200 MR HAYCRAFT on the Specific Heat of the Gases. 



By a slight consideration of this description of the apparatus, 

 which may be deemed rather prolix, and by an inspection of the 

 plate, it will be perceived that two Gases, contained in the two 

 parts of the machine, will be under circumstances precisely similar ; 

 the quantities of Gas transmitted through the calorimeters in a 

 given time will be the same ; the temperature of the surround- 

 ing media and the barometrical pressure will be equal ; the tem- 

 perature also of the Gases themselves must be the same, because 

 they passed through the same heating medium. In fine, the 

 size of the tubes, cylinders, calorimeters and valves, were the same 

 in the two parts of the machine. 



Therefore, the temperature communicated by the two Gases 

 submitted to a comparative trial, will be the direct ratio of their 

 comparative capacities for caloric, provided there be no dispro- 

 portionate escape by absorption in the calorimeters, arising from 

 the different temperatures of surrounding bodies. 



This source of fallacy was obviated by the arrangement of 

 Count RUMFORD, who contrived that the temperature of the sur- 

 rounding medium should be as much above that of the calorime- 

 ter at the beginning, as it was lower at the end of the experi- 

 ments. 



The quantity of Gas propelled through the calorimeter was 

 12 cubic inches during the action of the piston. Those actions, as 

 regulated by a second pendulum, which was suspended in the 

 apartment, being 120 every minute, the whole quantity would be 

 1440 cubic inches of Gas propelled through the calorimeter every 

 minute. There was no occasion, however, to take these quanti- 

 ties into the account, because they were precisely the same of 

 each Gas subjected to trial. 



My thermometers were adjusted by Mr AD IE of Edinburgh. 

 Each degree was divided into 5 parts, which were sufficiently 

 large to be divided by the eye into 4 parts ; so that the tempera- 

 ture could be ascertained to a 20th part of a degree, making al- 



