On the New Electromotive Force in the Voltaic Arc. 353 



i 



P 



o 



of the following construction was Fig. 1. 



used. In fig*. 1, n, n' and I, l ! re- 

 present four mercury-cups bored 

 in a mahogany base, in which the 

 various parts of the commutator 

 were fastened. In the middle was 

 a horizontal axis to which a lever 



was fastened, both whose ends » /"N x~~~\ 



were forked, and these forks were 

 bent downwards at right angles 

 to the longitudinal direction of 

 the lever. When the one half of 

 the lever was depressed, the cups 

 n and n! were connected with each 

 other, because the forked points 

 at the end of the depressed lever 

 were dipped in these cups. If 

 the other half of the lever was ~^= 

 lowered, the cups / and /' were 



connected, and the connexion between n and n ceased. By 

 means of a copper wire,^, the cups n f and /' were connected. The 

 negative pole of the battery was connected with the cup n, 

 and a wire went from l' to b, where the arc was formed. The 

 wire b q r led to the positive pole of the battery. From q a wire 

 led to the magnetometer g ; and the other end of the magneto- 

 meter-wire was in connexion with the cup /. When n and n 1 

 were connected with each other by the forked lever, a luminous 

 arc was formed at b. If the lever was reversed, the connexion 

 between n and n 1 was broken and the current ceased, upon which 

 the cups / and /' were connected. By this means, one end of 

 the magnetometer-wire was connected with each side of the lu- 

 minous arc b. Hence if, after the cessation of the principal cur- 

 rent, there were any residual electromotive force, this would send 

 a current through the magnetometer and produce a deflection. 

 The magnetometer was the same as that I had used in previous 

 researches. If a thermo-couple consisting of a German-silver 

 and an iron wire was connected with the magnetometer, an in- 

 crease of 1 degree in the temperature of the soldering produced 

 a deflection of 54 divisions. 



It is of course important that the connexion between / and V 

 ensue as rapidly as possible after the cessation of the connexion 

 between n and n r . But, on the other hand, care must be taken 

 that, in reversing the lever, n and n' are not connected with each 

 other for a brief space as well as / and /'. If this is the case, 

 part of the principal current passes during the simultaneous 

 closing through the galvanometer-wire, and the magnetic needle 

 gives in consequence a deflection which is due to the principal 



Phil. Mag. S. 4. Vol. 36. No. 244. Nov. 1868. 2 A 



