DTSTBIBTJTION OS INDIAN FRESHWATER FISHES. 339 



more turbulent coasts of Beloochistan, at the mouths of the 

 larger rivers and marine mud-flats off Sind and India, more espe- 

 cially Bengal and Burma. They are comparatively rare in the 

 ' open sea ; in fact at the Andamans they become scarce, except such 

 species as reside where small streams empty themselves into muddy 

 creeks. On the other hand, in the turbid sea near Moulmein 

 and amongst the islands of the Mergui archipelago they abound. 



In short, they resort to muddy waters ; and in such places their 

 feelers permit them to move about with ease, whilst their organs 

 of hearing enable them to ascertain the vicinity of any enemy or 

 the approach of some incautious prey. 



We perceive the same thing amongst those inhabiting the 

 fresh ; the larger and muddier the rivers the more are they 

 resorted to by Siluroids. Siluroids are found from the sea-coast 

 to the base of the Himalayas ; but as they generally deposit 

 their ova in the waters of the plains, and not (like many 

 Carps) in the rivers of the Subhimalayan range, they very sen- 

 sibly diminish on nearing the hills. Still one or two species of 

 Gallichrous and a Macrones find their way into the hill-waters, 

 where an Amhlyceps and Silurus, as well as the mountain -torrent 

 genera of Pseudecheneis, Glyptosternum, Euglyptosternum, and 

 Exostoma find a congenial home. Those genera Avhich are almost 

 exclusively found in hill-streams are provided with an adhesive 

 apparatus on the thorax, which, enabling them to adhere to stones, 

 prevents their being carried away by the stream. As might be 

 anticipated, their barbels are short, whilst in all the air-vessel is 

 more or less enclosed in bone. 



Eeverting again to the air-vessel, we find it in Siluroids in two 

 ways. In the marine forms it has an outer thick fibrous layer, 

 and is attached to the lower surface of the bodies and transverse 

 processes of some of the anterior vertebras, whilst from it a chain 

 of ossicles passes to the internal ear. As, however, we proceed 

 inland or towards mountains a diff'erence occurs, and in many 

 genera the air-vessel becomes partially or entirely surrounded by 

 bone. This may be eflfected by a trumpet-shaped extension of 

 the lateral processes of the first or second vertebra, or else by an 

 expansion of the most posterior of the auditory ossicles ; but in 

 either case the chain of bones is continued to the internal ear. 

 The majority of those forms which have the air-vessels enclosed 

 in a bony capsule being adapted to a mountain-torrent life leads 

 me to conjecture that such may have some connexion with sound ; 



