ON ELECTRIFICATION OF MOLECULES AND CHEMICAL CHANGE. 495 



gate the question whetlier the union of elements and compounds was in 

 any way connected with electrical discharge. The facts of electrolysis 

 point strongly in this direction. I gave up at once the idea of directly 

 investigating the question whether molecules of gases in contact were elec- 

 trically charged, but in an indirect way I have obtained some evidence on 

 the question. Taking a tube divided in the middle by a tap, and filling it 

 with a mixture of dried ammonia and hydrogen chloride gases in equal 

 proportions, I introduced at the two ends platinum plates which were 

 oppositely charged from the terminal of a Wimshurst machine. After 

 half an hour the tap was closed, and the gases in the two parts of the tube 

 were drawn through a solution of litmus. On admitting moisture white 

 fumes of ammonium chloride were produced, but there was found to be 

 residual gas in both halves of the tube. The gas from the part which had 

 contained the negatively charged plate turned litmus blue, and the gas from 

 the other part reddened the litmus. The experiment was often repeated, 

 and the result was the same. The separation was never very great, but the 

 evidence of some separation seemed conclusive. With other gases there was 

 also evidence of separation. With air dried by sulphuric acid there was 

 found to be 1*8 per cent, more oxygen in the part containing the positive 

 plate than there was in that containing the negative plate. Using a mixture 

 of hydrogen and oxygen dried by phosphorus pentoxide, analysis showed an 

 excess of 2-3 per cent, of oxygen in the part of the tube containing the 

 positively charged plate. 



It is possible, therefoi'e, that the molecules of gases which may under 

 certain circumstances combine together, may have an electrical charge. I 

 hope to extend these observations to the case of other gases, and to find 

 out, if possible, whether the charge on the molecules, if it exists, is theirs in- 

 trinsically, or if it only exists when molecules of a different nature are 

 in contact with each other. 



With the object of finding out if electric discharge bears any analogy 

 to the process of chemical combination, I undertook some experiments a 

 year ago to see if electric discharge in air was affected by drying the air 

 as completely as possible. It was found that sparks from a Ruhmkorff's 

 coil, when the discharge was very feeble, would leap across a space of 

 moist air, but that none would pass in the dried air though the sparking 

 distance was very much less. If, however, a strong discharge was used, 

 sparks were obtained in the dried air, and, more than this, the feeble 

 discharge would then easily pass in the dried air. These results entirely 

 agree with those published by Professor J. J. Thomson.' 



I am inclined to think, however, that there may be another interpreta- 

 tion of the latter part of the phenomenon besides that offered by Professor 

 Thomson. It may be that the strong discharge splits the molecules into 

 atoms, and that these can then carry on the feeble discharge. If this is 

 proved to be true, it will serve as yet another point of analogy between 

 chemical combination and electric discharge, foi-, as I have tried to show 

 above, chemical combination does take place between atoms, whatever the 

 state of dryness of their environment. 



With regard to this analogy there may be found some evidence in the 

 study of the action of actinic light on chemical combination and electric 

 discharge. We find that when a mixture of chlorine and hydrogen gases 

 is exposed to light, combination takes place only after a certain interval, 

 and that the interval is not shortened by exposing the gases separately to 

 the action of light. If the interval is spent in breaking up molecules into 



' Phil. Mag., October 1893. 



