£54 



Mr. A. Schuster. 



stein found, in some of his experiments, that when an electric current 

 passing through a gas is forced through a narrow opening many of 

 the phenomena seen near the kathode appear on that side of the 

 opening which faces the positive pole. It is also known that, if a 

 current passes through a funnel-shaped opening, the fall of potential 

 required is greater in one direction than in the other. Finally, if the 

 current is discontinuous, and a point on the outside of a glass tube is 

 connected to earth, certain phenomena are seen which have been spe- 

 cially investigated by Messrs. Spottiswoode and Moulton. All these 

 facts I believe to be capable of explanation if we remember that, when- 

 ever a current passes between solids of different conductivities, a 

 certain surface electrification is necessary to satisfy the conditions of 

 continuity. Gases do not follow Ohm's law, and there will in all 

 probability be an electrification whenever the cross-section alters. 

 The different behaviour of positive and negative electrification will 

 come into play, and this, together with the different rate of diffusion 

 of different ions, will, I believe, be found sufficient to explain the 

 phenomena. 



The effect of ultra-violet light on a negatively electrified body is 

 probably due, as has been pointed out, to a chemical action, but we 

 have further to assume that this action is not set up on a positively 

 charged body. If this view is correct, we shall have to take the law 

 of impacts between the gas and the metal to be modified in such a 

 way that a chemical effect only takes place when the metal is 

 charged negatively. 



Stratifications. 



It is generally considered that the most important test of any 

 theory of the discharge is to be found in the way in which it can 

 explain stratification. Very little is known about the circumstances 

 which produce stratifications, and they show by their lawless be- 

 haviour that they are rather to be considered as irregularities in the 

 discharge than as matters of primary importance. 



According to our view, the regular diffusion of ions in the positive 

 part of the discharge can only be maintained by a balance of very 

 delicately-adjusted phenomena. The two kinds of ions diffuse with 

 different velocities ; they will tend to recombine together, and will 

 occasionally do so. If so, and if the current does not cease to be 

 steady, we must have as many fresh dissociations as combinations in 

 each part of the tube. It does not seem impossible that there may 

 be several stable ways in which the current may pass. It is possible 

 that, besides the discharge which passes as I have just explained, 

 there may be another in which the tube is divided into a succession 

 of parts in which the decompositions alternately outnumber the 



