^8 Memoirs of the Indian Museum. [Yoh. II, 



in a few species. The only known species of Benthobatis has a dark ventral surface, 

 the only known species of Bcngalichthys a pale ventral surface. The genera of the 

 Torpedinida?, on the other hand, fall naturally into several groups separated by the 

 number (or absence) of their dorsal fins. Benthohatis belongs to one of these groups, 

 Bengalichthys to another. There can be little doubt, therefore, that the two genera 

 liave not had the same ancestry but have become like one another owing to parallel 

 or rather convergent lines of evolution. Although Bengalichthys does not live in the 

 dark abysses of the sea, we may suppose that its mode of life is very similar to that of 

 Benthobatis. Neither can be a powerful swimmer, but both, judging from the manner 

 in which the muscles of the disk are developed, must be powerful wrigglers and squirm- 

 ers. It must be remembered in this connection that the flabbiness of the flesh {i.e., 

 the muscles) of deep-sea fish which have been brought to the surface is mainly an artificial 

 condition,' not one that would be apparent if it were possible to examine the fish in 

 their natural environment. Now, quite apart from the question of the depth at which 

 a fish lives, it is quite clear that eyes may be inconvenient to an animal which wriggles 

 about in the mud at the bottom of the sea, and I have little doubt that both the fish 

 under discussion live in this way, perhaps actually burrowing into the mud, through 

 which the movements of their disks assist them to make their way. Their mouths, like 

 those of their nearest allies in both cases, are feebly developed and probably suctorial 

 in function. Neither they nor their allies can attack large organisms of any kind, and 

 it is clear that their electric organs must be weapons of defence rather than offence. 

 Perhaps both Benthobatis and Bengalichthys have become more perfectly adapted 

 for obtaining their food by sucking it from the miid owing to the degeneracy of cer- 

 tain organs that are of no use for this particular purpose. 



Bengalichthys impennis, sp. nov. (PI. iiirt, fig. 7.) 



Disk stout and muscular, pear-shaped, the broader end being in front ; the anterior 

 margin broadly convex ; the length about the same as the maximum transverse 

 diameter. Numerous white glandular pits are present on the dorsal surface, on 

 which the}' are arranged in much the same way as on that of Benthobatis mores- 

 byi. Pectoral fins consisting externally of a fringed ridge 2—3 mm. broad near the 

 edge of the dorsal surface of the posterior half of the disk. 



Tail stout, longer than the disk ; the caudal fins rounded posteriorly ; the dorsal 

 small, situated about half-way between the posterior border of the pelvic fins and 

 the caudal. The pelvic fins distinct from the disk and apparently situated on the 

 sides of the tail. 



Colour .—DorssA surface of disk and tail deep buff clouded with dark brown. Ventral 

 surface, margin of disk and fins, a large oval spot on each side of the back in front 

 of the root of the tail, a forwardly directed streak on each side of the tail, in front 



1 



The muscles of Torpedo marmorala. however, which is not a deep-sea fish, are much more flabby 

 than those of any species of Narcine (except Narcinc mollis), .Aslrape or Bengalichihvs with which I am 

 acquainted. 



