IN A CLOSED GALVANIC CIRCUIT. 253 



tested, and then previous to every experiment covered with a 

 ground glass plate. The resistances themselves were measured 

 by a rheostat of Jacobi's construction. The coils were of German 

 silver wire 0*62 miUim. in thickness ; the length of each coil, 

 which is assumed as the unit of resistance to conduction, 

 amounted to 220*5 miUims. The temperature was from 18° to 

 20° C. After determining the resistances, the liquids were 

 poured into the apparatus described in § 7? and the heights of 

 the mercury in the manometer were noted for currents of dif- 

 ferent intensities. The following table shows the results thus 



obtained. In it the numbers in the column ^ denote the per- 

 centage of crystalline salt (Cu S H- 5aq) in the solutions ; the 

 numbers under r are the observed resistances of the solutions ; 



the numbers under -7 are the mean ratios, each found from 



several experiments, of the height attained by each solution to 

 the intensity of the current. 



From this table it is evident, that, whilst the resistances in- 

 creased from 1 to 6, the ratios of the heights attained by the 

 liquids with currents of equal intensity to the corresponding 

 resistances of the liquids differed only yV from their mean value. 

 If, therefore, a general conclusion may be drawn from the 

 present experiments, it must be thus expressed : — 



The heights to which the several liquids are raised by the gaU 

 vanic current are^ under otherwise equal circumstances, directly 

 proportional to the specific resistances of these liquids, 



* The resistances of the very dilute solutions are perhaps here somewhat too 

 great, inasmuch as on the copper electrodes immersed in them, a thin stratum 

 of oxide was deposited during the process of determination. 



