8o CORRELATION OF PHYSICAL FORCES. 



film of oxide, the oxide will be removed in concentric spaces, 

 and increased in the alternate ones, showing a lateral alterna- 

 tion of positive and negative electricity, or electricity of 

 opposite character in the same discharge. 



It would be hasty to assert that in no case can the elec- 

 trical disruptive discharge take place without the terminals 

 being affected. I have, however, seen no instance of such a 

 result where the discharge has been sufficiently prolonged, 

 and the terminals in such a state as could be expected to 

 render manifest slight changes. 



The next question which would occur in following out the 

 enquiry which has been indicated would probably be : What 

 is the action upon the gas itself ? Is this changed in any 

 manner ? 



In answer to this, it must be admitted that, in the present 

 state of experimental knowledge on this subject, certain gases 

 only, appear to leave permanent traces of their having been 

 changed by the discharge, while others, if affected by it 

 which, as will be presently seen, there are reasons to believe 

 they are return to their normal state immediately after the 

 discharge. 



In the former class we may place many compound gases, 

 as ammonia, olefiant gas, protoxide of nitrogen, deutoxide of 

 nitrogen, and others, which are decomposed by the action of 

 the discharge. Mixed gases are also chemically combined by 

 it: for instance, oxygen and hydrogen unite and form water; 

 common air gives nitric acid ; chlorine and aqueous vapour give 

 oxygen, the chlorine uniting with the hydrogen of the water. 



But, farther than this, in the case of certain elementary 

 gases a permanent change is effected by the electrical dis- 

 charge. Thus, oxygen submitted to the discharge is partially 

 changed into the substance called ozone, a substance now 

 considered to be an allotropic condition of oxygen ; and there 

 is reason to believe that, when the change takes place, there 

 is a definite polar condition of the gas, and that definite por- 

 tions of it are affected that in a certain sense one portion 

 of the oxygen bears temporarily to the other the relation 

 which hydrogen ordinarily does to oxygen. 



If the discharge be passed through the vapour of phos- 



