CANOBIUS ELEGANTULUS. 175 



of vvhicli again there appears to be a circle of long, narrow, curved ossicles, whose 

 number is uncertain, apparently surrounding the entire orbit, which is proportionally 

 very large, and seems indeed to occupy almost the entire space between the snout and 

 the opercular bones. There is considerable difficulty in making out the exact form of 

 the jaw-bones. One thing is, however, certain— viz., that the maxilla has not the shape 

 usually found in the Palaeoniscidae, but has a somewhat triangular form more resembling 

 that in certain Platysomidae, such as Mesokpis, etc. The gape seems to be small, and 

 the mandible delicate ; no teeth can be seen on either jaw. The bones of the face are, 

 like those of the cranial roof, sculptured externally with tolerably fine and occasionally 

 flattened tortuous rugae. 



The bones of the shoulder-girdle present nothing calling for special remark ; their 

 external surfaces are sculptured in a manner similar to the bones of the head. 



The scales are moderate in size and mostly of the usual rhomboidal form all over the 

 body ; but there is a median row of specially large imbricating scales of a more or less 

 oval shape extending from the occiput to the origin of the dorsal fin, besides the 



Fig. 13. — Outline-sketch of tlie principal external head-plates in Canobius elegantulus, Traq. The sub- and 

 circum -orbital plates have been omitted owing to their not being preserved with sufficient distinctness, af., 

 anterior frontal or dermal ectoethinoid ; other letters as iu Fig. 12. 



usual V-scales along the upper margin of the tail. There are about thirty oblique 

 dorso-ventral bands of scales from the shoulder-girdle to the commencement of the 

 lower lobe of the caudal fin. The ganoid area of the flank-scales shows, in the first 

 place, a few delicate yet sharp vertical grooves close to and parallel with the anterior 

 margin, succeeding which the greater part of the exposed surface is sculptured with five 

 or six prominent straight ridges running across the scale, nearly parallel with the upper 

 and lower margins, ending in sharp points on the posterior margin. A very similar 

 sculpture pervades the entire squamation, though the corresponding ridges on the median 

 scales of the back are somewhat convergent, and the minute lozenge-shaped scales of the 

 ■caudal body-prolongation are nearly smooth. 



I have not seen either the pectoral or the ventral fins. The dorsal and anal are 

 nearly opposite each other, the former commencing only a little more anteriorly; both 

 fins are very similar in shape, being short-based and triangular-acuminate; each contains 

 about twenty rays, which are delicate, smooth, distantly articulated, and dichotomising 

 towards their extremities. The caudal is very heterocercal, deeply cleft, and inequilobate, 

 the upper lobe being elongated; the rays are delicate, smooth, and distantly articulated; 

 25 



