ELECTRIC GENERATORS. 



101 



charge a Leyden jar or battery having a given amount 

 of coated surface. 



The results are only approximate, especially those 

 by the first method, for the following reasons. Length 

 of spark is not a true index of energy ; since a short, 

 thick spark may have greater energy than a long, thin 

 one: and our estimate of the comparative energy of 

 each from its appearance, and the accompanying snap, 

 is liable to be very inaccurate. The spark accom- 



Fig. 30 Lane's Unit Jar. 



panying the discharge of a Leyden jar or battery is 

 generally quite short, though its energy often greatly 

 exceeds that of any single spark of much greater length, 

 given by the machine in charging it. 



The humidity of the air and its electric potential, 

 being liable to great variation, produce a corresponding 

 variation in the results obtained at different times. 



The charge and discharge of a Leyden jar of a given 

 capacity, in a given time, is a more reliable method. 



