Electric Residue in the Leyden Jar. 

 Table a. Table b. Table c. 



307 



The series of observations contained under b is represented 

 graphically in fig. 2, Plate VI., by the line L^ The curves of the 

 two other series of observations are not drawn. 



On the whole these curves present a similarity to parabolas, 

 but they satisfy a constant parameter for a short distance only' 

 It is seen, however, at first sight that they cannot be due to the 

 mere loss of electricity in the air, as they are much too curved 

 to be accounted for in this manner. Adhering to the results 

 stated m b, and assuming that the loss of electricity is solely due 

 to Its dispersion in the air, the charge, which at first amounted to 

 1-4968 accordmg to the arbitrary unit of the sine-electrometer 

 being reduced in 5370 seconds to 05266, the corresponding 

 curve would be shown by the dotted line N, in which the stronger 

 points are calculated*. 



* This is proved as follows :-Let Lo be the charge of an insulated con- 

 uuctor at the commencement, then we have 



where « is a constant dependent on the state of the air. 

 lategrating from t=() to t=t, we have 



Y2 



