A Supplementary Description of Cypridina mariae and Rediagnosis of the Genus Cylindroleberis (Ostracoda: Myodocopa: Cylindroleberididae)

The ostracod family Cylindroleberididae is based on the genus Cylindroleberis Brady, 1868, and has a complicated nomenclatural history. The type species of Cylindroleberis is Cypridina mariae Baird, 1850. Baird described only the carapace, which had been considered lost. Thus, there was no reference point for the concept C. mariae or the genus Cylindroleberis. Baird's material has now been found in the Natural History Museum, London, U.K., and is illustrated here. To clarify the taxonomic status of C. mariae and Cylindroleberis, specimens were obtained from near the type locality, and a supplementary description is presented. This includes description of appendages, particularly the first antenna and mandible, which contain important diagnostic characters. This supplementary description provides important information about C. mariae, allowing a revision of the genus Cylindroleberis, and establishing a framework for future biological research on this ostracod group.


Introduction
Cylindroleberidid ostracods are distinguished from other myodocopid families by their flat gills at the posterior of the body [1][2][3]. A cladistic analysis based on morphological and molecular characters suggests cylindroleberidids are a monophyletic clade within the Myodocopa [4]. The Cylindroleberididae have a global marine distribution and range from the intertidal to depths of 4500 metres [2]. The complicated taxonomic history and current species list of Cylindroleberididae was presented in Syme and Poore [5].
The Cylindroleberididae and Cylindroleberidinae are based on the genus Cylindroleberis Brady, 1868 [6]. The type species of Cylindroleberis is Cypridina mariae Baird, 1850 [7], subsequently designated by Sylvester-Bradley [8] despite the inadequacy of Baird's description. Like many ostracod descriptions in the nineteenth century, Baird's description was of the carapace only. Although Baird's specimens were considered lost [9], the material is stored in the Natural History Museum, London, U.K. We have examined this material and discuss it below.
Whilst other specimens have been described under the name C. mariae, we reject their synonymy for the following reasons. Brady [10] illustrated and briefly described the carapace and limbs of an ostracod that he called C. mariae. Skogsberg [9] considered the figures of the limbs to be ''incomplete and incorrect'', although he did not elaborate on the particular errors. The general problems of the figures are the apparently incorrectly-placed setae ( = bristles) (his Fig 1a), lack of some sutures (Fig 1b), truncation (Fig 1i), and uncertainty about whether the mandible, maxilla (fourth limb), fifth, sixth, seventh limbs and furca are from the male or female specimen (not stated in figure captions). More specific problems are the inability to determine the status of generic diagnostic characters: whether the s-seta of the first antenna has a proximal filament and whether the mandible has a lateral e-seta. Because of this, we agree with Skogsberg's view that Brady's specimens do not permit a certain identification and are not clearly referable to any other known material.
Brady and Norman [11] also illustrated specimens identified as C. mariae. Their figures are of an adult male and a juvenile male. The latter is referred to as a female, but the endopod of the second antenna has the typical ''robust'' form of juvenile males in the family Cylindroleberididae. Because of uncertainty in determining the conspecificity of the sexes, we do not consider these specimens as synonymous with Baird's concept of C. mariae. Other references to C. mariae by Cushman [12] and Juday [13] are of American species and so are not near the type locality for C. mariae Baird. Muller's [14] specimen of C. mariae was synonymized into of C. grimaldi vicina by Skogsberg [9].
Cylindroleberis mariae appears in various electronic databases and recent literature [15]. We consider a supplementary description necessary to clarify the nominal taxon Cypridina mariae (now Cylindroleberis mariae) and because it has an impact on the accepted definition of Cylindroleberis.
The type locality of Cypridina mariae Baird is ''off the Isle of Skye'', Scotland, U.K. Of the available material in the National Museum of Scotland, Edinburgh, specimens from the Shetland Islands, were chosen for examination. These are described herein. Besides Baird's material, the only illustrated specimens are carapaces from a male and female from Norway [8], which are no more informative. The supplementary description herein is of an adult female and juvenile male consistent with what little is known from Baird's material and description of the holotype (carapace shape). The description fits with the current diagnosis of Cylindroleberis except for the arrangement of filaments on the s-seta, and thus the genus diagnosis of Cylindroleberis is expanded to include the variation in this character.
The aim of this project was to clarify the morphology of the species C. mariae and, as a consequence, the genus Cylindroleberis. The illustration of the type material of C. mariae, along with a detailed supplementary description, clarifies important morphological characters of that species, and the genus Cylindroleberis has been revised in that context.  [7], subsequent designation by Sylvester-Bradley [8].
Remarks: The genus was rediagnosed in detail by Skogsberg [9] (as Asterope) and Poulsen [19] (as Asteropina). Poulsen's diagnosis defined the adult female s-seta as having a filament configuration of 1+6, which is true in all known species. Cylindroleberis mariae has a configuration of 0+9, and the A-1 male with 1+6. Historically, the presence or absence of a proximal filament on the s-seta has been considered a good generic character. However, within the family, the proximal filament varies continuously from absent to short to long, and the pattern has been interpreted differently. A ''long'' proximal filament can alternatively be interpreted as a ''short'' terminal filament. For example, species in the genus Bathyleberis Kornicker, 1975 [1], all with 7 filaments in total, show the full range of this character, with some described as 1+6 and some as 0+7. Further, the ontogeny of this character is not clear: juvenile females may lack the proximal filament where adults have it (Cylindroleberis vibex A-2 instar), and the reverse (Synasterope calix A-2 instar) [20]. The A-1 male of C. mariae described below with 1+6 differs in this character from the female.
Thus, the generic diagnosis of Cylindroleberis is expanded to include s-seta arrangements of either 1+6 or 0+7-9 filaments, i.e., between 7-9 filaments in total. The generic diagnosis of Polyleberis (monotypic: Polyleberis mackenziei Kornicker, 1974 [18]) includes an s-seta with 0+7 to 0+9 filaments, (differing between individuals in the species). Because this definition includes C. mariae, we consider it no longer justified to keep Polyleberis separated from Cylindroleberis on the basis of this character alone, and it is synonymized herein.
C. vicina was previously a subspecies C. grimaldi vicina (Skogsberg, 1920) [9]. However, we believe its differences from C. grimaldi warrant species status, and here raise it to the species rank as C. vicina. It is diagnosed by no setae on mandiblular basale dorsal margin at midlength, a carapace length greater than 1.5 mm, and fewer than 5 anteroventral setae on the sixth limb. C. nodulifera and some specimens of C. variabilis also lack setae at this position on the mandibular basale; these species are smaller than C. vicina (carapace length less than 1.5 mm). C. marranyin also lacks setae at this position on the mandibular basale, but has greater than 5 anteroventral setae on the sixth limb.
A further species, C. rangiroaensis Hartmann, 1984 [26], has a long mandibular exopod which excludes it from Cylindroleberis [21] . Here, we place it in Synasterope Kornicker, 1975 [1], because it shares with other members of this genus: first antenna s-seta with proximal+distal filament configuration 0+6, first antenna d-seta absent, mandible exopod greater than 50% length of first endopod article, and mandible e-seta absent.
Cylindroleberis   Carapace: elongate, incisur at midheight, posterior end evenly rounded (Figure 2A First antenna ( Figure 2D): Article 2 with 1 spinous dorsal seta and 1 lateral seta with faint spines. Article 3 with 1 short ventral seta and 6 dorsal setae-all setae with long spines except seta 3 (no spines) and seta 6 (shorter spines). Article 4 with 1 dorsal medial seta with short marginal spines and 2 ventral setae. Article 5 with sensory seta with no short proximal and 9 terminal filaments. Article 6 with medial seta with faint spines, reaching tip of a-claw. Article 7 with a-claw, b-seta with 5 marginal filaments, c-seta with 6 marginal filaments. Article 8 with minute peg d-seta, e-seta bare with blunt tip, f-seta bent dorsally with 4 marginal filaments, g-seta with 6 marginal filaments.  with stout a-, b-, c-, and d-setae, 1 slender seta proximal to a-seta; medial side with 7 ''cleaning'' setae, and 1 long g-seta distal to base of d-seta; lateral side with no e-seta between b-and c-setae, and 1 long f-seta between c-and d-setae. 3rd endopod article with stout dorsal claw, 3 stout setae with faint spines, and 2 slender shorter setae, the longer one attached laterally.

