Mr. F. S. Spiers on Contact Electricity. 75 



meter exactly counterbalanced the Yolta E.M.F. between the 

 plates E and F. A voltmeter gave the reading directly. 



§ 7. The method of heating the lower part of the tube was 

 to slip over it a tube of copper lined with asbestos. It was 

 then found quite possible to bring the tube safely to a bright 

 red heat with a blowpipe flame. The large distance between 

 the source of heat and the upper part of the tube, only about 

 6 or 8 inches of the lower part being heated directly, pre- 

 vented the joints and seals from becoming unduly heated, 

 and conveyance of heat by convective air-currents was pre- 

 vented by suitably placed sheets of asbestos cloth. It was at 

 first feared that a large ground-glass joint such as C, C (fig. 2) 

 would not be sufficiently air-tight for the very high vacua 

 that it was desired to work with, but the joint proved to have 

 been so very carefully ground* that when scrupulously clean 

 it would readily hold a vacuum as high as ^ooob mm «? which 

 was almost the limiting reading reliably given by the McLeod 

 gauge. I always found that I could get a far better fit, 

 although, on the other hand, it was then far more difficult to 

 open after exhaustion, when the joint was quite clean and 

 dry than when it was smeared with even a slight trace of 

 glycerine. 



In o:her respects the apparatus did not prove so satisfactory. 

 In the first place it was necessary that the fine platinum 

 wires attached to the movable plates E and F should be so 

 flexible as not to exercise any appreciable control on the free 

 motion of the system. On the other hand, if they were too 

 loose there was fear of one or other of them coming into con- 

 tact with the platinum inductors at some position through 

 which it passed. It proved an exceedingly troublesome 

 operation to make the exact adjustment of length necessary, 

 and even at their best the wires always exercised a certain 

 amount of control, so that it was never possible to revolve 

 the moving system completely through 18l>° of arc. This, of 

 course, materially diminished the sensibility of the arrange- 

 ment. There was another cause which operated still more 

 in reducing the sensibility, and that was the comparatively 

 large distance between the movable plates and the inductors. 

 This was in consequence of the necessarily great length of 

 the spindle G, G, so that when the plates were closer than 

 about -^ in., quite a small amount of side play caused by pull- 

 ing round the system was sufficient to bring the plates into 

 contact. 



* I might mention that the glass-blowing required in all these pieces 

 of apparatus was most satisfactorily performed by Mr. C. E. Muller. 



