ALTERATIONS IN THE DIELECTRIC 145 



the delay in phase of one ray relative to the other and showed * 

 that the delay is proportional to the square of the strength of 

 field as measured by a Thomson electrometer. 



Kerr investigated the effect for a large number of liquids f and 

 found that some, which he termed positive liquids, behave like 

 quartz with its axis parallel to the lines of force, or like glass 

 extended in a direction parallel to the lines of force. Of the 

 liquids in which he was able to measure the effect, carbon disulphide 

 shows it in the greatest degree. Paraffin oil, toluene, and benzene 

 are other positive liquids. Others, which Kerr termed negative 

 liquids, behave like Iceland spar with its axis parallel to the lines 

 of force, or like glass compressed in the direction of the lines of 

 force. Of these colza oil gives a strong effect ; others are olive oil 

 and seal oil. 



The effect may be expressed conveniently by the number of 

 wave-lengths difference between the two components in a length I 

 of the dielectric when the transverse electric field has intensity 

 E. Experiment shows that if S is the difference of path pursued 

 in the same time, and if X is the wave-length of the light used, the 

 number of wave lengths difference or S/\ is given by 



S/X = B/K- 



ulu-re B is a constant for a given substance and wave-length the 

 Kerr electric constant. The difference of phase is STT^/X, and 3 

 and X must both be measured in the liquid or both in air. It is 

 easilv shown that if // and //, are the refractive indices for the two 

 rays and X is the wave-length in air 



Mo - yu. = BX/E 2 . 



QuinekeJ was the first to obtain a value of B, working with 

 carbon disulphide. Since then various measurements for this 

 liquid have been made and the value of B for sodium light 

 appears to be very near to 3-1 x 10~ 7 . 



But the v&lue changes with the wave-length, increasing as X 

 diminishes. This was observed by Kerr, and he thought that /UL O /m t 

 varied inversely as the wave-length. Subsequent observers || have 

 not verified the simplicity of the relation though confirming the 

 general nature of Kerr's observation. Cotton and Mouton have 

 investigated the effect of change of temperature and have found 

 that B <i rapidly as the temperature rises (loc. cit.). 



Kerr observed in 1879 that though nitrobenzene is not a 



.}/.,. [5], 1880, vol. ix. p. 1.17. 

 t /'hi/. M'K.i. [5], vol. xiii. pp. 153 and 248. 

 Wied. Ann., XIX, 1883, p. 7 -".. 



A careful determination by McCorab is given in Phys. Rev., 29, 1909, p. 534. 

 References are given to earli> ; 



lilarkwell. /'/ \rad.ofArl -res, XLI, 1906, p. 147, McComb, 



/'"-. "'. Cotton and Mouton, Jmimnf *\, />/,,,.<;</>,<, 5th. ser., vol. I, 1911, p. 5. 



I 



