186 Mr. J. Daniel on the Polarization upon a 



being made periodically until the polarization becomes con- 

 stant, or nearly so. 



These are the principal direct problems that have been 

 attacked. Numerous minor observations have been made and 

 ■will be discussed in their place. The experiments have been 

 made in a room of fairly constant temperature, and no attempt 

 has been made to account for slight variations of temperature 

 as affecting resistance &c, as the quantities to be measured 

 are interesting mainly as regards their relative values ; but 

 especially because such a slight correction would be cloaked 

 by such unavoidable variations as irregular escape of gas from 

 the electrodes or partition, in case of its development there. 



Apparatus. 



The instruments employed as well as the plan of the experi- 

 ments are very simple. As a galvanoscope as well as current- 

 measurer, a w r ell-damped ring-galvanometer was used. The 

 galvanometer had four coils of about 22 ohms each, which 

 could be connected as desired. Deflexions were read with a 

 telescope on a scale divided into millimetres, fixed at a 

 distance of 1*5 metre. The instruments were supported by an 

 independent stone pier. As nearly as possible the same de- 

 flexion of the galvanometer was always used, the resistance 

 being so adjusted as to effect this; and the "figure of merit " 

 of the galvanometer for this position was determined by con- 

 necting a Daniell cell with the galvanometer in series with a 

 known resistance of such value as to give the deflexion to 

 be used, and found to be, by Ohm's law, 



y= 0*000000265 ampere per scale-division. 



For a new position used later, 7= '000000224. 



The connexions for the galvanometer circuit are shown in 



%. 1. 



(a) To measure the strength of current flowing through 

 the voltameter at any time, we observe the scale-deflexion, 

 and apply the following simple relations of current and 

 resistance, by Kirchhoff's law : — 



(1) 1=^ + 4. 



(2) hr 1 =i 2 r 2 , but 4 = 7^ and r 2 = 1100, always ; 

 .-. 1100 x 0-000000265 x a=i iri = i 2 r 2 . 



Here, r x being known, i x is determined, and from (1) I is 

 found. The shunt resistance, r 1} was changed as the main 

 resistance, R, was altered so as to maintain the same deflexion 

 of the galvanometer. 



