

Experiments on Positive Rays. 241 



I shall in what follows confine myself to the consideration 

 of lines which satisfy both the tests for a primary line given 

 on page 217. There are several lines satisfying the first of 

 these tests which cannot be explained by any known sub- 

 stance ; but as I do not feel sure that the possibility of their 

 being secondaries is excluded by this test, I shall not base 

 any conclusions upon them. There is a strong one corre- 

 sponding to an atomic weight 6 in fig. 20 (PL IX.) taken 

 with a mixture of helium and hydrogen. 



Existence of H 3 . — On several plates taken when the dis- 

 charge-tube contains hydrogen, the existence of a primary 

 line for which m/e = 3 has been detected. There can, I 

 think, be little doubt that this line is due to H 3 , and not to 

 the carbon atom with four charges; for if it were due to 

 carbon we should expect to find it conspicuous in carbon 

 compounds, which it is not, and it would be accompanied by 

 the lines due to the carbon atom with 1 and 2 charges, both 

 of which are easily developed. These do not, however, 

 accompany the line under consideration. 



The line mje does not occur, as far as we know, in abso- 

 lutely pure hydrogen, but if a trace of air or oxygen is added 

 to very pure hydrogen, the line appears under certain con- 

 ditions of pressure and current. 



Copies of photographs of plates showing this line are given 

 in figs. 21 a, 21 b, and 21 c (PL IX.) ; the line is the one 

 marked a ; above it are the lines due to the hydrogen atom 

 ^nd molecule. The other lines on the plate are due to the 

 oxygen atom and molecule and to the mercury atom. The 

 existence of this substance is interesting from a chemical 

 point of view, as it is not possible to reconcile its existence 

 with the ordinary conceptions about valency, if hydrogen is 

 regarded as always monovalent. The polymeric modification 

 of hydrogen seems to require special conditions for its 

 formation, for it cannot be detected on many of the plates 

 taken with hydrogen in the tube. 



Formation of N 3 or N 3 H by the Discharge. — When the 

 discharge passes through nitrogen at not too low a pressure, 

 a strong line is found for which mje is about 43, the numbers 

 ranging from 42 to 44. This value of m\e is about the most 

 embarrassing one there is for the application of this method 

 of analysis, for there are many compounds which might 

 occur in the tube which have values of mje equal or 

 nearly equal to 44. C0 2 for example, a common impurity 

 in the tube, has a molecular weight 44 and would give a 



Phil. Mag. S. 6. Yol. 24. No. 140. Aug. 1912. R 



