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



sphere whose temperature was gradually raised from the 

 temperature of the room up to 250° C. or 300° C. In this 

 paper only those experiments will be described where a 

 potential-difference of between 80 and 240 volts was used. 



§ 2. Apparatus. — The potential-difference was measured by 

 means of a Kelvin multicellular voltmeter (laboratory pattern). 

 The insulated terminal of this instrument was connected by 

 a copper wire to the insulated terminal of an air-leyden, A 

 (fig. 1). This air-leyden consisted of an outside metal box 

 34 cms. long, 9 cms. broad, 6 cms. deep : one end of this was 

 open; from the other and closed end a sheet of metal MM, of 

 the same breadth as the box, passed along the middle of the 

 box for about three quarters of its length and divided the 



Fijr. 1. 



'o Case 



space inside into two equal parts. Into each of these spaces 

 a metal plate P, 13 cms. long, 6 cms. broad, was put, insu- 

 lated from the box, and kept in position by a stout wire, w, 

 which was held by a wooden clamp, C, this latter being fixed 

 in a block of paraffin. These two metal plates met near the 

 open end of the box. The stout wire formed the insulated 

 terminal of the air-leyden. The box, A, was covered on the 

 outside by asbestos ; round the asbestos one or more layers 

 of wire gauze were wrapt. The whole was then supported 

 on two blocks of wood, W, W, and underneath it a strong 

 gas-flame could be applied. The wire joining the insulated 

 terminal of the air-leyden to that of the electrometer was 

 surrounded by a metal guard-screen, S. The box was made in 

 such a way as to render impossible the passage of fumes from 

 the gas-flame underneath into its interior. 



