223 Dr t'yfe on the Illuminating mid Heating Power of 



perfectly consumed ; the supply of air must also bo properly 

 regulated, otherwise the gas does not undergo that degree of 

 decomposition which is necessary for the separation of the car- 

 bon from the hydrogen, by the combustion of the former of 

 which white light is produced. A question therefore natu- 

 rally arose at the outset of the investigation on the heating 

 power. Is this subject to the same law \ or, when the gas is 

 completely consumed, is the heat evolved the same, whether 

 there is or is not previous decomposition ? I conceived it ne- 

 cessary to establish this first, because if it were found that the 

 heating power is not subject to the same law, the subsequent 

 steps of the investigation would be much facilitated. 



In the experiments on the heating power, the gas was mea- 

 sured by passing it through the same meter that was used for 

 the illuminating power. To ascertain the amount of heat 

 evolved, the method which most naturally occurred, was merely 

 to use the same quantity of water in the same vessel, always 

 at the same temperature, and to find the degree to which it 

 would be raised by the consumpt of equal quantities of gas. 

 This would not, of course, give the actual amount of heat, but 

 as the abject in view was merely to ascertain the comparative 

 heating power, it was considered sufficiently correct. The ap- 

 paratus which I used was merely a boiler, B, with its sides 

 extended farther down than the bot- 

 tom, and hollow, so as to form a ca- 

 vity A, into which the flame was 

 placed ; CO is a tube to carry off the 

 products of combustion. This was 

 preferred to a common pot or flask, 

 because the flame being in a great 

 measure surrounded by water, the 

 heat that would otherwise be lost by 

 radiation was absorbed ; besides, the 

 heated products were transmitted by 

 the tube through the water, so as 



to give off the greater part of their heat. In using this appa- 

 ratus, when the flame of the gas is placed too far up in the 

 cavity, owing to the products of combustion not being carried 

 off quickly, there is frequently a good deal of smoke ; it is nc- 



