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



kept tightly in place by means of the small nuts Z, Z which 

 screw on to G. The nuts are insulated from the bolts and 

 shaft and from one another with strips of mica. The spindle Gr 

 is mounted at L on a needle-point that rests in a small hole 

 in the glass support N, which is shaped as shown in the 

 figure. M is the end view of a small bar-magnet which is 

 screwed on the spindle. By means of an outside horseshoe- 

 magnet revolving on a turntable one is able to turn M, and 

 therefore the contact couple, through an arc of 180°, just as 

 in the apparatus described in § 4. 



Two very fine platinum wires are connected to the dis- 

 similar metals E and F and pass out of the top of the glass 

 tube through the seals at P and Q *. K l5 K 2 are the platinum 

 inductors (each J in. by 5f in.). These are bent into arcs of 

 circles as shown in the plan of fig. 2, and sealed into the 

 highly insulating glass tubes R and S. Platinum wires running 

 through these tubes pass from the inductors K : , K 2 to the 

 outside of the apparatus, into which they are sealed. The 

 wire which was always kept earthed is sealed at T, as shown 

 in fig. 2. In the case of the insulated one S, it was necessary 

 to prolong the glass tube as shown in fig. 3 (S, S), so that a 

 couple of inches of its surface could always be kept highly 

 insulating by the common device of slipping on it a small 

 flask containing a little strong sulphuric acid (T, fig. 3) . The 

 cork (U) is just slipped upwards when readings are being- 

 taken. This flask is not shown in fig. 2, as it would there 

 appear in front of the apparatus instead of at the side, as in 

 fig. 3. The outlet W (fig. 2) served to connect the tube to 

 the air-pump. U is a ground-glass joint sealed in mercury. 



§ 6. The method employed of measuring the contact P.D. 

 between E and F was a combination of Ayrton and Perry's 

 method (described in § 4) and the well-known compensation 

 method used by Lord Kelvin f. The plates E and F were 

 connected through the platinum wires ending at P and Q to 

 two points on a potentiometer. The one inductor K x was 

 joined to the earthed quadrants of the electrometer, the other 

 K 2 to the insulated quadrants. The variable potentiometer 

 contact was then adjusted until, on insulating K 2 and reversing 

 the contact couple by means of the outside horseshoe-magnet, 

 no deflexion of the electrometer needle took place. When 

 this stage was reached the impressed P.D. from the potentio- 



* It was an operation of sonie difficulty, even to the practised glass- 

 blower, to make so many seals into one small piece of glass. 



t B. A. Report, 1880J aud ' Nature,' April 14th, 1881, also Dr. Lodge's 

 Report B. A. 1885. 





