produced by the Electric Current. 475 



Iodide of Cadmium. 









Table IV. 











Tube No. 2. 



A-D. 



1. 



A. 



192-4 



D. 185-8 



6-6 



2. 



D. 



182-0 



A. 189-3 



7-3 



3. 



A. 



188-4 



D. 181-5 



6-9 



4. 



D. 



183-0 



A. 190-3 



7-3 



5. 



A. 



190-5 



D. 184-8 



5-7 



6. 



D. 



183-3 



A. 190-5 



7-2 



Difference A - D = 6S. 



The previsions given by the theory are, then, confirmed by 

 experiment, inasmuch as a difference is shown between the 

 intensities of the ascending and the descending current in a 

 column of an electrolyte, and the sign of the difference ac- 

 cords with that indicated by the theory. In nitrate of silver 

 the descending current is stronger than the ascending ; the 

 reverse is the case with iodide of cadmium. 



But the numbers given by experiment are twice or three 

 times less than those foreseen by the theory ; it is impossible 

 to attribute so enormous a difference to errors of observation. 

 The most probable explanation of this apparent anomaly is to 

 be sought in a peculiarity of feeble currents noted by Helm- 

 holtz*. Feeble currents, which polarize the electrodes with- 

 out producing definite decomposition of the electrolyte (their 

 electromotive force being less than the electromotive force of 

 polarization which they would call forth in passing through 

 the liquid), or the currents which arise from the depolariza- 

 tion of the electrodes, present according to M. Helmholtz the 

 following peculiarity : — Ohm's law seems to be inapplicable 

 to them : a very great metallic resistance introduced into the 

 circuit weakens their intensity but very little, which amounts 

 to the same as if the liquid or the electrodes were the seat of 

 an enormous supplementary resistance. Experiment proves 

 that this fictitious resistance is variable and augments when 

 the intensity of the current diminishes. These phenomena 

 present a striking analogy with the currents given by elec- 

 trical machines. M. Rossetti's experiments on the Holtz 

 machine f have shown that what may be called the internal 

 resistance of that machine in like manner varies with the 

 velocity of rotation, consequently with the intensity of the 

 current. My tubes presented this peculiarity in a very high 



* Poggeridorff's Annalen, vol. cl. p. 483. 



t D'Almeida's Journal de Physique, 1875 ; Pogg. Annalen, vol. cliv . 

 p. 507. 



