16 Prof. E. Edlund's Researches on the Passage 



the conclusion that e increases or diminishes with the intensity 

 of the current. I have, on the contrary, demonstrated, in a 

 previous investigation on the electromotive force of the voltaic 

 arc, that the above-mentioned force is independent of the 

 intensity of the current to which the arc is due. It was only 

 when the current employed was so weak that it was only just 

 able to produce an arc that the electromotive force appeared 

 to have diminished a minute fraction per cent.* If it is 

 otherwise when the experiments bear upon the electric spark, 

 that must depend on other conditions. When the electricity 

 traverses the rarefied-air space in the form of discontinuous 

 discharges, and the quantity of electricity which passed in the 

 way indicated in the preceding experiment is augmented or 

 diminished, it is possible that not only the duration of the 

 separate sparks, but also the curves indicating the increase of 

 the eletricity at the commencement and its decrease at the 

 end of the discharge, undergo modifications. The electromo- 

 tive force e probably depends on these circumstances, and is 

 not in direct relation with the quantity itself of electricity. 



Already in 1868, at the time of my first researches on the 

 electromotive force of the electric spark, I had found that the 

 deflections become smaller if an induction-coil is inserted in 

 the circuit conducting to the galvanometer. In those experi- 

 ments the spark was formed in a space filled with air. There- 

 fore, of the two induced currents due to the passage of the 

 discharge through the coil, only that which traversed the spark 

 in the same direction as the discharge had power to penetrate 

 the spark ; or perhaps, to express myself more correctly, that 

 current traversed the spark with more facility than the induced 

 current which traversed it in the contrary direction. As 

 there might be some interest in ascertaining what would be 

 the results of lowering the pressure of the air in the space in 

 which the spark was produced, the following experiment was 

 made : — 



Experiment 6. At w (fig. 1) an induction-coil was inserted, 

 consisting of a copper wire coated with gutta percha and 

 wound in 40 turns of a helix ; and then the deflection pro- 

 duced was observed. The coil was now removed, and replaced 

 by a German-silver wire having nearly the same resistance as 

 the induction-coil, and the deflection was observed in the same 

 manner as before. When no spark was formed between c and 

 d, and consequently the entire discharge divided itself between 

 the galvanometer and the bridge, the deflection amounted to 

 1*5 scale-division. In this way, at different air-pressures the 

 following series was obtained: — 



* Pogg. Ann. cxxxiii. p. 353 ; Phil. Mag. [4] xxxv. 



