104 Prof. E. Edlund on Galvanic Resistance as 



motor is inserted between the middle disk and the binding- 

 screws g and h ; so that the deflections obtained do not derive 

 their origin from any sort of diaphragm-currents or from any- 

 other extraneous cause. The following observation (to which 

 I would direct the attention of those who may desire to control 

 my experiments) does not contradict this : — If, as soon as a 

 series of experiments is concluded, the electromotor be with- 

 drawn and instead of it a simple conducting-wire connect the 

 middle disk with the binding-screws g and h, and thereupon 

 the liquid be set in motion, a deflection will be obtained, and 

 this in the same direction as the one a short time before ob- 

 tained while the electromotor was in the circuit ; and if the 

 glass tube be now inverted, we shall have, quite regularly, a 

 deflection in the opposite direction. It is easy to perceive 

 whence these deflections proceed. When the electromotor was 

 inserted the disks in the side-tubes were polarized, and the two 

 outer ones received a coating of the same sort of gas. Since 

 the disappearance of the polarization is gradual, on this account 

 a polarization-current went through each half of the water 

 column immediately after the removal of the electromotor ; but 

 the current in one half was opposite to that in the other, in the 

 same manner as when the electromotor was in the circuit ; and 

 when the liquid was put in motion, the resistance in one half be- 

 came greater, and the other less, than before. Thereby the two 

 currents acquired unequal intensity; and in consequence of 

 this the magnetometer-needle could not but make a deflection. 



IV. 



The result of the above experiments I have endeavoured to 

 control by a method of investigation in every respect different 

 from that just described ; it was the following : — 



Kohlrausch and Nippoldt * have [shown that alternating in- 

 duction-currents, succeeding each other at sufficiently brief 

 intervals, can be advantageously employed for making the re- 

 sistance of liquids independent of polarization. When water is 

 the electrolytic liquid, only a very thin covering of oxygen can 

 form on the one polar disk, and of hydrogen on the other, during 

 the short time that the current lasts ; and in the next instant 

 the directions of the current are reversed, so that the disk 

 which just before was receiving hydrogen receives now oxygen, 

 and vice versa. If the currents are of equal intensity and 

 short duration and follow close upon one another, it is easy to 

 see that the polarization produced by them must be reduced to 

 a minimum. Indeed the investigations of the above-mentioned 

 philosophers showed also that the polarization was without per- 

 * Pogg. Ann. vol. cxxxviii. p. 280. 



