570 GAS, COAL 



its closed end with two platinum wires, fused into the glass, for the transmission of 

 the electric spark. A drop of water, about the size of a pin's head, is introduced 

 into the upper part of the eudiometer before it is filled with mercury and inverted 

 into the mercuriul trough : this small quantity of water serves to saturate with aqueous 

 vapour the gases subsequently introduced. About a cubic inch of the residual ^ras 

 from the last determination is passed into the eudiometer, and its volume accurately 

 read off; about 4 cxibic inches of pure oxygen are now introduced, and the volume 

 (moist) again determined. The oxygon is best prepared at the moment when it is 

 wanted, by heating over a spirit or gas flame a little chlorate of potash, in a very 

 small glass retort, allowing, of course, sufficient time for every trace of atmospheric 

 air to be expelled from the retort before passing the gas into the eudiometer. The 

 open end of the eudiometer must now be pressed firmly upon the thick piece of India- 

 rubber placed at the bottom of the trough, and an electric spark passed through the 

 mixture ; if the above proportions have been observed, the explosion will bo but 

 slight, which is essential if nitrogen bo present in the gas, as this element will other- 

 wise be partially converted into nitric acid, and thus vitiate the results. By using a 

 large excess of oxygen, all danger of the bursting of the eudiometer by the force of 

 the explosion is also avoided. The volume after explosion being again determined, a 

 bullet of caustic potash is introduced into the gas, and allowed to remain so long as 

 any diminution of volume takes place ; this bullet absorbs the carbonic acid that has 

 been produced by the combustion of the light carburetted hydrogen and carbonic 

 oxide, and also renders the residual gas perfectly dry : the volume read off after this 

 absorption, when deducted from the previous reading, gives the volume of carbonic 

 acid generated by the combustion of the gas. 



The residual gas now contains only nitrogen and the excess of oxygen employed. 

 The former is determined by first ascertaining the amount of oxygen present, and 

 then deducting that number from the volume of both gases ; for this purpose a quan- 

 tity of dry hydrogen, at least three times as great as the residual gas, is introduced, 

 and the volume of the mixture determined ; the explosion is then made as before, and 

 the volume (moist) again recorded : one-third of the contraction caused by this ex- 

 plosion represents the volume of oxygen, and this deducted from the volume of residual 

 gas, after absorption of carbonic acid, gives the amount of nitrogen. 



The behaviour of the other three non-luminous gases on explosion with oxygen 

 enables us readily to find their respective amounts by three simple equations, founded 

 upon the quantity of oxygen consumed, and the amount of carbonic acid generated 

 by the three gases in question. Hydrogen consumes half its own volume of oxygen, 

 and generates no carbonic acid ; light carburetted hydrogen consumes twice its 

 volume of oxygen, and generates its own volume of carbonic acid ; whilst carbonic 

 oxide consumes half its volume of oxygon, and generates its own volume of carbonic 

 acid. If, therefore, we represent the volume of the mixed gases by A, the amount of 

 oxygen consumed by B, and the quantity of carbonic acid generated by C, and further, 

 the volumes of hydrogen, light carburetted hydrogen, and carbonic oxide respectively 

 by x, y, and z> we have the following equations : 



y + z = C. 

 From which the following values for x, y, and z t are derived : - 



V. Estimation of the Value of the Luminiferous Constituents. 



Wo have now given methods for ascertaining the respective quantities of all the 

 ingredients contained in any specimen of coal-gas, but the results of the above analy- 

 tical operations afford us no clue to its illuminating power. They give us, it is true, the 

 amount of illuminating hydrocarbons contained in a given volume of the gas, but it 

 will be evident, from what has already been said respecting the luminiferous powers 

 of these hydrocarbons, that the greater the amount of carbon contained in :v given 

 volume, the greater will bo the quantity of light produced on their combustion ; an. I 

 therefore, as the number of volumes of carbon-vapour contained io one volurag of tho 



