ON DEW, &c. 261 



immediate cause, the additional moisture, to 

 the peculiar humefying quality, which they 

 supposed that luminary to possess. This false 

 theory has, probably, contributed to discredit, 

 with the moderns, the circumstance which it 

 was employed to explain. 



VI. The last fact, of which I shall treat in 

 this Essay, is the formation of ice, during the 

 night in Bengal, while the temperature of the 

 air is above 32. 



I have seen only two original descriptions 

 of this process, both of which are contained in 

 the Philosophical Transactions ; the first, by Sir 

 Robert Barker, in the 65th volume ; the other 

 in the 83rd, by Mr. Williams. 



According to the method followed by Sir R. 

 Barker's ice-maker, square excavations, 2 feet 

 deep, and 30 wide, having been formed in a 

 large open plain, their bottoms are covered with 

 sugar-cane, or stems of Indian corn, dried, to 

 the thickness of 8 inches or 1 foot. On this 

 layer, are afterwards placed, in rows, near to 

 each other, small, unglazed earthen pans, i ,of 

 an inch thick, and 1 inch and -^ deep, filled 

 with boiled soft water. The pans are sufficiently 

 porous to allow their outer surface to appear 

 moist, after water has been poured into them. 

 Sir R. Barker adds ; that the nights, the most 

 favourable for the production of ice, are those, 



