of Edinburgh, Session 1881-82. 569 
distance of the tube was varied through a range extending from 
1 cm. to 7 cm. 
The method of measuring the potential required to produce the 
spark was that employed in Macfarlane’s previous experiments.* 
The special arrangement of the apparatus is indicated by the 
accompanying plan. 
A and B are the two electrodes of a Holtz machine, of which one, 
say A, was connected with the earth. An insulated wire joined B 
to the knob of a large Leyden jar C, and C was similarly connected 
with the disc D and with the insulated ball F. G, the other 
insulated ball, was connected with one of the electrodes H, of a 
Thomson quadrant electrometer, the other electrode K being earthed. 
An uninsulated metallic plate L, with a hole in the middle of it, was 
placed between F and G, for the purpose of diminishing the 
influence of F on G. This is a modification of an idea introduced 
by Professor Chrystal, namely, the placing of F inside a metallic 
cube having a hole in the side next G. 
* Trans. Roy. Soc. Edin., vol. xxviii. p. 633, or Phil. Mag., s. 5, vol. x. p. 
389 . 
