Ophiocymbium cavernosum Lyman, 1880

Martynov, Alexander, 2010, Reassessment of the classification of the Ophiuroidea (Echinodermata), based on morphological characters. I. General character evaluation and delineation of the families Ophiomyxidae and Ophiacanthidae 2697, Zootaxa 2697, pp. 1-154 : 55-56

Ophiocymbium cavernosum Lyman, 1880

Figures 4E–G; 12H,M; 15B; 17D, G; 27H–I; 40A–D

Ophiocymbium cavernosum Lyman, 1880: 7–8 ; Lyman, 1882: 163, plate 27, figs 1–3; Paterson, 1985: 59–60, fig. 24

Material. Two specimens, ZMMU D-795, R / V “Dmitry Mendeleev”, cruise 16, sta. 1290, 23 January 1976, 54° 33' S 159° 24,0' E, depth 5410–5450 m , Sigsbee trawl. Three specimens, ZMMU D-794, R / V “Akademik Kurchatov”, cruise 11, sta. 870, 29.11.1971, 55° 7' S 25° 2' W – 55° 8' S 25° 1' W, depth 4704– 4680 m , Sigsbee trawl. One dried specimen, ZMMU D-796, R / V “Vitayz”, st. 4074, 20.10.1958, 40° 19' N 175° 45' W, depth 6065 m . One ethanol specimen, ZMMU D - 797, R / V “Vitayz”, st. 4074, 20.10.1958, 40° 19' N 175° 45' W, depth 6065 m .

Description of a specimen from sta. 1290: The disk, 10.5 mm dd, is rounded, conspicuously, but irregularly swollen (probably artificially) not extending onto the proximal arm segments. The disk plates are numerous, most of them elongate and narrow, irregular-rhomboidal – to polygonal, entirely devoid of spines and granules. Radial shields are elongated plates slightly larger than disk scales, scarcely visible at the edges of the disk (see a detailed radial shield description under paratype variations). The interradii are slightly swollen, ventrally without spinelets. There are no evident genital slits, but between arms and disk a long groove superficially resembles a genital slit. A narrow adradial genital plate articulating with the radial shield can just be seen in the distal part of the ventral side of the disk. There is no abradial genital plate, though a very small elongated scale adjacent to the adradial genital plate similar in structure and shape to the disk scales could be considered its remnant. Each jaw bears a cluster of 3–5 short spiniform teeth apically, which can be termed apical papillae. There is no larger middle apical papilla. Few additional teeth are placed more dorsally on the dental plates, similar in shape and size to the apical papillae. The total number of teeth including apical papillae is 4–8, distributed in a cone-shape tapering dorsalwards. Oral papillae are weakly developed and represent rather 1–3 block-shaped elevations, more like a fringe on the jaw than true papillae. One small spiniform papilla is placed in some interradii laterally to the apical papillae, at the ventral edge of the dental plate. The adoral shield papillae are long, slightly flattened, similar in size and shape to the arm spines, one papilla placed in the middle of each adoral shield. The oral shield is relatively narrow, triangular with a straight or slightly convex rather long distal edge, conspicuously longer than wide, completely separated from the first lateral arm plate by the adoral shields. Distally the oral shield is adjoined by a supplementary narrow plate or just an elevation formed by adjacent interradial scales. The madreporite is considerably swollen. The adoral shields are wing-shaped laterally, broadly adjoining the arm, rapidly narrowing towards the jaws, retaining thus only a narrow bar between jaws and oral shield.

