518 Prof. E. Ketteler on the Dispersion of Light 



where G denotes a constant. If we further make the assumption 

 that the medium, though always composite, contains only a 

 single so-called active substance, there results approximately 

 by subtraction 



inasmuch as, on account of the smaliness of the difference of 

 the indices, their sum ?z x + n 2 may be put = 2n. Consequently, 

 further, 



n 1 — n 2 _ 1 _ 1 _ GL /2 ,^-j v 



x ~p p-p-u 2 ' • • * ^ l ' } 



But it is known that this expression is the rotation-angle ifr 

 proportional to the plane of polarization. If the division be 

 effected, since according to experiment I/ 2 is everywhere only 

 small, it will be finally written 



R S 



This relation differs from the dispersion-formula proposed 

 by Boltzmann, and recently verified on quartz by Soret and 

 Sarasin, only by containing I instead of \. 



Addendum. 



Since the printing of the preceding memoir for the present 

 year's Vevhandlungen des naturhistoHschen Vereins, I have 

 recognized that the problem treated therein is confined within 

 narrower limits than were necessary, and particularly that the 

 refraction-ratio of very large waves can differ from unity with- 

 out it being necessary to assume an essential alteration of 

 quality on the part of the internal aether. 



While, namely in § 2, for the reasons there alleged, a term 

 Kp was regarded as not indispensable, I now add it, and ac- 

 cordingly replace the system of the two equations I. and II. 

 by the following: — 



These being integrated by means of the expressions III., 

 instead of VI. the two following now join on to the relations 

 IV. and V.:— 



