Hemipholis cordifera

Hemipholis cordifera versus Hemipholis elongata

Hemipholis cordifera was described and originally named Asterias cordifera Bosc, 1802, by the French naturalist Bosc, who reported its occurrence off the coast of Carolina (Bosc 1802). Bosc did not designate a holotype, and specimens that he examined have not been located at the Muséum national d'Histoire naturelle (Paris), the Natural History Museum (London), or the National Museum of Natural History (Washington, DC) (C. Ahearn, A. Cabrinovic, and M. Eléaume, pers. comm.). This is not surprising as Bosc reported that dried specimens were difficult to conserve. Moreover, type specimens of other marine invertebrates that he named have been lost (e.g., Kohn 1981, Blake & Maciolek 1987, Wormuth 1988). Bosc provided a brief description and a rather imprecise figure of A. cordifera, but he specified that individuals have five arms that are nine times longer than the disk, round gray disk scales with white borders, cordiform radial shields separated by three small scales, and five slender arms with three white spines that are shorter than the arm’s width. Although he neglected to mention several distinguishing features of Hemipholis species, such as nearly naked interradii and the absence of bursal slits and infradental papillae, all of the features that he cited, especially the characteristic pairs of heart-shaped radial shields, match those of the Western Atlantic Hemipholis species. Particularly compelling was Bosc’s account of extensile red tentacles beneath the arm (“qui s’alonge à la volonté de l’animal”) (Bosc 1802:114). Similar tube feet filled with red coelomocytes have been reported in only three species of North American ophiuroids (Christensen et al. 2008), of which H. cordifera alone has five arms. It is the sole species from Carolina with the precise suite of features specified by Bosc, and is known as the “blood brittle star” (Ruppert & Fox 1998:70) because of its red tube feet.

A putative species of Hemipholis collected from Charleston Harbor, South Carolina, was described and originally named Ophiura elongata Say, 1825, by the American naturalist Say, a contemporary of Bosc. Say (1825) did not designate a holotype, and specimens of O. elongata that he studied are not in the Academy of Natural Sciences of Philadelphia where they were presumed to be deposited (Ives 1889, Spamer & Bogan 1992, Thomas 1962); neither are they in the Natural History Museum (London) where some of Say’s crustacean type specimens reside (Spamer & Bogan 1992, G. Patterson, pers. comm.). Say reported that the species has a pentagonal, finely scaled disk five mm in diameter, and slender arms about 35 mm long with ovoid dorsal arm plates, quadrate ventral arm plates, and three, short, blunt arm spines. The only potentially diagnostic detail that he mentioned is that the arm spines are “hardly more than equal to half the width of the segment; the intermediate spine … more obtuse than the others, and … minutely echinated.” In contrast, the spines of H. cordifera are tapered rather than obtuse or echinulate. In comparison with Bosc’s reasonably informative account, Say’s description of O. elongata is so vague that the species’ identity is indeterminable, even to family. Furthermore, Say’s (1825:146) assertion that the “species inhabits Gorgoniae” along with Ophiothrix angulata (Say, 1825) is perplexing, since H. cordifera invariably burrows in soft sediment. Stimpson (1852:226) found that in Charleston Harbor, “It is gregarious, living in companies of twenty or thirty. The existence of these groups is indicated at low water by spaces of about a foot in diameter covered with small holes, looking very much as if a charge of shot had been fired into them. If these spots are watched as the tide rises, from each hole an arm of the star-fishes will be seen to protrude and wave about in the water, with the red tentacular filaments, by which the respiration is effected, clothing the sides.” Thus, nothing in Say’s description shows that O. elongata is a Hemipholis species, or that it is identical to H. cordifera or another South Carolinian species, or that it resembles any ophiuroid species that is epizoic on gorgonians. Since the name cannot be interpreted, one must conclude that O. elongata and equivalent combinations such as H. elongata are nomina dubia, names of unknown application.

Discrepancies in the allocation of the species’ names first came to light when Lütken (1859:180) transferred ophiuroids that he regarded as Asterias cordifera and Ophiura elongata to the genus Amphiura. Lyman (1860:203) soon realized that specimens from Charleston, South Carolina, which Lütken had identified as Amphiura elongata, must have been the species that Bosc named Asterias cordifera. Additionally, Lyman recognized that it was actually a new species of ophiuroid from St. Thomas, The Virgin Islands, which Lütken (1859) had identified as Amphiura cordifera. As a result, Lyman advocated that the species from South Carolina should be called Amphiura cordifera, and that a new name was required for the species from St. Thomas, and after consultation he reported that “Dr. Lütken agrees with me … and wishes to change the name Amphiura cordifera (Ltk.) [i.e., the name of the West Indian specimens misidentified as A. cordifera] to Amphiura Riisei, (Ltk.)” (Lyman 1860:258). Subsequently, Verrill (1899) transferred A. riisei to his newly erected genus Amphiodia, as Amphiodia riisei (Lütken, 1859).

When Lyman (1865) erected the new genus Hemipholis, he fixed A. cordifera as its type species by monotypy. It is regrettable that he treated O. elongata as a junior synonym of H. cordifera without providing a justification, and it is not known whether Lyman or his contemporaries ever examined specimens that Say himself had identified as O. elongata. Regardless, the name H. cordifera remained in general use until Koehler (1914:39), who misconstrued Lyman’s evidence regarding priority of publication and who accepted Lütken’s mistaken identifications of A. cordifera and A. elongata, erroneously substituted the name H. elongata “instead of H. cordifera, under which it is usually known.” Regrettably, H.L. Clark (1915:237) accepted Koehler’s ill-advised decision in his “Catalog of Recent ophiurans,” which still serves as a standard reference for ophiuroid nomenclature. Other authors followed suit, and when Fell (1960) transferred Hemipholis to the Ophiactidae, he cited H. elongata as the type species of the genus. Recently, H. cordifera has been treated as a nomen oblitum, a forgotten name, based on the presumption that it had not been used in the literature since 1899 (Stöhr 2010). However, the name H. cordifera was indeed used after 1899 by Ludwig (1904), Koehler (1907), and by Mortensen (1920:4) who pointedly referred to “ Hemipholis cordifera, or as it should be named according to H.L. Clark Hemipholis elongata (Say).” Consequently, H. cordifera is not an eligible nomen oblitum since the name does not meet the conditions set in Articles 23.9.1 and 23.9.2 of the ICZN (International Commission on Zoological Nomenclature 1999). Rather, in accordance with the Principle of Priority, H. cordifera should be deemed the correct name for the Western Atlantic species of Hemipholis.