Combination of Bodies 'with Oxygen. 323 



the liquid, was again introduced, and the increment of heat 

 observed. 



The duration of the experiment was so short, that scarcely 

 any correction for the cooling and heating influence of the air 

 was required. The temperature of the air was in general a 

 little above the mean between the initial and final temperatures 

 of the apparatus : the heat, however, was given out so rapidly, 

 that the latter must have been nearly at the final temperature 

 during the greater part of the time. After each experiment, the 

 apparatus was again rotated for a period of thirty-five seconds, 

 and the loss of heat from cooling observed. 1 have assumed 

 one-half of this loss to be the required correction, except ni 

 the case of olefiant gas, when the initial temperature was a 

 little lower than usual. The correction so applied, it wdl be 

 seen, never exceeded 0^*005 C. 



The thermal values of the different parts of the apparatus 

 in terms of water were as follows : — 



Copper 170 grms. x 0-095 . . 16-15 

 Brass 111 ... x 094- . . 10-43 

 Solder 15 ... x 0-043 . . 0-64 

 Leather, cork, &c 0'^8 



Thermal value 27-70 



The amount of water was always determined by weighing 

 the apparatus with its contents after each experiment, and 

 deducting the weight of the same when dry. 



Hydrogen and Oxygen. 



The hydrogen gas was purified, according to the method 

 of M. Dumas, by passing it through a series of tubes in which 

 it was successively exposed to solutions of the acetate of lead, 

 sulphate of silver, and hydrate of potash. It was afterwards 

 collected over water in a graduated vessel. In this way it 

 became contaminated with a small quantity of atmospheric 

 air, the amount of which it was necessary to ascertain with 

 precision. This was effected by an independent experiment, 

 in which the gas was collected in exactly the same manner. 

 In the case of other gases, the true volume was inferred from 

 the diminution which occurred after the explosion. The dif- 

 ficulty of obtaining accurate results in experiments upon gases 

 collected over water (which for obvious reasons could not be 

 avoided in this inquiry) is so well-known to chemists, that I 

 deem it unnecessary to dwell upon this point. I have endea- 

 voured in every case to determine by experiment, and to apply 

 the necessary corrections for absorption, &c., but at the same 



Y2 



