212 MR HAYCRAFT on the Specific Heat of the Gases. 



that mercury contained in the hollow ball would rise into the 

 stem upon any increase of the expansive force of the air contained 

 in the ball, I filled the latter with air at a temperature of 60 ; af- 

 ter which the ball was immersed into boiling water. In a short 

 time the mercury rose into the stem to the height of 7 inches. 

 The experiment was repeated, excepting that a few drops of wa- 

 ter were put, together with the air, into the ball. The mercury, 

 after the immersion of the ball in boiling water, rose to 21 inches. 

 Afterwards, on passing a quantity of water into the ball, the 

 mercury, after its immersion, rose to 28 inches. Some months 

 afterwards, on repeating the experiment, the mercury rose in one 

 instance to 14 inches. Thus we have a series of expansive forces 

 of air united to watery vapour of 7, 14, 21 and 28 inches : it was 

 upon this principle that I contrived an air thermometer. The 

 form of it is similar to that of the differential thermometer invent- 

 ed by Professor LESLIE. One ball contained atmospherical air 

 dried by means of muriate of lime ; the other contained air in its 

 usual state. Interposed between the balls was a column of the 

 volatile spirit of turpentine. Upon any rise of the temperature 

 of the atmosphere, the column immediately rose at the side 

 of the dry ball. After some time, however, the instrument 

 seemed to have lost its power; and after a still longer pe- 

 riod, the ball containing dry air had the greater expansive force. 

 This I accounted for by supposing, that the vapour of turpen- 

 tine had in process of time combined with the dry air, and had 

 given it its greater expansive power. This thermometer is now 

 a remarkably delicate one, though its degrees are of very une- 

 qual length, and appear to vary by lapse of time. Probably hy- 

 drogen gas contained in two platina balls, in ope of which a 

 h'ttle mercury might be placed, connected together in the same 

 way, would make an accurate pyrometer, indicating tempera- 

 tures as high as the melting point of platina. 



