[mclennan-keys] electrical CONDUCTIVITY 41 



a Dowar flask C. A second copper wire B was also suspended in 

 the flask in such a way as to touch neither the wire A, nor the sides 

 or the base of the flask C. The wire B was held in position by care- 

 fully cleaned ebonite supports. It was also connected to the free 

 quadrants of a Dolazaleck electrometer and to one of the armatures 

 of a sulphur condenser of about 2,300 e.s.u. capacity. 



In one experiment with this arrangement the electrode B and its 

 connected system was charged when the flask was empty to one 

 quarter of a volt by means of an auxiliary battery. The steady de- 

 flection on the scale corresponding to this potential was 99-5 mms. 

 In order to test the insulating power of the system the connection of 

 the auxiliary battery was broken and the leak from the system ob- 

 served for an hour. The rate was found to be «48 mm. per minute, 

 which was not very high considering that the insulated system was 

 supported by a number of pieces of ebonite and that the air in the room 

 possessed a very considerable humidity since the experiments were con- 

 ducted in the summer time. The Dewar flask was then filled with care- 

 fully filtered liquid air and the electrode B and the connected system 

 again charged to ^ of a volt. The deflection again was found to be 99 • 5 

 divisions and when the auxiliary battery was disconnected the second 

 time, the leak from the system for an hour was found to be only at 

 the rate of '56 mms. per minute. This showed that the leak from the 

 whole system was practically the same, when the Dewar flask contained 

 liquid ail" as when it contained only ordinary gaseous air at atmos- 

 pheric pressure. 



In a second set of experiments the Dewar flask was in turn filled 

 with amyl, ethyl, and methyl alcohol, and the electrode B and attached 

 system was joined, with each liquid in the flask in turn, to the \ volt 

 auxiliary circuit. With the amyl alcohol the steady deflection obtained 

 under these circumstances was only 97-5 mm, with ethyl alcohol 

 76 mm., and with methyl 63 mm. With these deflections the current 

 from the battery to the electrode was equal to the current from the elec- 

 trode through the liquid to earth. In all three cases when the auxiliary 

 charging current was disconnected, the electrode system at once lost 

 its charge and the electrometer needle immediately returned to the 

 zero position. These experiments shew that of the three alcohols 

 the amyl was the best insulator and the methyl the poorest. The exper- 

 iments also shew that the liquid air was a very much better insulator 

 than any of the three alcohols. 



In another set of experiments a small air condenser of about 100 

 e.s.u. capacity was made up of brass plates and inserted in a brass box, 

 the latter being placed in a second box resting on and surrounded with 

 wool. This condenser was charged to a potential of 894 volts with a 



Sec. Ill, 1913—3 



