Widespread hybridization between Pleurocera laqueata and P. troostiana in streams of the Tennessee/Cumberland

Editor’s Note – This essay is considerably more data-heavy than we normally post in the columns of this blog.  We apologize in advance.  Readers with the fortitude to slog through a column yard of charts and graphs and regression analyses, however, will be rewarded with a real slam-bang finish next month, I promise you.  So, buck up.

In last month’s essay [18Sept24] we focused on Pleurocera laqueata, a widespread and common inhabitant of streams and rivers in Middle Tennessee, central Kentucky, and North Alabama.  The species was described by Thomas Say in 1829 [1] from specimens collected by Prof. Gerard Troost in the “Cumberland River,” an overly broad region which we ultimately restricted to Browns Creek, running through the state fairgrounds in Nashville.  The topotypic population of P. laqueata bears shells that are variably plicate but never striate, matching Say’s original description.

In direct contrast stands Pleurocera troostiana, also first collected by Prof. Troost [6Dec19] but described a bit later by Thomas Say’s successor at the ANSP, Isaac Lea around 1838 [2].  In its East Tennessee type locality, P. troostiana bears shells that are variably striate, but entirely without plication.  In a painfully detailed and ultimately exhausting series of six essays posted on this blog between December 2019 and July 2020, we reviewed the shell morphological variation demonstrated by populations of P. troostiana across Tennessee, Kentucky, and North Alabama, and the elaborate taxonomy that developed in the 19th century in an attempt to capture it.

So, if you have more than a casual interest in the taxonomy, systematics, and evolution of the North American Pleuroceridae, I would encourage you to go back and click through my 2019 - 20 series on P. troostiana from the links at footnote [3] below and download the pdf summary for your files.  Otherwise, here is a quick summary.

The range of variably-striate-but-never-plicate populations of P. troostiana, which we refer to as P. troostiana troostiana or P. troostiana sensu strictu (s.s.) is shown in blue above.  That is all you will find in East Tennessee.  West of Chattanooga, however, as the Tennessee River breaks through Walden Ridge into Alabama and Middle Tennessee, the range of P. troostiana begins to overlap with the range of P. laqueata, shown as a dashed line.  And populations of P. troostiana bearing shells that are both striate and plicate, variously identified under the subspecific nomina perstriata (yellow), edgariana (red), and lyonii (gray), begin to predominate.  This is not a coincidence.

I have hypothesized that P. troostiana hybridizes with P. laqueata several times previously on this blog, most prominently in my P. troostiana perstriata essay of [15Apr20].  But to show this, we will require a second, independent genetic character of some sort, beyond shell sculpture.  Let me back up a couple steps and refocus this entire essay away from shell sculpture, and toward shell shape.

Quite a few 19th century authorities remarked on the “spire elevation” or slenderness of the P. troostiana shell.  Isaac Lea, in his original description of 1838 [2], remarked that the shell of M. troostiana is “elevated.”  In 1841 Lea described M. teres (a troostiana synonym) as “remarkably elevated, spire much drawn out,” and ditto “spire drawn out” for a second troostiana synonym, M. strigosa [4].  John G. Anthony [5] described his M. arachnoidea (yet another troostiana synonym) as “rather thin, spire slender and much elevated” in 1854.

Now I daresay that no man nor beast who ever held a gastropod shell in hand, nor cracked it open with tooth, nor crushed it with claw, has ever in the history of this wide earth been more sensitive to that portion of the variance in shell shape that is not heritably genetic than the humble author of the present essay [6].  My filing cabinets bulge with papers vividly demonstrating ecophenotypic effects on gastropod shell morphology.  Bulge.  I cannot close them.  They remain ajar, to scar my wife’s shoulders should she dare enter the sanctum sanctorum wherein I lurk, writing quaint and curious blog posts such as this.

Shell shape and shell sculpture in pure populations.

