Marshall (1988) described three species from sunken driftwood and Waren (1992) described two additional species from more shallow water, about 100 m, in the Mediterranean. These had been found on bottoms very rich in Posidonia fibres and on sunken driftwood. No other species have been referred to Xylodiscula. Species of Xylodiscula are similar in shell characters to the pyramidellid genus Cyclostremella Bush, 1897 (see Robertson 1973). These species, however, have whorls of more rapidly increasing diameter, lack a radula, and have apaucispiral operculum with strong growth lines. Hickman (1980) described Cyclostremella chimaera from late Eocene deep-water deposits (deeper than 200 m), Washington, U.S.A. That species has a protoconch which is virtually identical to the species classified in Xylodiscula and the whorls increase their diameter quite slowly. We therefore, and because of its habitat (species of Cyclostremella live in shallow water, 0-30 m) find it likely that Hickman's species actually belongs to Xylodiscula.

Shell minute (diameter up to 1,75 mm), subdiscoidal, thin, glassy, umbilicate. Protoconch of 1-1,75 convex whorls, apertural rim with a varix that spreads onto adjacent part of protoconch. Teleoconch of up to 2,4 whorls, slightly or distinctly flattened above periphery, smooth apart from obscure spiral lines and fine collabral growth lines. Snout flattened, quadrate. Cephalic tentacles similar, large, strongly flattened, broad, tips rounded.
Description. Shell up to 1,75 mm in diameter, subdiscoidal, spire almost or entirely flat, thin, widely umbilicate, colourless and transparent. Periostracum thin and smooth, pale buff or colourless. Planktotrophic protoconch of about 1,75 convex whorls, tip very small and strongly inrolled. First 0,75 whorl delineated by a fine groove. First 0,5 whorl with minute granules. A rounded fold immediately behind sigmoidal apertural rim, thickest over adaxial half on adapical surface, adaxial extremity crisply granulate and spreading onto preceding whorl. Protoconch elsewhere smooth. Lecithotrophic protoconch of 1 smooth, convex whorl, tip broad; a rounded varix on adapical half immediately behind sigmoidal apertural rim, its adapical extremity spreading onto protoconch tip.
Teleoconch of up to 2,4 slowly and regularly expanding whorls that are rather evenly rounded or weakly angulated at shoulder and periphery. Whorls slightly or markedly flattened above periphery, base evenly curving into umbilicus. Smooth apart from obscure spiral lines and fine collabral growth lines. Umbilicus broad, perspective to protoconch. Aperture subcircular, peristome thin, very thin over rather broad parietal contact area.
Radula with the formula 2+1+0+1+2. Lateral teeth thin in section, broad, convex, outer basal limbs forward-swinging; cutting edge almost straight, finely serrated. Marginal teeth slender, cutting edge long, finely serrated.
Remarks. Xylodiscula species differ markedly from species of Orbitestella Iredale (1917) (type-species Cyclostrema bastowi Gatliff, 1906; Recent, southern Australia) and Microdiscula Thiele (1912) (M. vanhoffeni Thiele, 1912; Recent, Antarctica) in details of radula morphology, particularly in the lack of a central tooth. Although Orbitestella and Microdiscula species have highly distinctive jaws, I was unable to ascertain whether jaws were present in Xylodiscula. Possibly confamilial, although differing in habitat and shell morphology, are the type-species of Helisalia Laseron (1954) (H. liliputia Laseron, 1954; Recent, New South Wales) and Vitrinorbis Pilsbry and Olsson (1952) (V. callistus Pilsbry and Olsson, 1952; Recent, Equador): H. liliputia lives mainly on algae in shallow water, is reddish brown and very much smaller (diameter ca. 0,5mm), V. callistus presumably lives intertidally or in the immediate sublittoral and has a crisp shell microsculpture, a stronger peripheral angulation, and a thickened apertural rim. The type-species of Cyclostremella Bush (1897) (C. humilis Bush, 1897; Recent, Atlantic, North America) has a superficially similar shell but is in fact a skeneiform pyramidellid (Robertson, 1973). Although the structure of the animal and radula suggest affinities with Rissooidea (Ponder, 1967), the systematic position of the family Orbitestellidae remains uncertain.