ELECTRICITY. 73 



handles, each plate is separately electrified, one side of it being 

 positive and the other negative, showing very neatly and de- 

 cisively a polarisation throughout the intervening substances 

 by the effect of induction. 



Indeed, chemical action or electrolysis may, as I have 

 shown, be transmitted by induction across a dielectric sub- 

 stance, such as glass, but apparently only while the glass is 

 being charged with electricity. A wire passing through and 

 hermetically sealed into a glass tube, a short portion only pro- 

 jecting, is made to dip into water contained in a Florence flask ; 

 the flask is immersed in water to an equal depth with that 

 within it ; the wire and another similar wire dipping into the 

 outer water are made to communicate metallically with the 

 powerful electrical machine known as Rhumkorf's coil ; 

 bubbles of gas instantly ascend from the exposed portions 

 of the wires, but cease after a certain time, and are renewed 

 when, after an interval of separation, the coil is again con- 

 nected with the wires. 



The following interesting experiment by Mr. Karsten 

 affords evidence in corroboration of the molecular changes 

 consequent upon electrisation : A coin is placed on a pack of 

 thin plates of glass, and then electrified. On removing the 

 coin and breathing on the glass plate, an impression of the 

 coin is perceptible ; this shows a certain molecular change 

 on the surface of the glass opposed to the plate, or of the 

 vapours condensed on such surface. This effect might and 

 has been interpreted as arising from a film of greasy deposit 

 supposed to exist on the plate ; the impressions, however, have 

 been proved to penetrate to certain depths below the surface, 

 and not to be removed by polishing. 



The following result, however, goes farther : On sepa- 

 rating carefully the glass plates, images of the coin can be de- 

 veloped on each of the surfaces, showing that the molecular 

 change has been transmitted through the substance of the 

 glass ; and we may thence reasonably suppose that a piece of 

 glass, or other dielectric body, if it could be split up while 

 under the influence of electric induction, would exhibit some 

 molecular change at each side of each lamina, however 

 minute the subdivision. 



