472 Intrinsic Optical Activity of Isotropic Media. 

 which would give a relation of the same form as before 

 F = a l E + b 1 curl E, 



but in which 



^ e 2 /m be 2 /m 



s , 9 9 S ' . n t) 



nf—nr ,- ns — nr 



a^ 2 /?)i ' a«*/wi ' 



2 — « 2 7lr 2 — w' 



and, therefore, on such a theory 



^ fo 2 /m 



9 -^ 9 





ae 2 /m 



1-2 



This formula, although slightly more general than the 

 simple Drude formula, appears equally inadequate in both 

 the fundamental tests. 



3. Second modified theory. — In the complete form of the 

 theory due to Lorentz it is, however, shown that the force 

 on a contained electron must be represented in the general 

 •case by 



F=E+/(P)*, 



where f(P) denotes some undetermined vector function of 

 the polarization P of" the medium, the first approximation to 

 which is, however, aP, as above. In second order pheno- 

 mena of the kind under discussion other terms may arise, 

 and it is in these second order terms that Lorentz presumes 

 the asymmetric structure to present itself. 



It is therefore assumed that the next approximation in 

 isotropic optically active media is 



/(P) = aP + &curlP, 

 so that 



F=E + aP + 6curlP. 



This assumption appears reasonable, in so far as it merely 

 ascribes the observed chiral effect to the asymmetric struc- 

 ture of the various parts of the substance, without presuming 

 any result of the effect of such a structure on the surrounding 

 aether. 



This relation has been shown by the author to lead to a 



* < Versuch einer Theorie,' &c. (Leipzig-, 1906), pp. 78-81. 



