408 On the Residual Charge of the Ley den Jar. [Mar. 30, 



easily follows. Addiug the hypothesis, which at first sight appears most 

 probable to connect the induction and polarization with the time by two 

 differential equations, it follows that the potential E of a Leyden jar 



when insulated may be expressed in the form E=(A-j-Be~^)e~~^ 

 where A. and fi are constants for the material, and A and B depend on 

 the previous states of the dielectric. 



2. Observations made with the quadrant electrometer, the condenser 

 being a Florence flask containing sulphuric acid, show that E cannot be 

 so expressed. Glass is a mixture of different silicates, and it may be 

 supposed that each substance is capable of independently being electro- 

 polar ; there will thus be E and more than one polarity to be connected 

 with the time by more than two differential equations. Making a simi- 

 lar obvious hypothesis regarding these relations, E must be expressed 



in the form 2 A,, e r . If this be so, it would probably be possible to 







charge a Leyden jar in such a way that the sign of the return charge 

 after discharge and insulation should change once or more. 



3. This anticipation was verified by charging the flask alternately posi- 

 tively and negatively for successively decreasing periods. The author 

 learned after making this verification that Sir William Thomson had tried 

 similar experiments before, but had never formally published them. 



4. The analogy between coercive force in magnetism and the electro- 

 coercive force suggested that, as mechanical agitation shakes out the 

 magnetism from a magnet, so it might shake down the electropolar state 

 of a dielectric and unmask residual charge more rapidly than is the case 

 in quiescence. This was found to be the case ; a residual charge manifests 

 itself in the flask more rapidly when the flask is tapped than when it i's 

 quiet. It was also found that that portion of the return charge which 

 comes out last is more accelerated by vibration than that which comes 

 out first, and that, after tapping, the flask was less susceptible to the 

 effect of tapping than it was before it was touched. 



5. Experiment shows that, after a return charge has attained a maxi- 

 mum and is decreasing by conduction through the glass, the loss per 

 cent, per unit of time does not continuously increase from zero at the 

 point of maximum potential but may presently decrease. 



6. Sir William Thomson explained specific inductive capacity by sup- 

 posing every part of the dielectric to be electropolar under induction ; 

 by introducing time into that explanation, it is made to cover both specific 

 inductive capacity and that on which residual charge depends as re- 

 spectively rapid and slow cases of similar phenomena. 



