174 Prof. E. Edlund on the Path of the 



derations that, like those resulting from voltaic induction, these 

 currents are formed of two currents equal in quantity, one of 

 which has the same and the other the opposite direction to that 

 of the primary current*. As these currents are equal in quan- 

 tity, and in opposite directions, they cannot deflect the magnetic 

 needle ; but they can disengage heat, and, as their intensities 

 may be unequal, can also produce magnetic induction in har- 

 dened steel needles. This view as to the nature of the induction- 

 currents in question, which rests upon theoretical considerations, 

 has been confirmed since the discovery of the electrical valve has 

 furnished an unfailing means of distinguishing between the two 

 opposed currents. The electrical valve consists of a hollow glass 

 cylinder in which air is rarefied at pleasure. One end of this is 

 closed air-tight by a glass disk ; and at the other end is a brass 

 cap with a stopcock, by which it can be connected with an air- 

 pump. Through the glass disk passes a platinum wire, of 

 which one end is level with the inner surface of the glass disk, 

 and the outer end can be connected with a conducting-wire. 

 Inside the cylinder a brass rod extends from the brass cap ; the 

 rod terminates in a brass disk, which is parallel with, and at a 

 short distance from, the glass disk. When the air is adequately 

 exhausted, and the platinum wire connected with one and the 

 brass cap with the other end of the induction-spiral, it is proved 

 that only one of the two induction-currents can traverse the 

 valve; for Riess found that when a galvanometer is placed in 

 the circuit, the magnetic needle gives a deflection in a direction 

 which differs according as one or the other end of the induction- 

 spiral is connected with the platinum wiref. 



In my investigation on the electromotive force in the electrical 

 spark, there was no other spiral in the circuits than those which 

 were formed by the forty coils of the galvanometer J. In this 

 spiral induction-currents were of course formed when the elec- 

 trical discharge traversed them ; but it is readily seen, from the 

 manner in which the experiments were arranged, that these in- 

 duction-currents could have no influence upon the deflection of 

 the magnetic needle. In the adjacent figure, A B represents the 

 rotating induction-disk, and a b the two combs. An insulated 

 copper wire, a c, was directly connected with «, whereas the insu- 

 lated wire de terminated in a brass knob d in the neighbourhood 

 of b. From c and e insulated conducting-wires passed to the 

 knobs / and g. Two other conducting-wires went from the 

 points c and e to the galvanometer G. At m a rheostat was in- 

 serted, consisting of an insulated thin German-silver wire. Be- 



* Berzelius, Jahresbericht, vol. xx. p. 119. 



t Pogg. Ann. vol. cxx. p. 513. 



% Ibid. vol. cxxxiv. p. 337. Phil. Mag. S. 4. vol. xxxvii. p. 4\. 



