u 



Rev, Mr. PowelTs Remarh on some of [July, 



through very thin glass. This conclusion is in the first instance 

 deduced from a very simple and elegant experiment, which 

 consists in placing a hot body midway between the thin bulbS 

 of a large differential thermometer; the hquid becomes station- 

 ary; the exterior half of one bulb is then blackened; and the 

 liqu id on that side is found to fall the instant the hot body is 

 placed as before. Hence the author infers, that by rendering 

 that part of the glass opaquie, a portion of the heat is stopped ; 

 which therefore radiates through the transparent part of the bulb 

 in this case, and through both parts when the whole is transpa- 

 rent. With a view to examining the validity of this conclusion, 

 I conceived it desirable, in the first place, to repeat the experi- 

 ment, and verify the result. This I did with a differential ther- 

 mometer, having its bulbs about an inch in diameter. The half 

 of one bulb away from the source of heat was coated in some 

 instances with a cap of thin paper previously moulded to fit it 

 exactly, by which a facility of removing and replacing the coat- 

 ing was obtained ; in others with indian ink, or the smoke of a 

 candle. 



A hot iron ball was placed at an ec^ual distance from each 

 bulb; and before experiment, the liquid was observed to 

 remain stationary. 



On the bulb being pldced in its position, the action was inva- 

 riably greatest on the coated bulb. When the coating Was 

 removed (care being taken to keep the whole exactly in the 

 same position), and to heat the ball exactly to the same poitit ; 

 viz. having made it red-hot, to let it cool down precisely till it 

 ceased to be visibly luminous in the same dark place, the action 

 was greatest on the plain bulb. 



TEe results in one experiment were as follow ; the divisions 

 on the scale are arbitrary ; the graduation yrom the plain bulb. 



Here we perceive, with the coating, a considerable action on 



