62 S. L. Harding—Bichromate of Soda Cell. 
the moment of making until breaking circuit. By transferring 
the individual records taken under the same conditions to a 
sheet of codrdinate paper and referring them to the same axes, 
their relative changes from time to time are readily noted. The 
accompanying diagrams show a number of these records. Fig. 
1 illustrates better than many words the relative value of the 
two bichromate cells, at least as far as their constancy is con- 
cerned ; it gives two records of each of the two classes of cells, 
and, for the sake of comparison, a record of a Daniell’s cell. 
It is to be observed that in Series 1 and 2 circuit was made im- 
mediately after setting up the cells, before the liquid had time 
to diffuse itself through the pores of the inner cup. The in- 
creased resistance, due to this cause, and the gradual return to 
the normal resistance during the first hour are plainly recorded. 
The following table gives the length of the records or the total 
number of hours before the exhaustion of the depolarizer : 
Double Liquid Cells. 
Hours before 
Depolarizer. External resist. exhaustion. 
Series) see Bichr. of potash 8 ohms 554 
Series 2 .-=-- ce Seen: Sass: 564 
Series 3 ....-Bichr. of soda cya 78 
Series, 452-2 7“ aie Sirs 742 
Series 5 _---.% Sulphate of copper Soe 
During the first 24 to 30 hours the records show little differ- 
ence in the working of the two classes of cells, both remaining 
comparatively constant and, therefore, giving for records nearly 
straight lines; though it must be understood that even in the 
case of the most constant cells, such, for instance, as the 
Daniell’s cell, the curves are never absolutely straight lines but 
rather a series of slight but numerous fluctuations about the 
position ofa right line. These fluctuations are far more marked 
during the first 10 or 15 hours than later, but at no time are 
they entirely absent. Both the potash and the soda cells, in 
the above series, showed signs of weakening after 30 hours. 
This weakening, however, proceeded far more rapidly, and, 
consequently, exhaustion was reached much sooner in the 
former class of cells than in the latter. As can be seen from 
the table, the soda cells ran on an average fully 20 hours, or 
more than one-third, longer than the potash cells. 
The comparatively short life of the depolarizing liquids in 
all the above cases was due to the small resistance in circuit. 
As a proof that the soda cell is capable of long continued 
action on a larger circuit, I need only cite a record which 
lasted, with 35 ohms in the external circuit, over 200 hours 
without showing signs of exhaustion. On large circuits where 
