1887.] Deep Waters of the Bay of Bengal, 231 



formity of temperature also shows a uniform origin, for the change of 

 tempei'ature in large bodies of water is slow. The submarine inflow 

 which must come from the southward to make up for the great eva- 

 poration of the bay, is therefore probably uniform in temperature and 

 widely spread. It should be mentioned, however, that the observations 

 have been entirely made in the fine season between November and May ; 

 and, although surface disturbance by wind would be unlikely to affect 

 deep currents, yet it is just possible that the great climatic difference 

 between the two monsoons may upset the balance in other ways and so 

 alter the rate of submarine inflow. 



Slow-moving deep currents such as would exist at the bottom of 

 the Bay of Bengal cannot be accurately tested by any means yet de- 

 vised. It is true that the United States Coast Survey have lately de- 

 tailed a vessel specially to take the strength and direction of the Gulf 

 Stream in the Straits of Florida, and that a careful section was obtain- 

 ed, by means of special fittings, of the strength and volume of the 

 stream in that its narrowest portion. But even there, where consider- 

 able movement might have been expected, it never exceeded one mile 

 an hour at the bottom, which was only some 400 fathoms deep. It is 

 then unlikely that the same instrument, ingenious as it is, would be of 

 any service at the greater depths owing to the far greater sources of 

 error and the very much slower movement. 



Whatever current-meter is used will have to be very sensitive, and 

 must be anchored at the bottom for at least thirty hours, without being 

 fouled by fish or weed, to minimise errors of lowering it down, pres- 

 sure, hauling up, &c. 



Surface currents are, however, guaged with comparative ease, and 

 the depth of these as yet found by the " Investigator " range from 80 

 fathoms to 175 fathoms in the open waters of the bay. 



Off the Rivers Hughli and Ganges, a higher temperature was fre- 

 quently found at five to ten fathoms than at the surface, and this pro- 

 bably occurs when the fresh water of the rivers brings down a lower 

 temperature, and temporarily overlies the salt water. 



Future observations of temperature will be unlikely to vary more 

 than one degree Fahr. from the mean curve given in the accompanying 

 plate for depths greater than 100 fathoms ; but at depths between 10 

 and 100 fathoms they may vary 5° either way. 



A remarkable verification of depth by aid of the thermal reading 

 was made in March last. When tabulating the mean temperatures 

 last January, I noticed that the temperature shewn at a cast of 1400 

 fathoms, made in 1885, in Lat. 19°.34' N., Long. 91°.07' E., was far more 

 suitable to a depth of 1000 fathoms. Upon reference to the Sounding 



