718 SURGEON F. DAY ON THE [Dec. 5, 



archipelago, down to which point my investigations have more or 

 less been carried on. His limits of this archipelago, lying between 

 Asia and Australia, extend from the Nicobars and the Tenasserim 

 provinces on the west, to the Philippines on the east, thus forming 

 its northern boundary ; whilst the Solomon Islands beyond New 

 Guinea define it on the east. This space he subdivides into five 

 groups, which, for my purpose, it is unnecessary to enumerate. 



My collections were made in Malabar, Madras, Orissa, and Lower 

 Bengal, up the valley of the Ganges, in the North- West Provinces 

 and the Punjaub, and through Burmah to as far as Mergui in Te- 

 nasserim. In this wide extent of country many fishes have doubtless 

 escaped my observation ; still I have obtained at least a general in- 

 sight into their distribution. The facts recorded by Hamilton 

 Buchanan, M'Clelland, Sykes, Blyth, and Jerdon are valuable, as 

 either increasing one's knowledge of localities or verifying personal 

 observations ; some references, which I am doubtful about, I am 

 compelled to omit, as when the author is not so accurate in his geo- 

 graphy as is desirable for investigations of this description. 



Siluroids inhabit all fresh waters in India and Burmah — some 

 almost generally, others locally. For an explanation of this I must 

 give a short description of these localities, especially with reference 

 to their ichthyology. These pieces of water consist of rivers, lakes, 

 or tanks, and jheels or swamps. 



The rivers may be arbitrarily divided into three portions, viz.: — 

 those existing in mountainous districts ; secondly, from such to 

 within tidal influence ; and, lastly, the tidal portions. As certain 

 visible results exist due to these three various localities, it will be 

 necessary to explain what their local causes are. 



The hill-rivers, or rather those which take their rise in hill-ranges, 

 consist of two very distinct classes, namely those which have and 

 those which have not alpine sources. Generally speaking, the 

 rivers which possess alpine sources, as those which descend from the 

 Himalayas, are chiefly replenished by the melting of snow at their 

 origins during the hot months of the year, consequently a diurnal 

 rise and fall in them is apparent, corresponding to the distance from 

 their snowy sources. During the monsoon or rainy season, doubt- 

 less the rains also assist in the melting of the snow, exclusive of 

 which, however, they are sufficient to fill the rivers in a spasmodic 

 manner. They thus form torrents, rapidly rising and as rapidly 

 subsiding. During the cold season, unreplenished by rains or melting 

 snows, they dwindle down to a small size. Of the fish inhabiting 

 these places, some of the Siluroids possess suckers or adhesive organs 

 on their heads or chests, as is also perceived in the genera Disco- 

 gnathus and Oreinus amongst the Carps. By means of these suckers 

 they retain their hold against rocks and thus prevent themselves 

 from being washed away. 



In the rivers destitute of alpine sources, as those of the Neilgher- 

 ries and the Wynaad in Madras, where snow but rarely falls and 

 never remains for months, we have a different state of affairs. 

 Amongst these must be classed the substreams or affluents of the 



