-S71.J FRESHWATER SILUROIDS OF INDIA. /19 



larger snow-fed rivers ; and it is in these places that all the hill-fishes 

 breed (if we except the Loaches). 



The rivers of the plains are of course merely the continuations of 

 those descending from the hills ; but the daily rise from melted 

 snows becomes less and less apparent the further we go from their 

 snowy sources. They may be divided into two classes : in the first, 

 as the Indus, Irrawaddy, &c, a fair supply of water is always pre- 

 sent ; the second class, as the Soane or the Cauvery, become nearly 

 dry during the hot weather — this result of course being mainly due 

 to their being replenished or not by melting snows. 



The Siluroids are very extensively distributed in India and 

 Burmah, where they appear to delight in muddy water, avoiding 

 that which is clear, especially if it has a stony or gravelly bed. 

 Some which are marine are only temporary visitors to the fresh 

 waters, whilst others live entirely in fresh water ; a few appear to 

 live in estuaries, sometimes ascending the rivers, at others extending 

 their range along the sea-coast, as necessitated by the abundance or 

 absence of food. 



A few small species always reside in the streams of mountains or 

 in those flowing near their bases, whilst the larger forms prefer the 

 rivers of the plains. Some inhabit tanks only ; others prefer running 

 water ; whilst a few are common to both. Some of these fishes 

 sestivate during the hot months of the year. 



Cold does not appear to suit Siluroids, the number of genera and 

 the species rapidly decreasing as cold climates are approached. At- 

 tempting to introduce some on to the Neilgherries a few years 

 since, they perished on the journey and apparently from cold. All 

 of these belonged to the division having the air-vessel not enclosed 

 in bone. But this cannot be assumed as the reason; for I received 

 two species belonging to this group from the Pegu hills {Akysis 

 and Olyra), whilst I have also obtained Sihc?-us punctatus from an 

 elevation of 2500 feet in the Wynaad ; in none of these situations, 

 however, were the rivers snow-fed. 



The Siluroid forms which I have collected from the snow-fed 

 Himalaya rivers or those streams in the Subhimalayan range all 

 belong to the division with the air-vessel small and enclosed in a 

 bony capsule, as Pseuclecheneis, Glyptosternum, Amblyceps, and 

 Exostoma. There are, however, many other genera of this group 

 which are not found on the hills. 



Countries possessing large and muddy rivers in their plains, such 

 as Burmah, are more suited for Siluroids than such localities as 

 Madras, where the rivers are smaller, the waters clearer, and the 

 beds more stony. When investigating the fish-fauna of large tracts 

 of country this becomes very evident : thus at Hurdwar, near where 

 the river Ganges debouches into the plains, the large Ganges canal 

 commences ; both the bed of the river and that of the canal are 

 stony, the waters pretty clear, and Siluroids are rare. Near Dheeri 

 and in the Soane river, which has a pebbly bed, it is exceedingly 

 clear water when floods are absent ; but few Siluroids are found there, 

 and these mostly the little Macrones carcio, H. B. A few miles 



Proc. Zool. Soc.— 1871. No. XLVI. 