But the heritable component of shell shape in gastropod mollusks is equally undeniable.  Working with Physa acuta in controlled conditions, I have estimated the heritability of simple shell length (SL) as h^2 = 0.429, and that of body whorl length (B) as h^2 = 0.321 [7].  In recent years I have favored the simple regression of shell width on shell length [8], or body whorl height (B) on apex height (SL), as a quick and reliable method of extracting the heritable component of shell morphological variance [9, 10] correcting for the age structure variance inevitable in wild populations.

So, last month I reported the collection of 29 topotypic P. laqueata from Browns Creek in Nashville, mapped as “L” at the top of this essay.  Of those, N = 25 were adults.  I measured total shell length for each (SL) and body whorl height (B), then calculated apex height as SL – B = A.   These data are plotted on the figure above.  The regression of B on A  was A = 0.70B – 1.19 (R = 0.77), a good fit.

I also measured N = 25 shells from a sample of P. troostiana troostiana collected from Steekee Creek at Loudon, Tennessee (35.7252, -84.3482), mapped as “T” way up above [11].  This is the type locality of J. G. Anthony’s (1854) Melania arachnoidea [5], synonymized under Isaac Lea’s Melania troostiana, see my essay of [7Jan20], FWGNA Circular 2 [pdf], or FWGNA Volume 6: 41 – 49 [publications].  The regression of body whorl height on apex height for troostiana was A = 0.98B – 1.11 (R = 0.90), an excellent fit.

Between the two elongated clusters of shell measurements, I have drawn a dashed line corresponding to the function A = 0.7B.  As a convenient approximation, it would appear that the two species can be distinguished by the ratio of shell apex height to body whorl height, greater than 0.7 for P. troostiana and less than 0.7 for P. laqueata.

Example Pleurocera from Spring Creek

So now let’s examine the Pleurocera in habiting Spring Creek (Wilson County, TN), a small tributary of the Cumberland River about 45 km east of the state fairgrounds in Nashville, mapped as “h” way up above (36.1800, -86.2411).  The Tennessee Department of Environment and Conservation took a quantitative sample of the Spring Creek macrobenthos back in August of 2014, using a kick net along three linear meters of creek bank to good effect, returning N = 185 Pleurocera [12].

I subsampled the N = 30 largest adults, measured their shells, and categorized the sculpture on their body whorl, ultimately recognizing (with some head-scratching) N = 9 striate (only), N = 8 plicate (only), and the remainder N = 13 as both striate and plicate.  The result is graphed below.

There is clearly a significant relationship between shell sculpture and shell shape in this sample of 30 pleurocerid snails, such that the fraction bearing smaller body whorls for their apex height (A > 0.7B) tend to bear striation (only) on their body whorl, and the fraction bearing larger body whorls for their apex height (A < 0.7B) bear plication (only) on their body whorl. With just those two open circles misclassified above the dashed line above, the Fisher’s exact probability is p = 0.002 [13].

Shell shape and shell sculpture in Spring Ck.

The most likely explanation for this phenomenon, which we have labeled “character phase disequilibrium” [4Jan22] is nonrandom mating.  The data graphed in the figure above strongly suggest some sort of reproductive isolation between the slender-shelled striate pleurocerid population of Spring Creek and the fat-shelled plicate population.  But the data also suggest that reproductive isolation is incomplete.  The largest fraction of the sample, 13/30 = 43%, seem to be hybrids, bearing both plication and striation on their body whorls.

Pleurocera laqueata and Pleurocera troostiana are distinct, reproductively isolated, biological species that hybridize extensively in rivers and streams throughout Middle Tennessee, southern Kentucky, and North Alabama.  Pleurocera troostiana populations are more common in the small creeks, and P. laqueata in the larger rivers, and the mixed populations in streams of intermediate size may comprise more hybrids than purebreds.

In keeping with taxonomic tradition, let us reserve the name P. laqueata for populations bearing shells entirely without striation, and P. troostiana troostiana for populations entirely without plication.  Then the subspecific nomina perstriata, edgariana, and lyonii will apply to the hybrids, according to their degree of shell sculpture.

