40 THE VOYAGE OF H.M.S. CHALLENGER. 



July 2, 1875, and is 12 mm. long and 1 mm. broad. It represents evidently a some- 

 what less advanced stage of the same group of fishes than the following. The snout is 

 much distorted, but the general form and structure of its component parts were probably 

 the same as in the next largest specimen, although rather shorter. The e5^e is extremely 

 large and of an oval shape, occupying nearly the whole depth of the head. The termin- 

 ation of the tail is diphycereal ; the embryonic fin-fringe commences on the back as a 

 transparent rayless membrane in about the middle of the length of the fish, the dorsal 

 rays becoming a little more distinct further behind, all being much more feeble than 

 those of the ventral side. The anal fin becomes distinct behind the middle of the length 

 of the body, is higher than the dorsal, and after having proceeded for a short distance, 

 the rays become abruptly longer, stronger, and more closely set, and are supported by 

 haemal elements nearly to the end of the chorda. The situation of the vent, and the 

 position of the paired fins, if they be present, cannot be ascertained. 



The second specimen (PI. V. fig. B) is 14 mm. long, and was obtained on May 

 3, 1876, on the surface of the North Atlantic. It is elongate, band-shaped, with 

 pointed, subconical snout. The mandibulary and maxillary bones are long, and the 

 cleft of the mouth extends to below the eye. Teeth, as represented by Richardson in 

 Prymnothonus, cannot be detected in this specimen. The eye is large. The vent is 

 clearly distinct at a distance from the head which is nearly equal to the length of the 

 latter. Behind the vent a low embryonic fringe commences, which is continued round 

 the tail, terminating on the back in about the middle of the length of the fish. The 

 fringe is striated or rayed throughout, but the rays are exceedingly fine in the anterior 

 half of the anal portion, become then suddenly longer and stronger, forming a kind of 

 lobe, and are shorter again round the caudal extremity. The termination of the 

 vertebral column is heterocercal, the end of the chorda being bent upwards and continued 

 beyond a group of haemal elements sujiporting an assemblage of stronger rays which in the 

 adult fish would be developed into a distinct caudal fin. Of the paired fins the pectorals 

 are clearly develoj^ed, but I cannot detect a trace of ventrals. Behind and above the 

 pectorals a larger and two smaller roundish black spots are visible below the transparent 

 integument, which are the bver and other abdominal organs. A similar black spot 

 existed in the specimen of Prymnothonus hookeri, and was misunderstood by Eichardson, 

 who, having a drawing only for his description, considei-ed it to represent the gill- 

 opening. As far as I am able to judge from the specimen described here, the gill- 

 opening is wide, and at its usual place. 



The specimen of which Richardson {loc. cit.) has given a figure, reproduced here 

 (PI. V. fig. C), seems to have been intermediate, as regards development, between that 

 just described and the third discovered by the Challenger. The termination of the tail 

 is homocercal, with a distinctly difi'erentiated caudal fin, which, however, is continuous 

 with the dorsal and anal. The dorsal extends only a short way forward on the tail, but 



