212 On the Determination of the [Sept* 



body will become gradually hotter and hotter, till it reaches a 

 temperature when it loses as much heat as it receives. At this 

 point its temperature will become stationary, if that of the air 

 does not vary. 



On the otner side, it is a principle generally admitted, and the 

 justice of which cannot be questioned when confined to small 

 differences of temperature, that the quantity of heat lost every 

 instant by a hot body suspended in the air is proportional to the 

 excess of its temperature above that of the surrounding air. 



It is evident, from these two principles, that if we subject a 

 'body, A, to the action of different uniform sources of heat, the 

 ratio of their intensity will be equal to that of the excess of the 

 temperature of the body, A, when it becomes stationary, above 

 that of the ambient air; since the body, A, arrived at this maxi- 

 mum, receives at each instant the heat which it loses. 



Now let us conceive a thin copper cylinder, A B, (Fig. 1, 

 Plate X.) 6 inches long, and S in diameter, filled with distilled 

 water, and traversed by a serpentine of about 5 feet in length, 

 forming 8 spiral turnings, the two ends of which open without 

 the vessel, the one at the top, the other at the bottom. If we 

 make a regular current of gas traverse this serpentine, main- 

 tained before its entrance at an elevated and constant tempera- 

 ture, this current may be , considered as an uniform source '.of 

 heat, and the cylinder, A B, as the body, A. Of course, if 

 we repeat the same experiment upon each Of the gases, each 

 current will raise the temperature of the cylinder, A B, to a 

 fixed point, where it will remain stationary ; and it follows from 

 the principles announced above, that, reckoning from this point, 

 the excess of the temperature of the cylinder, A B, above that 

 of the ambient air, will be proportional to the quantity of heat 



fiven out by the current of gas that passed through the cylinder. 

 lence we shall obtain by this method, with great exactness, the 

 relative specific heats of the gases subjected to this kind of expe- 

 riment. There are likewise, two methods of comparing them 

 with water. 



The first consists in subjecting the cylinder, A B, which we 

 shall afterwards call the calorimeter, to the action of a current of 

 water, perfectly regular, and so slow that it will hardly produce 

 a greater effect than the current of the different gases. 



The second method consists in determining by calculation the 

 real auantity of heat v, hich the calorimeter, come to its stationary 

 temperature, can lose in a given time; for since after it reaches 

 this point it does not become hotter, though the source of heat 

 continues to be applied to it, it is evident that it loses as much 

 heat as it receives. We shall employ, in the sequel, thete two 

 methods in succession. 



Tt is obvious that it would have been exceedingly tedious to 



