206 SIR ANDEEW NOBLE: RESEARCHES ON EXPLOSIVES. 



volume of gas taken for analysis, in order to obtain the volume of uncontaminated 

 gas in the sample, and hence the correct percentages of the various constituents. 



Carbon monoxide is next removed by prolonged treatment with two successive 

 portions of acid cuprous chloride solution. After absorption in the first cuprous 

 chloride pipette the gas is directly transferred to a second pipette containing a 

 solution which has not previously absorbed more than a trace of carbon monoxide, 

 this transference being accomplished in practically the same manner as the return of 

 the gas to the measuring tube, which takes place after transference to a pipette 

 containing a little water, which removes the traces of hydrochloric acid derived from 

 the cuprous chloride solution. An excess of oxygen* is then added, and, after 

 measuring, the mixture is transferred to the explosion pipette, where it is exploded 

 by means of an electric spark after expanding to such a volume as to prevent any 

 marked oxidation of the nitrogen, whilst ensuring the complete combustion of the 

 methane and hydrogen. The residue is next measured in order to ascertain the 

 reduction of volume resulting from the explosion, and the carbon dioxide, produced 

 by the combustion of the methane, is determined by absorption with potash. The 

 volume of the carbon dioxide produced is equal to that of the methane originally 

 present. The contraction due to the combustion of the methane, or in other words, 

 twice the volume of the carbon dioxide, is deducted from the total contraction 

 resulting from the explosion, and two-thirds of the corrected contraction so obtained 

 is equal to the volume of hydrogen. 



Finally, the excess of oxygen remaining after explosion is determined by means of 

 alkaline pyrogallol as a check upon the amounts of hydrogen and methane calculated 

 as above. The nitrogen is estimated by difference. 



The above represents the routine determination of carbon dioxide, carbon monoxide, 

 hydrogen, methane and nitrogen, as usually carried out, but additional tests have 

 also been employed in order to ascertain whether certain other bodies were present in 

 measurable quantities, but with negative results. Thus some of the gas samples 

 were examined for unsaturated hydro-carbons (ethylene, &c.) immediately after the 

 removal of the carbon dioxide, by shaking the gas with fuming sulphuric acid,t and 

 removing acid fumes in the potash pipette before again measuring. No change of 

 volume was ordinarily observed, and in no case did the change exceed O'l per cent., hence 

 the samples did not contain any appreciable quantity of unsaturated hydrocarbons. 



The ordinary determinations of contraction resulting from explosion, carbon dioxide 



* The oxygen is prepared by the electrolysis of dilute sulphuric acid in a Hof mann voltameter and freed 

 from traces of hydrogen by treatment in a Winkler combustion pipette. A supply is stored over mercury 

 in one of the ordinary absorption pipettes ready for use. 



t Fuming sulphuric acid was used in one of the ordinary absorption pipettes, provided with a guard 

 tube containing sulphuric acid, in order to prevent moisture from gaining access to the upper bulb D. 

 Of course no mercury was employed in the absorption bulb, and that in the capillaries was driven into the 

 bulb F when sending the gas into the pipette. 



