556 ME. E. w. L. HOLT ON THE [June 21 



last described, the dorsal sinus is collapsed and flattened from side 

 to side, haviiig therefore the appearance of a skinny median ridge. 

 A simiLar condition appears to have almost certainly furnished 

 the most striking feature of Vaillant's genus Aiiomaloptertis 

 (Exp. Sci. Travaill. Talism., p. 160, pi. ix.), which is founded on a 

 specimen of 60 mm. having a kind of adipose fold (" repli, sorte 

 d'adipeuse") occupying the eutii'e length of the back from the head 

 to the dorsal fin. Presuming in an allied family such a develop- 

 mental increase in the size of the eye as we have seen to occur in 

 Scojodus, it appears to me quite possible that Anomalopterus 

 pinguis is only a young stage of Bathytroctes, the dorsal fold 

 being merely a larval sinus ^ 



A younger stage of Scopelus is represented in the Faeroe 

 collection by a larva of 8 mm. (as slightly bent), shown in fig. 5. 

 The general conformation appears to clearly associate it with the 

 stage last described. The proportions of head, eye, and snout 

 appear to be as 15, 4, 3, but the posterior boundary of the head is 

 ill-defined and may be farther back than is indicated by my 

 measurements. Kelatively to the snout the eye is certainly a little 

 larger than in the last stage. There is a continuous marginal 

 fold, ampullated in the anterior dorsal region, the walls of the 

 sinus extending some little way on to the sides. The dorsal 

 is represented by a prominent interspinous ridge, beset with 

 embryonic rays, but destitute of definite permanent rays. The 

 rest of the dorsal fold bears embryonic rays, the adipose being 

 merely indicated by a prominence of the dorsal contour. Com- 

 paring the various stages observed, it would seem that the 

 development of the adipose proceeds on the same lines as the 

 first dorsal and anal, since in the 12 mm. stage (fig. 2) there is 

 an indication of the formation of true rays, which, however, is 

 never consummated. The caudal is in an advanced stage of the 

 familiar metamorphosis, the tip of the urochord projecting very 

 slightly. The anal, more developed than the dorsal, already shows 

 the proximal parts of 16 true rays. Thickened processes of the 

 bodj^-wall external to the origin of the rectum probably represent 

 the developing pelvic fins. The alimentary viscera apjjear to be 

 in much the same condition as at 11 mm. The anterior part of 

 the abdominal tract is masked by the base of the pectoral and the 

 liver. The rest of the cavity is occupied by a voluminous intestine 

 lined with well-marked annular or spiral ridges. There is little 

 or no black pigment in the peritoneal roof. Externally black 

 pigment is distributed as shown in fig 5. An aggregation near 

 the lower end of the preopercular ridge and another above the 

 middle of the anal fin appear to represent photophores, though no 

 supra-anal photophore is indicated in the more advanced stage of 

 11 mm. There are about 33 myomeres, of which about 11 or 12 



^ In Vaillant's plate {loc. cit.) is a figure of B. melanocephalus iibove that of 

 A. pinguis. Allowing i'or developmental changes on the lines indicated above 

 the two are much alike, but the radial formula given in the text are not quite 

 in harmony. 



