MEASUREMENT OF VERY GREAT RESISTANCES. 385 



The extremity A is connected with a constant source at a very 

 high potential, like the internal armature of a battery of great 

 capacity, or one of the poles of a battery composed of a great 

 number of couples, the other armature or the other pole being 

 to earth. The end B might be connected with a discharging 

 electroscope (824). The number of contacts in unit time will 

 be inversely as the resistances. 



The extremity B might also be connected with a condenser. 

 If R is the resistance to be determined, C the capacity of the 

 condenser, V the potential of the source, and V that of a con- 

 denser after the time /, we have the ratio 



(f -' 



-J^l V 

 R~/ 'V-V" 



or, if V is very small in comparison with V, very nearly 



/v 

 R cv 7 ' 



985. Without directly bringing in the potential of the source, 

 we may still join the two armatures of a condenser by the resist- 

 ance in question, and determine the time / required for the 

 difference of potential to increase from V to V. We have then 



- 



R / ' V ' 



In this case the ratio of the potentials V and V will be de- 

 termined by an electrometer, or by the swings given to the needle 

 of the ballistic galvanometer, when the condenser is discharged in 

 its initial and in its final state. 



The choice of units is immaterial for comparative experiments. 

 For absolute measurements, the resistance R is expressed in elec- 

 trostatic or electromagnetic units, according as the capacity C is 

 evaluated in units of the former or of the latter system. 



If the condenser leaks, then, whatever be the cause, it is 

 equal to that which would be produced by a resistance R inter- 

 posed between the two armatures. R is first determined by the 

 same method. By connecting then the armatures of the resist- 

 ance R, the condition of the condenser is the same as if it were 

 closed by two parallel plates. Lastly, a fresh experiment gives 



the sum -f - - of the two conductivities, from which the value 

 K. K 



of the resistance R is deduced by difference. 

 VOL. n. c c 



