564 



Drs. De La Rue and H. W. Miiller. [June 17, 



experiments we are about to describe, and which we select from a 

 long series, will show that there is really no such polarisation of the 

 terminals. 



We will first take the case of the voltameter ; the arrangement of 

 the apparatus employed in our experiments is shown in fig. 1, where 

 K K' represent a special key for breaking connexion between any piece 

 of apparatus and the battery, B, and connecting it afterwards instan- 

 taneously with a galvanometer, Gr. The terminals of the battery are 

 attached respectively to the screw-clamps, c and d, carried by an ebonite 

 bar, supported on two ebonite columns, touch-points in metallic contact 

 with c and d, pass through this bar, so that when the brass springs, 

 ae and bf, are allowed to rise and press against these touch-points, 

 the battery is in metallic connexion with the screw-clamps, a and b, 

 and any piece of apparatus attached to them, as, for example, the volta- 

 meter, V. The voltameter employed has a resistance of 20 ohms ; its 

 platinum electrodes are 2'5 inches long and O025 inch in diameter, 

 and are separated 0"25 inch ; it is charged with dilute pure sulphuric 

 acid in the proportion of 1 volume of acid, sp. gr. 1*84, to 8 volumes 

 of water. When the springs are suddenly pressed down by the fingers 

 placed on the ebonite disks, e and/, the connexion is broken between 

 the voltameter and the battery, and it is connected instantaneously 

 with the galvanometer, Gr. The galvanometer actually employed was 

 not that shown in the figure, but a Thomson galvanometer, whose 



constant, C,= — - — =1,874 scale divisions. On connecting a 

 1 megohm 



battery of 10 chloride of silver cells with the voltameter for a minute 

 or less, and then suddenly pressing down the springs e and /, there 

 was a deflection to the left, say, of more than 1,000 scale divisions, 

 although the shunt was used to reduce the current through the 

 galvanometer ; therefore, the deflection without the shunt would have 

 been more than 1,000,000 divisions. By comparing this deflection 

 with that produced by a half microfarad condenser, charged with 

 240 cells, it was ascertained that the deviation produced by the volta- 

 meter was equivalent to that of 111 microfarads. 



The small condenser shown at C, fig. 1, was substituted for the 

 voltameter ; it is made of a thin plate of glass 2 inches square and 

 00115 inch thick ; the tinfoil coatings being 1*5 inch square, its 

 capacity was found to be 0*00055 m.f. When charged with 3,600 

 cells, and afterwards connected with the Thomson galvanometer 

 through the £ shunt, by pressing down e and /, the deflection was 

 136'5 divisions to the left; this multiplied by 9*92, the value of the 

 shunt was equivalent to 1,354 divisions. 



The apparatus, shown in fig. 2, which was constructed for another 

 object already described,* was connected with a and b ; it consists 

 * " Phil. Trans.," vol. clxxi, p. 76. 



