402 RHODIN, ON STORAGE BATTERIES. 



From obvious physical reasons it follows: — 



Ut^cpIJt (6) 



Hence, X must keep constant or decrease. If it keeps constant 

 this diffusion theory is not applicable to the explanation of cessa- 

 tion or current Output which would then depend upon complete 

 reduction of the Pb0 2 or other chemical changes. If X decreases, 

 it finally assumes a value equal to Cineg : n (1) and the current 

 ceases_(or cannot be kept up), tmeq:n (5) has then the value 

 T, and T is then a function of the density of the electrolyte 

 (5 by integration): 



T=xpd (7) 



Hence the capacity of the cell is a function of the density of 

 the electrolyte or 



y = lyd (8) 



That this is a fact for plates like the Laurent-Cely plates which 

 contains a large excess of peroxyde of lead is evident from the 

 contemplation of the curves in Fig. II and III which indicate a 

 very simple relation betweeu the two characteristics of the accu- 

 mulator, capacity and conductivity of the electrolyte which might 

 be expressed in a form like the following: — 



V- = k + e z -' n (9)' 



where y = capacity and z = conductivity of the electrolyte. e y 

 k and m are numerical constants. This last formula is not 

 strictly correct, but shows the introduction of a function of ß 

 depending upon the nature of the peroxyde of lead and its 

 porosity etc, I do not wish to press this point as being very 

 important. When we made the previous reasoning, it was ne- 

 cessary to assame a large excess of peroxyde of lead as being 

 present. If my reasoning is correct, it follows that the Laurent- 

 Cely plate and all other plates which show a rapid variation of 

 capacity with the density of the electrolyte must have peroxyde 

 of lead or spongy lead in too thick layers for the free admission. 