OK, fine.  What might such widespread hybridization suggest about the evolution of the Pleuroceridae in North America?  Tune in next time.

Notes:

[1] Say, T. (1829) Descriptions of some new terrestrial and fluviatile shells of North America.  New Harmony Disseminator of Useful Knowledge 2(18): 275 – 277.

[2] Lea, Isaac (1838-39) Description of New Freshwater and Land Shells.  Transactions of the American Philosophical Society (New Series) 6: 1 – 154.

[3] Dillon, R.T., Jr.  (2020) The four subspecies of Pleurocera troostiana (Lea 1838), with synonymy.  FWGNA Circular 2: 1 - 5. [pdf]  This is a summary document for the observations, arguments, and hypotheses I advanced in a series of six blog posts to the FWGNA Blog:

• On The Trail of Professor Troost [6Dec19]
• CPP Diary: The Many Faces of Professor Troost [7Jan20]
• Huntsville Hunt [15Apr20]
• A House Divided [10May20]
• What is Melania edgariana? [5June20]
• The Return of Captain Lyon [6July20]

[4] Brief Latinate descriptions:

• Lea, Isaac (1841) Proceedings of the American Philosophical Society 2: 11 – 15.

More complete English descriptions with figures:

• Lea, Isaac (1843)  Description of New Fresh Water and Land Shells.  Transactions of the American Philosophical Society 8: 163 – 250.

[5] Anthony, J.G. (1854) Descriptions of new fluviatile shells of the genus Melania Lam., from the western states of North America.  Annals of the Lyceum of Natural History of New York 6: 80 -132.

[6] In fact, I have designated an entire topic entitled “phenotypic plasticity” in the list of “labels” at the right margin of the present blog.  If you click that link you will find 24 essays (as of October 2024) touching upon the component of shell phenotype that is not heritably genetic.  Among the most prominent:

• New clothes for The Emperor [7Feb23]
• Elimia livescens and Lithasia obovata are Pleurocera semicarinata [11July14]
• Pleurocera acuta is Pleurocera canaliculata [3June13]
• The Lymnaeidae 2012: A clue [9July12]
• Shell morphology, current, and substrate [18Feb05]

[7] Dillon, R. T., Jr. & S. J. Jacquemin (2015)  The heritability of shell morphometrics in the freshwater pulmonate gastropod Physa.  PLoS ONE 10(4): e0121962. [html] [pdf]  For a review, see:

• The heritability of shell morphology in Physa h^2 = 0.819! [15Apr15]

[8] Wethington, A.R., J. Wise, and R. T. Dillon (2009) Genetic and morphological characterization of the Physidae of South Carolina (Pulmonata: Basommatophora), with description of a new species.  The Nautilus 123: 282-292.  [pdf]

[9] Dillon, R. T. & J. D. Robinson (2016) The identity of the "fat simplex" population inhabiting Pistol Creek in Maryville, Tennessee.  Ellipsaria 18(2): 16-18. [pdf]  For a review, see:

• The fat simplex of Maryville matches type [14Oct16]

[10] Dillon, R. T. (2016)  Match of Pleurocera gabbiana (Lea, 1862) to populations cryptic under P. simplex (Say, 1825).  Ellipsaria 18(3): 10 - 12.  [pdf]  For a review, see:

• One Goodrich Missed: The skinny simplex of Maryville is Pleurocera gabbiana [14Nov16]

[11] I would have preferred to do these measurements on a sample of troostiana from Lea’s type locality at Mossy Creek, about  50 miles NE of Steekee Creek, but my sample size is insufficient.

[12] The gallon jug containing this (whole, unsorted) bulk sample was released to me by TNDEC-DWR personnel in Nashville on 14Jan21.

[13]  Here I count cases above the line and striate = 9, above and plicate = 2, below and plicate = 6, below and striate = 0.  The Fisher’s exact probability of that relationship between shell shape and shell sculpture would be p = 0.002.