Thin Metal Partition in a Voltameter. i97 



Then 



4=7*= -000000224 x 605 = *0001355, 



i> 1= 2 2 r 2 =-0001355 x 1100 = -14907 = ^, since r 1 = I, 



I=« 1 + «2= , 1492 ampere. 



2 

 w=R— B/=two scale-divisions = ^ = \L. 



20 



(20 = o-= scale-divisions for one ohm change of R.) 

 Iio=zx= 0*014 volt. 



In this manner, from such a table in the note-book, one 

 line in Table II. was obtained. The sample-page of my 

 notes for J 7 gold was selected to illustrate the method of 

 observation and reduction, because it also serves as a fair 

 sample of many efforts to measure quantitatively the polariza- 

 tion on gold-leaf in good-conducting (30 %) H 2 S0 4 . One 

 has only to look at the time-variation of the polarization 

 on the end electrodes to be able to account for such slight 

 variations from zero as occur here (see Table III.). Having 

 become satisfied from many such series of observations that 

 there was no measurable polarization on gold-, aluminium-, 

 and platinum-foil for currents as strong as used, I then 

 adopted an efficient method of testing them for polarization 

 in various solutions, viz., to allow the strongest current to 

 flow through the plate long enough to become constant, and 

 then, without opening the circuit, to remove the metal parti- 

 tion with a glass rod and simply observe whether there was a 

 change of deflexion of the galvanometer. On this method of 

 observation is based the statement under the tables that for 

 these thinnest partitions there is no polarization for the 

 currents used. The qualitative results of Table I., with the 

 old voltameter, seem to indicate pretty clearly a small pola- 

 rization on ft 7 gold in very dilute (badly-conducting) acid. 

 A preliminary test in the new voltameter of water containing 

 only a few drops of acid, using gold-leaf as partition, gave 

 sufficient polarization to develop visible gas. 



From Table IV. (p. 198) it seems that for a given current- 

 strength the polarization is fairly proportional to the square 

 of the thickness of the partition for plates near the lower 

 critical thickness. The last column shows this ratio as worked 

 out for # 5 and # 6 gold. 



Phil' Mag. S. 5. Vol. 37. No. 225. Feb. 1894. 



