﻿42 



Prof. E. Edlund's Experimental Proof that 



neously between/ and# and between 

 h and d. If the wires are connected <& 

 in the simple manner represented by J~ 

 the figure, a difficulty is met with "Q 

 which at first I could not entirely 

 remove. The electric discharge does 

 not, of course, entirely pass through 

 the spark/^, but part passes through 

 the galvanometer G. The galvano- 

 meter-wires coated with gutta percha 

 acquire thus an electric charge which 

 acts electroscopically upon the mag- 

 netic needle and draws it on one side. 

 This charge was retained for some 

 time after the machine was at rest, 

 and caused a deflection. It is possible 

 that this charge might have been 

 avoided if the galvanometer had been 

 differently constructed, and the con- 

 ducting wire had been otherwise in- 

 sulated. The charge in question 

 was ultimately quite removed, and the position of equilibrium of 

 the needle made quite independent of it, by directly connecting 

 the points i and k with each other by a German-silver wire which 

 could be lengthened or shortened at pleasure. The point k was 

 moreover directly connected with the water-pipes of the house 

 by the wire /, by which it was placed in conducting communica- 

 tion with the earth. Part of the electric charge passed now 

 through h instead of through the galvanometer, and the electri- 

 city of either one kind or the other which might have been in 

 excess passed to the earth through /. After these wires had been 

 introduced, no charge was observed in the galvanometer. In all 

 these experiments the jars belonging to the electrophorus ma-' 

 chine were used. Without these there was indeed an electrical 

 discharge between b and d-, but it was impossible to produce 

 sparks between /and g, however great was the resistance at m. 



It is well known that the electric discharge of a battery acts 

 upon the magnetic needle and produces a deflection which in- 

 creases with the quantity of electricity. The direction in which 

 the deflection of the magnetic needle occurs is the same as if the 

 magnetic needle were surrounded by a voltaic current which went 

 from the positive to the negative coating. If, then, the knobs/ 

 and g are so far apart that no discharge ensues between them, 

 the magnetic needle must make a deflection when the disk is 

 turned. This was always the case; and when the disk rotated 

 with a constant velocity, the sparks between b and d followed 



