160 



On the Ioniscdion of Atmospheric Air. 



being embedded in the sulphur. The electroscope formerly used to 

 test the constancy of the potential of the supporting rod was dis- 

 pensed with; all need for external wires was thus removed. Only 

 the end of the wire by which the charge was put into the condenser 

 protruded from the sulphur, and this was covered as shown in the 

 figure, except at the moment of charging, by a small bottle containing 

 calcium chloride ■ this fitted tightly on a conical projection of the 



Fig. 2. 



sulphur, through the centre of which the wire passed. The suffi- 

 cient constancy of potential of the supporting rod under these con- 

 ditions was shown by the fact that when it had been put, by means 

 of the magnet, in momentary electrical connection with the leaking 

 system, a second contact, made twenty-four hours later, caused the 

 gold leaf, which indicated the potential, to return to within two micro- 

 meter scale divisions of its position immediately after the first con- 

 tact. The change in the potential of the leaking system produced 



