Coelosphaera (Coelosphaera) lissodendoryxoides, Van, Rob W. M., 2017

Van, Rob W. M., 2017, Sponges of the Guyana Shelf, Zootaxa 1, pp. 1-225 : 133-134

Coelosphaera (Coelosphaera) lissodendoryxoides sp. nov.

Figures 82 a–g

Material examined. Holotype RMNH Por. 10512, Suriname, ‘ Snellius O.C.P.S. ’ Guyana Shelf Expedition, station F46, 6.312°N 56.57°W, depth 25–29 m, bottom sand, 7 May 1966 .

Description. The holotype consists of two fragments ( Fig. 82 a), which presumably were originally united into a single mass of 7 x 3 x 2 cm. The shape is in the form of closed fistules and digitations, 0.5–2 cm in length, erected on an encrusting base of similar sizes. Diameter of the erect parts <1 cm. Surface smooth, consistency compressible, not leathery as is usual in Coelosphaera .

Skeleton. The surface skeleton is a thickly felted mass of intercrossing megascleres, overlying a confused choanosome of loose spicules, with few spicule tracts.

Spicules. ( Figs 82 b–g) Tylotes, arcuate isochelae, sigmas.

Tylotes ( Figs 82 b–c), smooth, fusiform, thicker in the middle and subterminally constricted, with elongated tyles, in a large size range, divisible in two size classes with some overlap, (1) larger ( Figs 82 b,b1) 378– 413 –456 x 11 – 13.4 –17 µm, and (2) smaller ( Figs 82 c,c1) 236– 260 –336 x 6 – 9.2 –13 µm.

Arcuate isochelae ( Figs 82 d–e), curved, in two distinct size classes, (1) larger ( Fig. 82 d), with the flanking alae characteristically indented near the shaft, 28– 32.3 –36 µm, and (2) smaller ( Fig. 82 e), with alae proportionally longer than those of the larger isochelae, 16– 18.8 –21 µm.

Sigmas ( Figs 82 f–g), shape unremarkable, in two size classes, (1) larger ( Fig. 82 f), common, 34– 40.4 –48 µm, and (2) smaller ( Fig. 82 g), rare, only a few were encountered (n=5), 18–29 µm.

Distribution and ecology. Guyana Shelf, sandy bottom at 25–29 m depth.

Etymology. The name refers to the resemblance to species of the genus Lissodendoryx .

Remarks. The new species is unlike most Coelosphaera species in shape and consistency, lacking the leathery hollow fistules. In fact, it resembles Lissodendoryx (Lissodendoryx) sigmata ( De Laubenfels, 1949) (originally as Xytopsene ) in shape and general spiculation. However, a distinct difference, necessitating allocation of the present material to the genus Coelosphaera is the densely felted surface skeleton, which is not found in Lissodendoryx . Other details (spicule sizes, and categories) are also unlike L. (L.) sigmata and conspecificity is judged to be unlikely.

Only two other Coelosphaera (Coelosphaera) species in the Central West Atlantic region lack raphides/ trichodragmas like the present new species. These are Coelosphaera (C.) tunicata ( Schmidt, 1870) (originally Desmacidon tunicatum ) and Coelosphaera (C.) fistula Little, 1963 . I was able to examine a slide of the type of Desmacidon tunicatum , BMNH 1870.5 . 3.42, from Florida, 186 m depth (103 fthms). The type specimen is kept in the Museum of Comparative Zoology , Inv.Zool. PORb-90. It has the usual shape and fistules, while the megascleres are strongyles, the chelae are thick-shafted with short alae, 32 µm according to Schmidt , with sigmas 69 µm. Together these data do not conform to those of the present species. Schmidt (l.c.) reported Desmacidon tunicatum also from Portugal, with different sizes of chelae and sigmas. Topsent (1920) (p. 17) redescribed this material kept in the Strasbourg Museum , MZS Po 128, with label ‘type’, and discovered that it concerns a Hymedesmia . From Schmidt’s treatment of Desmacidon tunicatum and its illustration (his pl. V fig. 21) it is clear that the MCZ Florida material is the true type material.

The second species lacking trichodragmas, C. (C.) fistula , has much smaller tylotes (only up to 240 x 5 µm) and the chelae are tiny (9–12 mm) and apparently unguiferate. Assuming this species is a Coelosphaera , it is obviously different from our new species.

Other Central West Atlantic species of Coelosphaera (Coelosphaera) all have raphides/trichodragmas.

Of Coelosphaera (C.) raphidifera ( Topsent, 1889) (originally Fibularia raphidifera ) from the Campeche Bank in the Mexican part of the Gulf of Mexico, I was able to study a slide of the type, MNHN D.T. 1848. This confirmed Topsent’s (1889) (p. 16) description and subsequent ( Topsent 1894) (p. 34) discussion of its characters: megascleres strongyles of 300 x 6 µm, large isochelae with sharply pointed alae, and raphides of 60 µm long. The type is not separately preserved, but is located on a large piece of coral along with seven other different sponges, all dry and some also type specimens. The combination Coelosphaera raphidifera was also subsequently used by Hechtel (1969) (p.13) for an obviously different species of Coelosphaera from Barbados: the raphides of his specimen are much longer than Topsent’s species (430 µm) and there are two size categories of sigmas. I propose here to remove the secondary homonymy by naming Hechtel’s material Coelosphaera (Coelosphaera) barbadensis nom. nov., with holotype YPM 7745 . There is also a record of the species from Brazil ( Hechtel, 1976, in a table), but this needs further substantiation.

Further regional raphidiferous species are Coelosphaera (C.) biclavata ( Priest, 1881) (originally as Polymastia ) and Coelosphaera (C.) hechteli Van Soest, 1984 . Apart from the presence of trichodragmas and the characteristic leathery surface, there are also additional spicule differences with the present new species: only a single category of sigmas (both), only a single category of isochelae ( hechteli ), only a single category of tylotes ( biclavata ).

Figures

FIGURE 82. Coelosphaera (Coelosphaera) lissodendoryxoides sp. nov., a, habitus of holotype (two fragments) RMNH Por. 10512 (scale bar = 1 cm), b – g, SEM images of spicules, b, large tylote, b 1, details of b, c, small tylote, c 1, details of c, d, large arcuate isochela, e, small arcuate isochela, f, large sigma, g, small sigma.