Discussion
The supplementary description herein agrees with the size and shape of the holotype, and with Baird's description [7], which states ''carapace valves elongate, oval, of exactly the same size at each extremity; extremities rounded. Dorsal and ventral margins nearly plane, or very slightly arched'', ''notch or ventral margin of anterior extremity blunt, leaving the upper and lower margins of the notch very obtuse''.
Cylindroleberis mariae is similar to Polyleberis mackenziei Kornicker, 1974 [18]. The holotype of P. mackenziei was described [18] but was considered not to be mature. The type locality is Gulf of Naples, Italy. Further specimens were described later from Mauritania and descriptions provided of an adult female and juvenile male [22]. The carapace of C. mariae is slightly longer than that of P. mackenziei-2.15 mm compared to the longest P. mackenziei specimen recorded of 1.96 mm. There are 10 furcal claws compared to 9, and there is no anterior spine on the upper lip. The A-1 male of C. mariae described here has an s-seta configuration of 1+6. This is similar to the juvenile male of P. mackenziei which is described as having a configuration of 0+7, with the first terminal filament being half the length of the others. The differences are insufficient to warrant a generic distinction.
Apart from the different s-seta configuration, Cylindroleberis mariae can be distinguished from other species of Cylindroleberis in carapace size (only C. vix also has a carapace length of greater than 2 mm in the adult female), and the presence of the distomedial seta on the second antenna protopod (only present in C. vibex).
This supplementary description allows clarification of the attributes of C. mariae, and reduces uncertainty in the concept of this species. With this information, the genus Cylindroleberis is also revised, providing a sound taxonomic framework for ecological, physiological and evolutionary research on this ostracod family.

Materials and Methods
Material was examined under compound and dissecting microscopes (up to X600 magnification). Dissections were made using tungsten needles, and appendages were mounted on microscope slides. Pencil illustrations were made using a camera lucida, and were then scanned and digitally traced using Adobe Illustrator software.