100 Dr. J. C. Beattie on the Leakage of Electricity from 



the leak of electricity, whether positive or negative, was 

 not perceptibly increased in these circumstances. That is, 

 we have here the well-known fact demonstrated that steam 

 coming away from electrified water and forming an atmo- 

 sphere of steam round the insulated electrified body, causes no 

 greater leakage of electricity from the electrified body than 

 there is from it when it is surrounded by air at the ordinary 

 room temperature (15° centigrade). 



§ 4. Mode of conducting the Experiments. — In making the 

 experiments the same procedure was observed throughout. 

 The insulated parts of the electrometer and of the air-leyden 

 were charged to a potential of about 220 volts by means of an 

 electrophorus. These were then left to themselves, and the 

 leak observed for five or ten minutes with no flame. The 

 flame was then applied and the continued fall of potential 

 was again noted, usually for periods of ten to twenty minutes. 

 (In the case of those substances where no increased leak was 

 observed, it was found that heating for an hour or an hour and 

 a half produced no increased rate of leak.) The flame was then 

 stopped, and the fall of potential observed for another five or 

 ten minutes. At times an attempt was made to measure the 

 temperature. To do this a mercury -thermometer rending to 

 350° C. was placed with its bulb between the bottom of 

 the box and the lower of the two insulated plates, and in such 

 a position as to touch neither. Usually, however, in cases 

 where a comparison was being made, the heating was con- 

 tinued for the same time in the different cases. 



§ 5. Results obtained with Zinc Plates insulated in Iron 

 Box. — It has already been mentioned that with zinc plates 

 insulated in the iron box no increased leak was produced 

 when heat was applied, either when the zinc plates had 

 nothing on them or when they were covered with water. 

 The heating in some cases lasted for an hour. 



No increased leak was produced when iodine was placed on 

 the zinc plates charged to 200 volts, and heat applied for 

 upwards of twenty minutes. Iodine vapour began to come 

 off after heating for five or six minutes. Similarly, no 

 increased leak was produced when bromine was placed on 

 the insulated zinc plates and allowed to vaporize. Also, if 

 potassium bichromate or potassium chromate was spread 

 over the upper surfaces of the insulated zinc plates, and 

 those charged to 220 or 240 volts, no increased leak was 

 observed after heat had been applied for upwards of twenty 

 minutes. All that has been said so far applies whether the 

 charge was positive or negative. On the other hand, if 

 potassium bichromate was placed on the zinc plates and 

 iodine sprinkled over it, then on heating for eight or ten 