True genital slits are inconspicuous small openings under the adoral shields, but a long furrow between the proximalmost arm segments and ventral disk interradius forms a false slit. Arm length is about 1.3 times the disk diameter. On all arms dorsal plates are absent for the proximal 3–4 segments, covered only with thin skin in which few small scales are embedded and the vertebrae are clearly visible, or skin apparently very thin and transparent, exposing the vertebrae. The dorsal arm plates are well developed but rather small, triangular, well separated on all segments (to about half of the length of the dorsal arm plate proximally and twice the length of the dorsal plate distally). All dorsal arm plates have pointed proximal and slightly convex distal edges. Many of the dorsal arm plates have a pit in the middle of the distal edge. Arms have weakly developed nodes, but arm spine articulations are placed on the lateral plates without forming an elevation. Laterally between two adjacent lateral plates there is a narrow space of uncovered soft tissue. On the first arm segment there are two spines, on the most proximal segments under the disk are three, on the proximal free segments four, and distally three spines. All spines are similar in size, flattened, and pointed apically. There are no hookshaped spines. The first ventral arm plate is broadly triangular; other ventral arm plates have a characteristic pole axe shape throughout the entire arm length, with straight proximal edge and convex distal edge with a small elevation. The ventral plates are separated throughout the length of the arm, except for the first two segments. The tentacle pores are large along the entire arm length. There are no tentacle scales except for the first arm segments; there is a single small scale placed on the inner part of the lateral arm plate, differing from the following arm spines only in size and absence of a defined articulation ridge.

Variability. A single specimen, also from sta. 1290, ca 12 mm dd, is similar to the holotype and shows only minor differences. Adoral shield papillae are two in number. Genital slits seem to be entirely absent, but long grooves between disk and arm segments under the disk are very evident. Block-shaped oral papillae are more evident, 2–3 in number. Two paired proximalmost oral papillae are rectangular and pointed. There are 2–3 spines on the most proximal lateral arm plates under the disk, and 4–5 on the proximal lateral plates. The spines are flattened, hollow and apically pointed. No tentacle scales on any segment.

Internal and microstructural characters. Radial shield and genital plate articulate very tightly, and have the appearance of an entire plate, broken by a fissure ( Fig. 4G). The radial shield is a very small irregular plate, with a slightly elevated articulation surface ( Fig. 4F). The articulation surface of the genital plate has a slightly elevated condyle ( Fig. 4E). The abradial genital plate is absent. Jaws are slightly elongated. Adradial sides of the jaws bear folds distally ( Fig. 12M). The dental plate is large and without folds, with few small rounded sockets ( Fig. 12H). Arm spine articulations are well developed and clearly belong to the doubleopening type ( Figs 27H–I). The muscle opening has a distinct distal lobe within the ribbed area. Nerve opening placed laterally below the level of the muscle opening. Vertebrae rather long, not keeled and with well developed zygospondylous articulation ( Figs 17D, G). Vertebral dorsal median groove almost indistinct. Podial basins are large.

Remarks. The present material covers a wide geographical range, including southern and northern Pacific and Atlantic Oceans. All studied specimens agree well with the holotype of O. cavernosum ( Figs 40A–D) in general appearance of the disk, shape of the dorsal and ventral arm plates, presence of the large uncovered tentacle pores, with only very small spiniform tentacle scale on one proximalmost segment. The oral structures of the holotype also agrees well with the present specimens in having a small cluster of spiniform apical papillae, distinctly block-shaped distal oral papillae and 2–4 additional long spiniform papillae on the adoral shield and around the second tentacle pore. The radial shields of the holotype are very small ( Fig. 40A, left) articulating with the small bar-shaped genital plate. Similar radial shields and genital plates were discovered in all present specimens ( Fig. 4E–F). The abradial genital plate is absent in all studied specimens. The holotype originated from the Kerguelen Islands region from 3568 m depth, thus relatively close geographically and bathymetrically to some specimens of the present material from the Sandwich trench. Arm spine articulations are essentially similar in both Atlantic and Pacific specimens ( Figs 27H and I) and represent a typical ophiomyxid pattern with large ribbed muscle opening and a small opening adjoining it laterally, similar to the genus Ophiomyxa (compare Figs 27G and 27H, I). Along a single lateral arm there is a transition from double-opening articulation to single-opening rounded or slightly subparallel ridge. Such variability is also characteristic for ophiomyxid articulations. No signs of the ophiacanthid volute-shaped articulations were found.