430 



On the Determination of the 



[Dec. 



130*3 i/°. Hence the specific heat of air is to that of water as 

 203-60 to 7-3/ I ; or as 0-2813 to 1 ; a ratio which does not 

 differ much from that which we ohtained by means of a direct 

 current of water. 



It now remains to prove, that if the heat lost by the calori- 

 meter in ten minutes had not been supplied by the hot gas, its 

 temperature would have sunk 2*8793. 



We began by determining, by means of the following experi- 

 ments, how much the temperature of the calorimeter would 

 have sunk, after having reached its maximum, in a given time, 

 if the current of hot gas had been suddenly stopped. For this 

 purpose the temperature of the calorimeter was raised, by means 

 of a lamp, to about 104°. Then leaving every thing in the 

 same situation as when the current of hot gas was circulating, 

 even the tube, F G, (fig. 5,) which served to heat the gas, being 

 kept full of steam, but not passing any current of gas through 

 the apparatus, we left it to itself, and marked the rate of its 

 cooling. By this method we obtained, in two successive experi- 

 ments, the two following series : — 



Now we see by our first table that the current of atmosphe- 

 rical air there indicated raised the temperature of the calorimeter 

 23-321° above the influence of the steam tube employed to heat 

 the gas, or, which comes to the same thing, above the influence 

 of the ambient air, supposing it increased by 4-5° (the eflect of 

 the steam tube). Let us make choice in our first series of obser- 



