756 Mr. G. H. Livens on the 



ionize the mercury molecules is that produced by their 

 passage through a potential difference of" 5 volts. This value 

 agrees well with that obtained by Franck and Hertz, viz. 

 4*9 volts, although their value approximates still closer to 

 the theoretical value 4*84 volts deduced from the quanta 

 theory. 



LXXX. On the Theories of Rotational Optical Activity. 



To the Editors of the Philosophical Magazine. 

 Gentlemen, — 



IN a very interesting letter addressed to you and published 

 in the August number of your magazine, Dr. G. Bruhat 

 discusses, only too briefly, some important experimental 

 results he has obtained in their bearing on the theories of 

 the rotational optical activity, and he concludes from them 

 that the original form of the theory given by Drude is more 

 satisfactory than the Lorentz form of the theory recently 

 developed in detail by me in several papers published by you. 

 Until the appearance of Dr. Bruhat' s work in full I am of 

 course unable to discuss the details of this letter, but I should 

 like to have an opportunity of justifying my own general 

 conclusion that Drude's theory is incomplete. 



I am extremely indebted to Dr. Bruhat for his trouble in 

 testing my formulae, and I am probably less surprised than 

 he is that they are not entirely satisfactory. I do not expect 

 any such simple explanation of such complicated phenomena 

 to be anything like complete, but I still maintain that the 

 apparent agreement at first sight of my formulae must be 

 more than accidental *. In any case, however, the theory 

 could hardly be less inadequate than Drude's theory. 



As I have already pointed out, Drude's theory furnishes 

 a formula which applies only in an extraordinarily small 

 number of the cases experimentally investigated, and no 

 satisfactory explanation has yet been discovered of the dis- 

 crepancies in all the other cases where it does not apply. 

 This is by itself almost sufficient evidence to condemn the 

 theory, but it is not all that can be advanced against it. We 



* A still more remarkable apparent agreement of my formulae is 

 provided in their application to explain the variation of the rotation of 

 a given acetic acid solution to which varying quantities of substances 

 such as sodium and potassium molybdate and wolfrarnate are added. 

 The curves representing these variations are more complete than any 

 others of a similar type, and their general form is identical with that 

 given by my theoretical formula. 



