CHAP, iv.] ELECTRICITY AND MAGNETISM. 225 



cation employed for the moment. This arrangement, 

 when employed in conjunction with an induction coil 

 (Fig. 148) and a rapid commutator, admits of the in- 

 ductive capacity being measured when the duration of 

 the actual charge is only very small, the electrification 

 being reversed 12,000 times per second. Such an instru- 

 ment, therefore, overcomes one great difficulty besetting 

 these measurements, namely, that owing to the apparent 

 absorption of part of the charge by the dielectric (as 

 mentioned in Art. 53), the capacity of the substance, 

 when measured slowly, is different from its " instantane- 

 ous capacity." This electric absorption, which gives 

 rise to residual charges, appears to depend upon the 

 fact that many dielectrics conduct slightly, and hence 

 their apparent capacity is greater if the measurement is 

 made slowly ; for then a larger quantity of electricity 

 will have to be imparted in order to raise the potential 

 to a given degree, than would be necessary if there were 

 no such absorption. For this reason the values assigned 

 by different observers for the inductive capacity of various 

 substances differ to a most perplexing degree, especially 

 in the case of the less perfect insulators. The following 

 Table summarises Gordon's observations, which are 

 probably much nearer the true " instantaneous " values 

 than any others : 



Air . . . . I -oo 



Glass . . . . 3-013 to 3 -2 58 



Ebonite .... 2-284 



Guttapercha . . . 2-462 



Indiarubber . . . 2-220 to 2-497 



Paraffin (solid) . . . 1*9936 



Shellac . . . .2-74 



Sulphur . . . 2-58 



Hopkinson, whose method was a " slow " one, found 



for glass much higher inductive capacities, ranging from 



6-5 to io- 1, the denser kinds having higher capacities. 



Cavendish observed that the apparent capacity of glass 



Q 



