42 



Memoirs of the Indian Museum. [Vol. II, 



T. hardwickii ^) in the collection of the Indian Museum , but unfortunately I have not 

 been able to trace its history or provenance. The specific characters of the only known 

 species of the genus are so imperfectly recorded that I hesitate to identify the specimen , 

 which is both immature and faded. 



Benthobatis is a deep-sea genus, but the relations between it and Narcine correspond 

 so closely with those between my new genus Bengalichthys (which is a shallow-water 

 genus) and Astrape , that I am obliged to discuss it at some length. 



Both the species and the genera of the family are in some cases closelj' related , 

 and it is often difficult to distinguish between them unless fresh or well-preserved 

 material is used. This is particularly the case as regards the genera Narcine and 

 Torpedo. Stuffed skins of these forms are practically useless for purposes of identi- 

 fication, unless the greatest care has been taken to preserve the natural features. The 

 naked, glandular skin is particularly liable to become distorted, while the colour, al- 

 though sufficiently permanent in spirit unless the specimens are exposed to a strong 

 light, invariably darkens in skins, so that all the markings disappear. 



The ridges or processes on the roof and floor of the mouth form in some genera as 

 useful a diagnostic character as is the case in certain genera of Trj^gonidse, but one 

 that can only be investigated properly by dissecting out the mouth. 



The Torpedoes do not appear to be used as food in India, at any rate habitually. 

 Those caught in the seine-nets at Puri are invariably thrown away. 



Genus Torpedo, Dumeril. 



Disk distinct from tail, which is stout, bears a well-developed caudal and two dorsal fins 

 and has a fold on either side. Pectoral fins well developed ; pelvic fins not joined 

 behind the anus. Eyes distinct^ separated from spiracles. Nasal valves con- 

 fluent into a quadrangular flap. Teeth pointed, with a single cusp and a flat 

 base ; cleft of mouth wide and U-shaped. 

 Only one species has as yet been found in Indian seas, or at any rate definitely 



identified. I am much indebted to Mr. Boulenger for examining photographs of an 



Indian specimen of this species and confirming my identification. 



Torpedo marmorata , Risso. (Pis. iiirt, fig. 4, and v^, fig. 3.) 



T. galvanii, Bonaparte, Faun. Ital., vol. iii, Pesci, plate (1832-41). 

 T. marmorata, Gtinther, op. cit., p. 450. 



Disk broadly truncated in front, broader than long, rather longer than the tail. The 

 latter very broad at the base. The dorsal surface covered with minute wrinkles, 

 which run longitudinally on the anterior part of the disk and transversel)^ on the 

 posterior. The eyes nearly as large as the spiracles, from which they are sepa- 

 rated by a distance less than their own diameter. The posterior and lateral 

 margins of the spiracles provided with a row of somewhat stout, irregular, conical 

 proces.ses. 



The mouth is situated between two deep, rather lengthy longitudinal folds, which 

 approach one another posteriorly and diverge anteriorly. The teeth occupy nearly 



