Meiomeniidae, Salvini-Plawen, 1985

Bergmeier, Franziska S., Haszprunar, Gerhard, Todt, Christiane & Jörger, Katharina M., 2016, Lost in a taxonomic Bermuda Triangle: comparative 3 D-microanatomy of cryptic mesopsammic Solenogastres (Mollusca), Organisms Diversity & Evolution (New York, N. Y.) 16 (3), pp. 613-639 : 626-628

Meiomeniidae

The two meiomeniid genera Meioherpia and Meiomenia were erected by Salvini-Plawen (1985a) based on the presence/ absence of a dorsoterminal sense organ (DTSO) and absence/ presence of copulatory spicules. Both of these morphological characters are taxonomically highly problematic ( Morse and Norenburg 1992): a DTSO is not mentioned in the original description of M. swedmarki which does not mean, however, that it is truly absent. Moreover, reinvestigation of M. arenicola from Florida revealed the presence of a DTSO ( Morse and Norenburg 1992). Reinvestigation of the type material of M. swedmarki and/or additional material re-collected at the type locality (B topotypes^) is required; Meiomenia might possess a DTSO, which can easily be overlooked with light microscopy or even SEM when mounted in a non-suitable angle. Therefore the use of this character to distinguish both genera is dubious at present. The presence vs. absence of copulatory spicules is also problematic. Morse (1994) reports that these spicules are not always present in adult specimens and might only be formed shortly before or lost during copulation in the male phase of the hermaphrodites, and thus cannot serve as a reliable diagnostic character. Initially, both genera were described with a common atriobuccal opening by Salvini-Plawen (1985a), but later the separation of atrium and oral opening in Meioherpia was added as a diagnostic feature by García-Álvarez and Salvini-Plawen (2007), based on reinvestigations of M. atlantica ( Todt 2006) . Our own reinvestigation of the meiomeniid type species M. swedmarki , based on a series of semithin histological sections of material re-collected at the type locality by Dr. K. M. Kocot, confirmed a single anterior opening from which both atrium and pharynx diverge. Therefore, at present this is the only reliable character for discrimination of the two genera within Meiomeniidae . Our study points at further characters which might be useful in delineation of genera or species (i.e., presence/absence of a pericardium, simultaneous/sequential development of sperm and oocytes), but further comparative data is needed.

At species level, only M. swedmarki from the Pacific Northwest Coast of the USA is clearly delineated by geographic distribution and morphological characters ( Morse 1979). All subsequently described meiomeniid species,

preparation. e Overview of 3D-reconstructed M2, laterodorsal view, indicating plane of section f Section through the midgut and male gonads of M2. Arrowheads point to sperm packages; asterisks indicate intracellular cnidocysts in the midgut. Abbreviations: agl adhesive gland, ft foot, gd gonoduct, gn gonad, ln lateral nerve cord, m muscle strands likely associated with the copulatory spicules, mg midgut, n nucleus with dark nucleolus of yolky oocyte, pc pallial cavity, pgl prostatic gland, sd fused spawning duct, vn ventral nerve cord however, which are at least partially co-occurring in the temperate Western Atlantic, are taxonomically problematic: M. arenicola , M. stygalis , and M. atlantica were all originally described based on scleritome and radula characteristics investigated via light microscopy based on only one or two specimens, lacking information on potential intraspecific variation ( Salvini-Plawen 1985a). M. arenicola was originally described from North Carolina, USA ( Salvini-Plawen 1985a) and later redescribed based on material collected from Florida, USA, adding data on live observations, and details of the scleritome via SEM and histological investigations of one serially sectioned specimen ( Morse and Norenburg 1992). But since identification based on radula and scleritome characters is problematic (see present study and B Discussion^), the conspecificity of the meiomeniid from Florida and the original M. arenicola needs to be tested via molecular markers (DNA barocding) and morphological investigations from specimens collected at the type locality.

Based on the original description, there is only slight variation in the density of laterally projecting mantle sclerites and in the shape of some body scales between the two Meioherpia species and the distinguishing occurrence of abdominal spicules in M. atlantica , which B could not be ascertained^ in M. stygalis ( Salvini-Plawen 1985a) . Meiomeniid radulae show little variation useful for taxonomic purposes: both genera of Meiomeniidae are characterized by distichous radulae and only differ in the number of denticles (four to six in Meiomenia , three median denticles plus one lateral one in Meioherpia ) and the number of rows ( García-Álvarez et al. 2000). But the number of rows can vary during ontogeny with a possible intra-individual variation in the number of denticles ( Scheltema et al. 2003), and therefore, this character is considered unreliable for species delineation.

The Meiomeniidae investigated herein all belong to the genus Meioherpia , based on the distinct separation of atrium and buccal opening. Unfortunately at this point, based on the absence of clearly distinguishing features and uncertainty on intraspecific variability of scleritome characters, the specimens of Meioherpia investigated in our study cannot be allocated to one of the existing species. A varying degree in the amount and visibility of abdominal spicules within a population might indicate that only one species of Meioherpia is valid. This is contradicted, however, by microanatomical variation in the shape of the pallial cavity and anterodorsal midgut caecum as well as the number of central cells within the pedal commissural sac investigated herein (albeit the intraspecific variation of these characters also still needs to be tested). Molecular barcodes (own unpublished data) of Bermudan Meiomeniidae also support the presence of at least two different co-occurring meiomeniid lineages. A reinvestigation of the type material is urgently needed to assign these anatomical differences to described species, but unfortunately the type material could not be found in the museum’ s collection and might be lost (late Prof. Salvini-Plawen, personal communication). Hopefully, the currently unavailable holotypes can be retrieved to reinvestigate the material for further characters. But even in case the types reappear in the future, it remains unclear whether we will be able to retrieve any microanatomical information or molecular data from these whole-mounts. A designation of neotypes (or lectotypes, if paratypes exist), which present unambiguous, distinguishing character states might be needed in the future to reliable solve the taxonomic dilemma and are necessary to retrieve the Meiomeniidae from this Bermudan triangle of taxonomy.