Behaviour of Solid Insulators, 183 



Kohlrausch's view of the behaviour of insulators is accordingly 

 completely untenable ; and the question therefore arises, what hy- 

 pothesis can be set up in its stead ? 



It is well known that the theoretical investigations of Kirch- 

 hoff and the experimental researches of Kohlrausch have led to 

 the following mode of regarding the movement of electricity in 

 conductors. As soon as a current is established, no more free 

 electricity exists in the interior of the conductor, but only at its 

 surface, and at the points of contact of heterogeneous metals. 

 These free electricities act at a distance upon the interior of the 

 conductor so as to cause continual decompositions and recombi- 

 nations of the electricities in each smallest particle, and thus 

 produce a movement of one electricity in one direction, and of 

 the other in the opposite direction. 



Now can this conception be transferred to insulators also? 

 can they be regarded simply as bad conductors, which differ from 

 good ones only by the forces which are required to produce the 

 separation of equal quantities of electricity being extraordinarily 

 much greater ? 



In other words, Is it the action of the electricities distributed 

 upon the coatings which occasions decompositions in the smallest 

 particles, consequently setting the two electricities in motion 

 towards the two sides, and thus giving rise to the phenomena of 

 the residual charge ? 



A theoretical investigation shows that in this case also the 

 influence of very thin intervening layers must be imperceptibly 

 small, as it is in Kohlrausclr's view, and that differences of thick- 

 ness would be equally without effect. It is moreover easy, upon 

 this hypothesis, to determine the form of the curve which repre- 

 sents the available charge (potential) as a function of the time ; 

 and this does not agree with that actually observed. We are 

 thus brought to conclude that it is certainly not merely the action 

 at a distance of the electricities on the coatings which brings 

 about the movements of the electricities in the interior of insu- 

 lators. But, on the other hand, it can also be proved by experi- 

 ment that this action at a distance is not completely excluded. 

 Thus if an uncoated glass plate is introduced between the plates 

 of an air condenser, so as to be separated from them by spaces 

 filled with air sufficiently great (according to special experiments) 

 to make the passage of electricity between the plates impossible, 

 we still find, on the one hand, after charging the air condenser, 

 a greater diminution of the charge than can be explained by the 

 mere loss of electricity into the air, and, on the other hand, after 

 discharging the condenser, a reappearance of a residual charge. 



Another essential difference between insulators and conductors 



