Mr. J. J. Waterston on Electric Conduction. 



441 



FlGM. 



Yfi 



only on the outer surface of a conductor when in its normal static 

 condition of charge. Thus let A, B, C, D, fig. 1, be four metallic 

 plates parallel and equal to each other. 

 Let A and D be connected to earth or 

 by return wire, and let B C be joined by / J 



a long metallic cylinder B L L C. Let 

 there be interposed between A and B a 

 thin dielectric plate, and another be- 

 tween C and D. 



Let B and C be equidistant from 

 A and D respectively, and the whole 

 BLLC charged with positive electricity. 

 The greater part of the charge will be 

 located on the sides of B and C that are 

 adjacent to A and D, being inductively 

 retained there as in the magic pane. 

 Let this charged system BLLC be 

 supposed to move closer to A, the sur- 

 face B draws nearer to A, and simulta- 

 neously the surface C withdraws from 

 D (the parallelism being strictly main- 

 tained). The attractive force between 

 B and A now becomes greater than 



between C and D. Also the quantity of electricity engaged 

 between B and A becomes greater than between C and D. 

 Electricity leaves C and travels to B ; and this takes place more 

 and more as B gets closer to A. Thus we have a current of 

 positive electricity travelling from C to B and conducted by 

 means of L L. If the motion of the system is towards D, a 

 positive current travels back from B to C. Thus a current may 

 be made to flow in either direction by mechanical force applied 

 to the system, and it may be made to flow with greater or 

 less speed according as the movement is made more or less 

 promptly. 



Let an electroscope be placed at G : the leaves will show maxi- 

 mum divergence when the system is exactly in the middle between 

 A and I), and will close as C approximates D orB approximates 

 A. We may assume the distance between B and A to be less 

 than any given magnitude, in which case the tension of the 

 electricity becomes indefinitely small (although discharge has 

 been prevented by the interposed dielectric). By applying 

 mechanical force and expending work, we separate the plates 

 and raise the tension of the electricity, which consequently 

 moves along the conductor ; and this may be supposed to be 

 effected with an indefinitely great quickness, so as to represent 

 the discharge of a Leyden jar. Those who affirm that electricity 



Phil. Mag. S. 4. Vol. 30. No. 205. Dec. 1865. 2 G 



