482 M. R. Kohlrausch's 7%eo?7/ of the 



The line v^, which in Plate VI. fig. 3, represents the curve of 

 the loss of electricity, does not differ much from a right line, 

 although the corresponding observations lasted an hour and a 

 half. In the present case, where the observations lasted only 

 eleven minutes, the curve will differ much less from a right line, 

 so that we shall incur little error by assuming the loss to be 

 proportional to the time. 



If the loss duiing the whole operation was V, then up to the 

 time T, when the jar was first discharged, it would be 



F 

 V. 



and hence, approximately. 



t F 



_ V 



T' •F + 0" 



If, in another experiment, the primitive charge had been Qq' 

 instead of Qo, but the condition of the atmosphere the same, then 

 would ,_i V -1- ^ 



This expression, however, requires a correction if the condition 

 of the atmosphere with respect to the loss of electricity be dif- 

 ferent. To this end the loss of electricity which is sustained by 

 any insulated fi'eely placed body must be determined by means 

 of the torsion balance. Let fu, represent this loss in the original 

 observations on the jar to which the Tables a and a' and the un- 

 accented letters Qq, V, F, cf) refer ; and fJ the loss in the expe- 

 riment now under consideration, then 



^t- ^ 'T-' F + c^-Qo' 



Further, 

 therefore 



L/=Qo'-'V-<. 



(6 .'n+l\ 



If for Q/ we substitute its value Q^—vl, and for «/ the above 

 expression, we have 



IV. L/ = Qo'[l-j.(l-.-»-ltl''"^') 



u p) ^ •t'^-f+<^-QoJ- 



Hence if the charge Q^' which was originally imparted be 

 known, L/ can be found ; or if the latter be given, Qg' can be 

 found. 



