454 MEASUREMENT OF CAPACITY. DIELECTRICS. 



distant from all other conductors. We have seen how Sir W. 

 Thomson gets over this difficulty by using a guard ring. The 

 influence of the edges is not altogether eliminated in consequence 

 of the small interval which must be left between the plate and the 

 ring ; this is allowed for by replacing the area a of the movable 

 plate by the mean between this area and the aperture a of the ring. 

 The capacity C formed by the plate a, with a parallel disc at a 

 distance e, and of such dimensions in reference to the plate as to 

 be considered infinitely small, is expressed by 



8** ' 



1044. It is desirable in practice to have the capacity of a 

 condenser in electromagnetic units ; as the measurement deduced 

 from the dimensions is expressed in electrostatic units, it is necessary 

 to know exactly the ratio of the units in the two systems (610). It 

 is best to determine the capacity of standards directly, by electro- 

 magnetic methods. 



The capacity of a condenser being the ratio of a charge of 

 electricity to a difference of potential, the measurement of capacity 

 in absolute value will be determined by that of a charge (883) and 

 of an electromotive force (1000). 



We observe, further, that the capacity may be regarded (609) as 

 the quotient of a time by a resistance, or of the square of a time by 

 a coefficient of induction. The measurement of a capacity may thus ^ 

 be referred to that of a resistance, or of a coefficient of induction. 



The practical unit of capacity is the microfarad equal to io~ 15 

 C.G.S. units (613). 



Sheet condensers are made which have only very slight absorp- 

 tion, and which, conveniently subdivided, form boxes analogous 

 to resistance boxes. 



The association of capacities in cascade (85) gives a result 

 comparable with that of conductivity boxes (928). The reciprocal 

 of the capacity of a battery of condensers arranged in cascade is 

 equal to the sum of the reciprocals of the capacities which form 

 the cascade. 



It is interesting to remember that Cavendish had used an ar- 

 rangement analogous to the actual boxes.* He used Franklin's 



* CAVENDISH. Electrical Researches, published by Maxwell, p. 157. Cam- 

 bridge, 1879 



