concerning the Gases of the Atmosphere, 101 



■contains about ^-^ of hydrogen in addition to variable 

 amounts of hydrocarbons. It appeared to me that a spectro- 

 scopic confirmation would be interesting. 



Using the old apparatus, in which the tubes conveying 

 the gas and the electrodes were fitted into a rubber cork, I 

 could not succeed in getting quit of C from the spectrum of 

 somewhat powerful sparks, however carefully the air were 

 dried. The coil was excited with five Grove cells and a large 

 leyden-jar was connected with the secondary in the usual 

 way. This observation was of course consistent with the 

 presence of hydrogen in the atmosphere ; but it was 

 suspicious that the best approach to evanescence was obtained 

 with a somewhat brisk rather than with a slow current of air, 

 indicating that the source of the hydrogen was in the 

 apparatus rather than in the atmosphere. As it seemed 

 desirable to apply heat, I discarded the old apparatus, sub- 

 stituting for it a simpler one consisting merely of a small 

 bulbous enlargement of the gas-leading tube. Into this the 

 platinum electrodes were sealed. The gases under examina- 

 tion were stored under slight pressure and on leaving the 

 reservoirs were partially dried with sulphuric acid. A three- 

 way tap allowed the easy substitution of one gas for another. 

 After passing this tap the gas was further dried by phosphoric 

 anhydride on its way to the sparking-bulb. 



The application of heat to the bulb and to the short length 

 of tubing between the bulb and the phosphoric anhydride led, 

 as was expected, to a recrudescence of C. Subsequently 

 there seemed to be an improvement. Not only was C less 

 conspicuous, but its visibility remained about the same 

 although the rate of flow were varied. It is difficult to 

 describe in words the effect upon the eye, but I may say that 

 with the actual spectroscopic arrangements including a some- 

 what wide slit the line could be certainly and steadily seen. 



The above was the appearance with a stream of (country) 

 :air. When air to which 5( ^ part of hydrogen (by volume) 

 had been added was substituted, the visibility of C was 

 markedly increased ; and the difference w r as such that one 

 could easily believe teat the proportion of hydrogen actually 

 operative had been doubled. This conclusion would be in 

 precise agreement with M. Gautier, could we assume that 

 the smaller quantity of hydrogen really accompanied the air. 

 But the facts now to be recorded render this assumption 

 extremely doubtful. 



In the first place the visibility of C with ordinary air was 

 not perceptibly diminished by passage of the air over red hot 

 >cupric oxide included between the sulphuric acid and the 



