242 Sir J. J. Thomson : Further 



line in this region, and one test for the presence of C0 2 , 

 the existence along with it of GO, cannot be applied in 

 the case of the discharge through nitrogen, as N 2 has the 

 same molecular weight and would of course be present. 

 N 2 has again the molecular weight 44, and might be 

 formed if oxygen were present, and N0 2 has a molecular 

 weight 46. We have, therefore, to apply further tests before 

 we can identify this line. If the line were due to C0 2 present 

 as an impurity, it is difficult to see why the presence of 

 nitrogen should make the line so much stronger. Again, if 

 it were due to C0 2 , the oxygen lines would be found, as 

 these are readily developed on both the positive and negative 

 sides when oxygen, free or combined, is in the tube, But 

 on some of the plates, when great care had been taken to get 

 rid of oxygen, the line under consideration was strong while 

 the oxygen lines were too faint to be detected. This is un- 

 favourable to the view that the line is due to C0 2 or N 2 0, 

 and leaves us with N 3 and N 3 H as alternatives. The hydrogen 

 lines were present on all the plates, as, indeed, they always 

 are, but I could not detect any increase in the intensity of 

 the line under consideration when a considerable amount of 

 hydrogen was mixed with the nitrogen. I am inclined, 

 therefore, to regard the line as due to N 3 and not to N 3 H. 

 This line is not nearly so strong as the line due to the nitrogen 

 atom, and, judging from the evidence afforded by these photo- 

 graphs, I should be inclined to attribute the u active form of 

 nitrogen," investigated by Professor Strutt, to atomic nitrogen 

 rather than to N 3 . It is interesting to find that in the dis- 

 charge-tube nitrogen exists in an exceptionally large number 

 of modifications. N++, N + , N" 2+ , N 3+ are found on all plates 

 when the tube contains nitrogen, and on one or two plates I 

 have found a line with m/e = A'7, which I attribute to ^ ++ + . 

 I have never found a line corresponding to nitrogen with a 

 negative charge. 



On some Lines observed when the Discharge passes through 

 mixtures of Hydrogen and Oxygen. — We have already alluded 

 to the negative line in the neighbourhood of m/e = 32 when 

 the discharge passes through oxygen containing a small pro- 

 portion of hydrogen. There is also a line on the positive side 

 with a value of mle about 50 ; its occurrence is not, as in the 

 case of the negative line, limited to those mixtures which 

 contain a small quantity of hydrogen ; it occurs in all 

 mixtures. I have not found it, however, in very pure oxygen. 

 The most natural origin of the line would be ozone. Against 

 this, however, is to be set the fact that it does not occur in« 



