Loss of Charge by Evaporation. 



491 



present in the vessel ; and therefore in the experiments now 

 to he described, a method was adopted which should not be 

 open to this objection. 



The method of experiment and the apparatus employed 

 can be most easily described by reference to the annexed 

 diagram. A cylindrical metal box, A, had a platform of 

 metal, B, inside it, insulated from the bottom of the box by 

 ebonite plugs. The lid of the box had a similar platform, a disk 

 of metal, C, underneath it, similarly insulated from the lid. 

 A wire passed from B through a hole in the side of the box 

 to the insulated quadrants of an electrometer, and had that 



Fig. 1. 



portion of its length which was outside the box surrounded 

 by a metal guard-tube D. The vessel, V, containing the 

 •water was of lead, 8 cms. in diameter and 1-5 cms. in depth, 

 and was placed on the platform B. The distance between C 

 and the surface of the water was 2 cms. The disk C was 

 connected by a wire passing through a hole in the lid to one 

 terminal of a battery of storage-cells, while the other ter- 

 minal of the battery was put to earth by connecting it to a 

 gas-pipe. The uninsulated quadrants and sheath of the 

 electrometer, the box A, and the guard-tube D, were con- 

 nected by a wire with one another and with the gas-pipe. 



The experiments are performed as follows : — The water- 

 vessel resting on B is put to earth by joining the two pairs of 

 quadrants. Thus the box A and the water- vessel are at the 

 same potential. Next C is connected to the positive terminal of 

 the storage-cells (a sufficient number being used to give the 

 desired potential), and the negative terminal put to earth. 

 Then the connexion between the two pairs of quadrants is 

 broken, so that the water now has a negative charoe ; but it 

 is at the same potential as the box from which it is insulated 

 so that there should be no tendency to leak down the ebonite 

 supports. If the water in V, however, loses some of its 

 charge by evaporation, this equality of potential will no 



