146 STATIC ELECTRICITY 



sufficiently good insulator to allow a persistent field to be estab- 

 lished in it, yet on making a spark-gap in one of the conncct- 

 ing wires, at each spark the field lighted up, due no doubt to 

 the oscillations producing a large oscillating field between the 

 conductors. 



Schmidt* succeeded in measuring the effect in nitrobenzene, 

 He used a compensation method in which the ray of light pas-cd 

 through the liquid in one cell and then through another cell con- 

 taining a liquid for which B was known, and arranged so as to 

 have an effect of the opposite kind on the phase relation of the 

 two components. It was adjustable so that the effect of the two 

 cells exactly neutralised each other when the same rapidly alter- 

 nating potential difference was applied to each. He found that 

 the Kerr constant is sixty times as great for nitrobenzene as for 

 carbon disulphide, and that it is about the same for the latter and 

 for water. McComb found that the effect is large for other 

 aromatic compounds.! 



Cotton and Mouton (loc. cit.) discou-ml I hat BlllobeBieM 

 exhibits magnetic double refraction, that is that a polarised ray 

 transmitted perpendicularly to a magnetic field established in the 

 liquid travels at different speeds according as its plant- of polarisa- 

 tion is parallel or perpendicular to the lines of force. They found 

 that the effect could be expressed by a formula exactly corre- 

 sponding to that for the Kerr electric effect, vi/. : 



S/\ = C/H* 



where H is the magnetic intensity and C is the constant of magnet ic 

 double refraction for the substance for the wave-length u-t-d. I'm 

 nitrobenzene they found that C/B is nearly the same for different 

 wave-lengths. This has been confirmed for several otlu i aromatic 

 compounds by McComb and Sk inner, t 



In Cotton and Mou ton's paper there is some reference to theories 

 of the subject by Voigt, Havelock, and themselves. These, how- 

 ever, belong to optics rather than to electricity. 



We may state the Kerr effect in terms of the electro- magnetic 

 theory of light by saying that K is slightly different in t la- 

 electric field. Since the change in velocity is given by 



dv _ldK 

 v 2 K 



and Dr. Kerr's observations show that is proportional to K 2 . 



7|- 



where E is the intensity, then -^ is proportional to E 1 , and 



* Ann. derPhys. [4], 7, 1902, p. 142. 



f A number of values will be found in Tables Annucttci Intcrrjationole* de 

 C nstantes I. 



J PJiys. Rev., 29, pp. 525 and 541. 



