay, however, does not:

• Isaac Lea drives me nuts [5Nov19]

[10] Tryon, G. W. (1873) Land and Freshwater shells of North America Part IV, Strepomatidae.  Smithsonian Miscellaneous Collections 253: 1 - 435.

 

[11] Goodrich, C. (1927) Some misplaced pleurocerids. Nautilus 41: 57 – 62.

 

[12] Ayala, F. J., M. L. Tracey, D. Hedgecock & R. C. Richmond (1974) Genetic differentiation during the speciation process in Drosophila.  Evolution 28: 576-592.

 

[13] Chambers, S. M. (1980) Genetic divergence between populations of Goniobasis (Pleuroceridae) occupying different drainage systems.  Malacologia 20: 63 – 81.  For more about my buddy Steve and a review of his groundbreaking research on the North American Pleuroceridae, see:

• Fred Thompson, Steve Chambers, and the pleurocerids of Florida [15Feb17]