237 



find any trace of teeth whatever ; but by its anal fin being very 

 short, formed of six branched rays, its lateral line running along 

 the medium line of the tail ; its dorsal fin with a strongly 

 serrated spine, situated a little behind the ventrals, its abdomen 

 compressed into a sharp ridge behind the ventrals, and the 

 absence of barbels, it seems to come very near to Dr. Bleeker's 

 genus, Roliteicthys ; but the very large scales which cover its 

 body make it distinct from this Indian genus, and its enor- 

 mously developed belly gives it a very particular appearance. 



I have just said tbat on the fresh specimens there is to trace 

 whatever of teeth ; but on the dried one there is visible on each 

 side of the palate a very large horny tubercular one, having the 

 form of a sugar loaf; but this seems only to have appeared through 

 the skin by the effect of dessication. 



NEOCAEASSUIS VENTRICOSTJS. 



The body is very high, very thick, and the lower profile 

 remarkably rounded and convex ; the mouth is rather extensible. 

 The height of the body is contained two and one-third times in 

 the total length ; head four and a-half times in the same ; orbit 

 four and four-tenth times in the length of head. No barbels ; 

 head without scales, except under the eye. The scales on the 

 infraorbital are very large, in small number, radiated, and serrated 

 on their edge ; operculum covered with arched, convergent striae ; 

 scales of the body very large — thirty on the lateral line, fifteen 

 on the oblique one, of which seven are over the lateral line 

 (which runs over the eighth) ; they are covered with, fine concentric 

 striae ; they appear, when seen through a lens, finely serrated. 

 The lateral line is always straight, rather elevated near the head, 

 and runs over the middle of the tail ; it is formed of a succession 

 of elevated ridges, which only extend over the two first thirds of 

 the scale, and end (on the living specimen) bj a rounded point 

 of an obscure colour. The first dorsal is formed of four spines, 

 of which the first is very small, the second rather longer, the 

 third nearly three times as long as the second, and the fourth 

 very long, rather strong, with numerous spinelets on its inner side ; 

 it is four times longer than the third. The rays number seven- 

 teen, and the first are longer than the longest spines, but they 

 decrease in height as they extend backwards. The caudal is 



