304 



PROF. W. A. BONE AND OTHERS ON 



The mean results of the foregoing experiments, so far as the distribution of oxygen 

 between methane and hydrogen is concerned, are summarised as follows : 



TABLE IX. Mixtures CH 4 +0 



It is at once evident from the results with the mixture CH 4 + 2 + 2H 2 that the 

 affinity of methane is at least twenty to thirty times greater than that of hydrogen 

 for o.i'i/gen in explosions. The actual distribution of oxygen when a particular 

 mixture i,s exploded is undoubtedly influenced to some extent b}'' the walls of the con- 

 taining.vessel but not by the absolute initial pressure. The influence of the containing 

 walls would presumably disappear after a certain -limiting ratio of area/volume 

 is attained, and, had the resources at our disposal permitted, it would have been 

 interesting to have made further experiments with a still larger bomb than B. 

 An examination of the summarised results in Table IX. leads to the important 

 conclusion that the influence of successive increases in x, the volume ratio of H 2 to 

 CH i in the mixture exploded, upon the actual oxygen distribution is for a given vessel 

 proprotional to ,r 3 . This can hardly mean other than that in explosion flames 

 hydrogen is directly burnt to steam as the result of trimolecular impacts, 



2 = 2H 2 0. 



PART IV.- EXPERIMENTS ON THE KELATIVE AFFINITIES OF METHANE AND 

 CARBON MONOXIDE FOR OXYGEN IN EXPLOSIONS. 



(With Messrs. HAMILTON DAVIES and H. H. GRAY.) 



The success of the experiments described in the preceding section led us to make 

 a similar attempt to determine the division of oxygen between methane and carbon 

 monoxide when mixtures of the general composition CH 4 + O 2 +o;CO are exploded 

 under pressure. In this we were not completely successful, owing to a slight 

 separation of carbon in the explosions whenever x exceeded 3'0 or thereabouts. Pro- 

 vided, however, that x did not exceed this limit, no carbon was deposited during the 

 explosion, and such experiments may be given in detail as quite reliable so far as the 

 question of the oxygen distribution between the two inflammable constituents is 

 concerned. 



