On Rotatory Polarization in Liquids. 91 



mainly on the recent work of Franck and Hertz on the 

 •collisions between electrons and atoms. 



The results of: some measurements published previously 

 are considered in the light, of the theory. They seem in 

 general accordance with that work. They cannot be ex- 

 plained quantitatively at present, but it appears that they 

 make it necessary to believe that 8 rays are emitted from 

 ionized molecules of gases with velocities of the same order 

 of magnitude as those from metals. 



Leeds University, 

 August 1918. 



VII. An Experiment on Rotatory Polarization in Liquids. 

 By E. Talbot Paris and Alfred W. Porter, F.R.S.* 



THE experiment recorded here was carried out in order 

 to see if any evidence could be obtained of a directive 

 action on the molecules of an optically active substance in 

 solution by light passing through the solution. If such a 

 directive action existed (the light tending to orientate the 

 molecules in a certain direction relative to its direction of 

 propagation), it would be expected that when light first 

 passes into the liquid a small interval of time would be 

 required for the full rotatory power to be developed. In 

 this case, if an intermittent source of light were used, giving 

 flashes of sufficiently short duration, the rotation of the 

 plane of polarization would be less than that for a steady 

 source. In the experiment described below the rotation 

 produced. in a long tube of cane-sugar solution was measured, 

 first using a steady source of light (a mercury lamp), and 

 afterwards an intermittent one in the form of a platinum- 

 mercury spark. 



The arrangement of the apparatus is shown in fig. 1. 



Light from the source S (the mercury arc or spark as the 

 case may be) passes through the collimating lens L x and the 

 polarizing nicol N 2 . Sensitiveness is obtained by means of 

 a half-wave plate of mica at M, placed across the middle of 

 the field so as to give a "three-field" arrangement. The 

 sensitiveness can be altered by rotating Ni. The sugar 

 solution was contained in a long iron tube AB, fitted with 

 plate-glass ends. To prevent the solution becoming con- 

 taminated with the iron, this tube was galvanized ; the 

 inside was painted with photographic black to avoid 

 reflexions. The length of the solution in the tube was 

 .about 165 cm., and the rotation fior the green mercury line 



* Communicated bv the Authors. 



