60 On Secondary Currents and a new Secondary Battery. 



presents a vertical section of the instrument through a pair of 

 springs across the axis of the cylinder. The plates C and D are 

 thrown out of contact with the springs, which, through the now 

 horizontal brass rods across the cylinder, become connected me- 

 tallically in pairs, as shown in the figure 3; i. e. Aj toB„ A 2 to 

 B 2 , . . . A n to B n , so that R x is connected with S, R 2 to L d , R 3 

 to L 2 , ... T to L 10 , the whole ten sets being thus joined toge- 

 ther in zigzag fashion ; the forty cells will then be con- 

 nected in series and discharged through the conductors S and T. 

 Complicated (and, it is feared, too much so) as this description 

 may be, the instrument and its action are in fact very simple ; 

 and as this apparatus " functions" remarkably well, and may 

 be used for other and analogous purposes, the account has been 

 made somewhat elaborate. 



15. The current from this battery had for a secondary cur- 

 rent considerable duration ; after one excitation, repeated defla- 

 grations could be obtained with tin- or platinum-foil ; with the 

 latter of the thickness of *1 millim. a hole was pierced 4 mil- 

 lims. in diameter, the edges of which were fused. Its electromo- 

 tive force was measured by a Thomson's portable electrometer, 

 and the results are given in Table VII. 



Table VII. 



Potential. 



Five Grove cells 3 



Forty secondary cells 19 



Forty secondary and five Grove cells . 22 



(In order to throw the Grove battery into the same circuit 

 and in the same direction as the secondary battery, a very simple 

 and obvious addition was made to the connexions.) These 

 results were repeatedly obtained without any variation, and 

 are in complete accord. The electromotive force, therefore, of 

 the forty cells w T as equal to that of 31*6 Grove cells. 



16. After some months' inaction this battery had seriously 

 declined in power; and on examination it was found that the 

 flannel was almost entirely destroyed. The couples were ther- 

 fore reconstructed, and on a smaller scale. Each of the copper 

 plates in each pair now made exposed an effective surface of 108 

 square centims., and was separated from the other by pieces of 

 ebonite at a distance of rather less than 2 millim s. On charging 

 forty of these couples by five Grove cells, and connecting the 

 terminals with the electrometer, the following results were ob- 

 tained : — 



