Electricity in a uniform plane conducting Surface. 479 



battery used was a single cell of Smee; the galvanometer was a 

 reflecting one of low resistance. The disk was a circular piece 

 of tinfoil, of 27*8 centim. diameter, pasted on the bottom of a 

 tray or shallow box of hard wood, on the edges of which rested a 

 wooden bar supporting the contact-pin C. This last slid in a 

 vertical tube fixed to the wooden bar j a spiral spring inside the 

 tube kept the pin out of contact with the sheet except when it 

 was pressed down. When the right position on the sheet had 

 been found by means of the blunt point of the pin, this position 

 was marked by pressing the pin down harder, so as to make a 

 slight indentation in the tinfoil. 



The mode of making contact at A and B which we found to 

 answer best is shown in fig. 2. A S is a brass spring with a 

 projecting pin A, and a screw Fig. 2. 



S with nut and washer. The 

 screw passes through the wood 



of the tray and is fixed by screw- 

 ing up the nut, the conducting 

 wire being clamped between 

 the nut and washer. The rounded end of the pin A rests on the 

 tinfoil surface, pressing it firmly and making a good circular 

 contact. It was found necessary to lubricate the bar carrying 

 C with blacklead to prevent its causing vibrations when sliding 

 on the edges of the tray, and so deranging the contact of the 

 springs A and B. 



49. After a series of points have been marked on the tinfoil 

 in the manner described, a line may be drawn through each set 

 by hand. A set of lines obtained in this way for the case of two 

 equal opposite poles on the edge of a circular disk correspond to 

 a portion of the diagram in Plate IX. (Part I.) bounded by a 

 complete line of flow. P being any one of the dots marking 

 such a line,' if they are circles with the poles as inverse points, 



AP 



the ratio ^> will be constant for any one line (Part I. § 16) ; 



and if they are at equal differences of potential, the values of 

 this ratio will vary from one line to the next by a constant factor 

 fM (Part I. § 18). The numbers given in the following Table 

 show to what extent these requirements are fulfilled by the ex- 

 perimental lines. Nine or ten points on an average were deter- 

 mined for each line ; but as a specimen of the results it seems 

 sufficient to give the ratios found for four points only, viz. (in 

 the case of each line) for a point near either end, a point on or 

 near the straight line joining the poles, and a point about halfway 

 between this and the more distant end of the equipotential line. 



