224 Sir J, J. Thomson : Further 



quantity of hydrogen is small, the line is visible, but unless 

 the oxygen is greatly in excess it is invariably absent. I 

 ascribe this line to the formation of H 2 2 . The question 

 whether the actual carrier of the negative electricity is the 

 molecule H 2 2 , or H0 2 , or even 2 , is difficult to determine, 

 as the lines are not strong enough to allow of the very finest 

 tubes through the cathode being used. The following are 

 the values of m/e obtained for this line on different plates, 

 the value of m/e for the hydrogen atom being taken as 

 unity :— 33-5, 35, 33, 35, 32, 33, 33*5, 34, 35, 33, 32, 31 ; 

 mean 33*3. This suggests that the carrier of the negative 

 charge is not 2 but either H0 2 or H 2 2 . The stronger 

 negative line, with a value of m/e near 16, is, I am inclined 

 to think, when hydrogen is present, not but OH, as the 

 measurements of this line on plates, taken with mixtures of 

 oxygen and hydrogen, uniformly gave values of m/e slightly 

 greater than 16. If the hydroxyl radicle is the carrier for 

 this line, we can understand why the line is especially bright 

 when the discharge passes through the vapours of some 

 organic compounds, such as methyl alcohol, when the hydro- 

 xyl radicle is very likely to be liberated in the process of 

 ionization. Though the line 33 only appears in mixtures of 

 hydrogen and oxygen when the oxygen is greatly in excess ; 

 it appears in mixtures of ethylene and oxygen even when 

 only a small quantity of oxygen is present. 



The negative lines which appear when the discharge passes 

 through the vapours of hydrocarbons are interesting. 

 When there are no bonds between two atoms of carbon, as 

 for instance in marsh gas, carbon monoxide, or carbon 

 dioxide, the only line on the negative side which can be 

 ascribed to carbon is the line for which m/e is 12, which is in 

 general very bright, When, however, we take compounds in 

 which there are bonds between the carbon atoms, such as 

 ethylene and acetylene, we frequently find that in addition 

 to the line for which m/e = 12 there is another for which 

 m/e is about 24, showing that we have a system with a 

 strong affinity for negative electricity formed from two 

 carbon atoms, with perhaps one or two atoms of hydrogen. 

 When the discharge passes through benzene vapour we get 

 in addition to the lines 12 and 24 a line near to 36 ; and I 

 have sometimes thought that I could detect one near 48, 

 though this is exceedingly faint : thus we can get strongly 

 electronegative radicles containing three or more atoms of 

 carbon. In very carefully purified helium the negative lines 

 due to impurities are exceedingly faint, much fainter than 



