Shell resembling Scutellina but with a blunt subcentral apex. Soft parts resembling Acmaea except in the following details: Animal blind, with the front part of the head between the tentacles and above the muzzle much produced upward and forward, extending considerably farther forward than the end of the muzzle, which is marginated with lappets at the outer corners. Jaw thin, translucent. Gill exactly as in Acmaea; sides of foot and mantle edge simple, nearly smooth. Dental formula 0 (1.0.1.) 0; teeth large, with transverse pectinated or denticulated cusps, the serrated edge of which is turned toward the median line. The number of teeth is the smallest in any known limpet.

Shell 8.00-39.8 mm long at maturity, white, teleoconch I smooth, teleoconch II radially and concentrically sculptured (protoconch unknown: deciduous). Radular formula 0+1+0+1+0, each tooth considerably longer than basal plate, comprising three or perhaps four fused laterals. Digestive system containing symbiotic bacteria that are involved in digestion of decaying wood upon which they feed.

The highly distinctive Pectinodonta radula comprises a single, alternately slightly offset pair of anteriorly converging, elongate, dorso-ventrally short, straight, dorso-ventrally curved, saw-like teeth arranged in a long series. The large anterior cusp in each crossrow seats in the crossrow in front and in the crossrow behind (but does not originate from them), extending about a third of the distance along the crossrow behind that: in other words each tooth is roughly three times longer than the crossrow to which it is attached. Each tooth has been interpreted as a fusion of three laterals (Habe 1949; Hickman 1983; Marshall 1985; Lindberg 1988), the boundaries of which are defined by sutures behind the two (anteriormost) unicuspid teeth. The rest of the tooth has been interpreted as a multicuspid third lateral, though the presence of a small, blunt subterminal projection (vestigial cusp?) and the deeper separation between the fourth cusp and adjacent ones, suggests that the third and fourth cusps may represent the third tooth and the remainder a multicus¬pid fourth lateral. Conversely, Okutani et al. (1992) considered that the putatively fused teeth in Pectinodonta "all come out from the common shaft" and thus comprise only a single pair of lateral teeth, each of which consists of three parts. They also did not accept that the lateral teeth in Serradonta vestimentifericola were the result of fusion, though sutures are clearly defined and it seems likely to us that each tooth in this species is indeed the result of fusion of three teeth.