An ersatz Valvata from Bernheim Forest Kentucky

by Harry G. Lee

    The intersection of mollusk shells and productions of creatures assigned to other phyla was imprinted in my mind early on. As a college freshman, I found several specimens of an unfamiliar ~ 5 mm. snail living on rocks at the bottom of a brook tributary of the Hoosic River in Williamstown, MA. Only after a few days had passed, when I viewed them under the stereoscope in the biology lab, did I see that these shells were composed of cemented mineral grains and each contained a bristly worm-like critter. The "conchological" resemblance to certain species of the genus Valvata, e.g., <>, however, was nonetheless still quite uncanny.

    I regret that I promptly lost track of these MA specimens, but in 1976 I placed some similar specimens collected by a botanist friend from a creek near Piha Beach, Watakaries, North Is., New Zealand in my collection. A third encounter occurred on October 19, 2014 when I found another bunch of these oddities living on small nearshore rocks at the confluence of Wilson and Harrison Creeks in the Bernheim Arboretum and Research Forest Nelson-Bullitt Cos., KY [Fig. 1]. The habitat was eerily reminiscent of my first encounter over a half-century before, and the find launched a "cold case" investigation into this mystery.

    Over a century before the first of the above events, Philadelphian Isaac Lea, who with some regret I must admit is neither an ancestor or even a close relative of mine, had a similar encounter. Lack of kinship notwithstanding, Dr. Lea shared my initial read on such "shells," and dubbed them Valvata arenifera [Latin: sand-bearing] (I. Lea, 1831: 104-105, pl. 15, figs. 36a, b) [Figs. 2 and 3]. Although apparently of opposite (counterclockwise) growth, these objects are quite reminiscent of my MA material. The Quaker's specimens came from Nashville, TN, where they were taken from the Cumberland River. Lea convinced himself he saw opercula in his specimens, and he believed the agglutinated mineral matter was simply a reinforcement of the snails' shells.

    The following year Constantine Rafinesque (1832: 122; fig. on p. 121 <>) named a "new tubular fresh water shell of the Alleghany [sic] Mts." Psephides paradoxa n. gen., n. sp. Not certain it was the production of a mollusk, he did write: "This strange shell has something mysterious in it. It appears a mass of gravel; strongly cemented ..." The figure depicts a tubular structure of the same fabric seen in my and Lea's shells. Although "conchologically" quite distinct, I think neither Lea nor Rafinesque, would find the eventual taxonomic proximity, not to mention placement, of their respective species anything short of incredible. Nearly as incredible is the fact that no further mention of the Rafinesque genus (or species) except Neave could be found in the literature.

    Not much later Thomas Swainson (1840) treated a  shell looking even more like my MA specimens, his Thelidomus braziliensis, as a gastropod mollusk. The nomenclatorial context of this action was almost as bizarre as the animal involved and would benefit from a short explanation. The author actually proposed the genus-group Thelidomus twice - and in the same work (1840: 191-192, 330; and 228, 353)! The first usage appears without mention of any constituent species on p. 191-192. That taxon was made available on p. 330, where its monotype is given as Helix striolata Guilding [now known to be a synonym of the "camaenid" H. incerta Férussac, 1821]. The second usage of this generic epithet initially pops up on p. 228 next to text figure 41, which depicts a "shell" very much like mine but with no associated species group name. On p. 353 the name Thelidomus reappears again with the same text figure (now no. 113). However, this time "Braziliensis Sw[ainson]," the monotype, appears in the text block [Figure 4].* Years later, the Fist Reviser, Pilsbry (1894: 96) remedied this shocking example of Swainsonian homonymy. He gave the land snail seniority based on "position priority" (page number 330 vs. 353), an attribute which no longer mandates such preference, and thereby rendered Thelidomus Swainson, 1840: 353 non Swainson 1840: 330 permanently invalid.

    Sensing that nobody had put a generic name on such shells, but possibly quite ignorant of the Thelidomus fiasco, Carl Theodor Ernst von Siebold named the group Helicopsyche. The German zoologist, well-grounded in entomology, was a bit more savvy and thorough than his predecessors cited above. Not only did he realize these shells were the product of caddisfly larva rather than a gastropod, he mentioned the likelihood that Lea's Valvata arenifera was a congener. He even lifted and republished its type figure in support his assertion!

    Here's a taxonomic recap of these snail impersonators provided by the Entomology Department, Swedish Museum of Natural History:
Phylum Arthropoda
Class Insecta
Order Trichoptera (caddisflies)
Family Helicopsychidae Ulmer, 1906 [four genera]
Genus Helicopsyche von Siebold, 1856 [~ 230 named species in five subgenera]
Type species Helicopsyche shuttleworthi von Siebold, 1856 [subsequent designation Flint, 1964].

    Helicopsychidae is worldwide in distribution. Interestingly, in the above system Valvata arenifera Lea, 1834 [sic; error pro 1831] is treated as an invalid synonym of the later name, Helicopsyche (Feropsyche) borealis (Hagen, 1861). Perhaps a worker somewhere along the way overturned the priority of the Lea name by invoking the nomen oblitum option (ICZN 1999, Article 23.9).
    Whether von Siebold was the first to recognize the caddisfly as the perpetrator of this inter-phylum imposture is not clear, but his taxonomic initiative struck the path for proper understanding of the players involved. This geographically far-flung conchological masquerade is a stunning instance of evolutionary convergence in the geometry of an animal production - approached, but not exceeded only by certain tubicolous polychaete annelids and symbiotic arthropod-anthozoan (final plate of Abbott and Dance ,1982), and arthropod-bryozoan  <> species.

* Note the following entry stating that pleurotomarians were known only as fossils. The first living species, dubbed Pleurotomaria quoyana by Paul Fischer and A.C. Bernardi the same year as von Siebold named Helicopsyche <>, was discovered in 1855, 15 years later (Dance, 1969: 47).

Abbott, R.T. and S.P. Dance, 1982. Compendium of seashells. E.P. Dutton, New York, x + 1-411 + [1], incl. numerous text figs.

Dance, S.P., 1969. Rare Shells. University of California Press, Berkeley. (1)-128 + 24 plates.

ICZN (International Commission for Zoological Nomenclature), 1999. International Code for Zoological Nomenclature Fourth Edition. International Trust for Zoological Nomenclature, London. pp. 1-306 + i-xxix. <>

Lea, I., 1831. Observations on the naiades and descriptions of new species of that and other families. Transactions of the American Philosophical Society 4: 63-121 + pls. 3-18.   <>

Pilsbry, H. A., 1893-1895. Manual of Conchology (second series) 9. Helicidae, vol. 7; index to the helices. Academy of Natural Sciences, Philadelphia. frontispiece + xlviii + 1-366 + 71 pls. Nov. 16 to April. <>

Rafinesque, C.S., 1832. (Article) 22. Conchology. - A new tubular fresh water shell of the Alleghany  [sic] Mts. The Atlantic Journal and Friend of Knowledge 3: 121-122. <>

Swainson, W., 1840. A treatise on malacology; or the natural classification of shells and shellfish. London. vii + 419, figs. <>

von Siebold, C.T.E., 1856. Wahre Parthenogenesis bei Schmetterlingen und Bienen. Wilhelm Engelmann; Leipzig. 144 pp. <>