292 K L. Nichols and W. S. Franldin — Electromotive 



galvanometer could then be sensitized to any desired extent, 

 but the fluctuations in the electro-motive force of the iron cell 

 were such that the balance was never more than momentarily 

 maintained, and the galvanometer drifted continually. 



In very weak acids and in solutions of FeS0 4 , FeCl„ NH 4 C1, 

 and similar salts, these fluctuations were not such as to pre- 

 clude the possibility of making measurements, but in concen- 

 trated acids they were for the most part so rapid and irregu- 

 lar as to put galvanometer readings out of the question. In 

 nitric acid, of considerable strength, these fluctuations were 

 so remarkable as to deserve special mention. The electro mo- 

 tive force, amounting to a considerable fraction of a volt, 

 changed sign continually, carrying the spot of light across the 

 galvanometer scale, to and fro, with great rapidity. The fre- 

 quent and irregular oscillations continued with undiminished 

 violence until the electrodes were entirely dissolved or the 

 acid nearly neutralized. Upon first closing the circuit one of 

 the iron terminals would be slightly more active than its fel- 

 low. The tendency of the less active electrode would then be 

 to protect the other from the attack of the acid, rendering it 

 temporarily passive, as a piece of platinum would do under the 

 same circumstances. The passive terminal would then react in 

 like manner upon the first and electro-motive force would be 

 reversed again and again until the electrodes were consumed. 



In those cases in which the flu'ctuations were not such as to 

 make the attempt at compensation ineffectual, we were able 

 to make measurements of the initial electro -motive force of the 

 cell and of the changes in electro-motive force caused by the 

 action of the magnet. 



One of the iron terminals was placed between the poles of 

 a large electro-magnet. To obviate any direct effect of the 

 magnet upon the galvanometer needle, the galvanometer was 

 set up in a room several hundred feet distant from that in 

 which the former instrument was located. The " iron " cell 

 having been connected with the wires leading to the galvan- 

 ometer room, the initial electro-motive force was balanced by 

 the method already described, and the galvanometer was 

 brought to the desired degree of sensitiveness by means of a 

 governing magnet. One observer then proceeded to make 

 galvanometer readings at intervals of fifteen seconds, while 

 another magnetized and demagnetized the electro-magnet every 

 two or three minutes, reversing the magnetizing current each 

 time. The electro-magnet in question has been described in a 

 previous paper.* 



Like most large instruments of the kind, it required several 

 seconds after the circuit had been closed to attain its full 



* This Journal, vol. xxxi, p. 272. 



