486 Mr. B. V. Hill on Accidental 



which were two parallel cylinders, external to one another. 

 Between these the light passed. While this apparatus required 

 a larger volume of liquid, local heating due to friction was 

 reduced to a minimum, so that a half-shade polariscope could 

 be used, and a sensibility one to two hundred times greater 

 could be obtained. His observations were confined mainly 

 to the crystalloids water and hyposulphite of soda, his sen- 

 sibility for the former being '000025 \ and -00005 \ for the 

 latter, but no effect could be detected. 



The present research was undertaken for the study of this 

 phenomenon in dilute solutions of the colloids. Gelatine was 

 used principally, as representative of the gelatinizing colloids. 

 Umlauf has given some results for gelatine, but most of his 

 experiments were made near the temperature 33° C, which is 

 the melting-point of jelly. There was a large discrepancy 

 between some of his results, which left doubt as to the laws 

 of this substance. 



Apparatus. 



The apparatus used in my work is the same as that employed 

 by Mr. Almy. The solutions studied possessed a rotary 

 power of their own, which necessitated using homogeneous 

 light in order to obtain a perfect match, and also to measure 

 the double refraction, which is a function of the wave-length. 

 The greatest sensibility was obtained by using sunlight 

 passed through absorbing solutions, made according to the 

 formulas of Landholt*. In this way a red was obtained 

 between 7 L8 /Li/* to 639 /x/a, which gave a bright, uniformly- 

 coloured field. The green was bright and sufficiently homo- 

 geneous, but absorbed so much light that the sensibility was 

 greatly reduced. Moreover, the eye was not so sensitive to 

 changes of intensity in the green and blue as in the red. The 

 data given below are for the red. With a clear solution, 

 settings could be made to from o, 003 to o, 005. This gives 

 a sensibility of from '000017 X to -000028 X. With water a 

 sensibility of '000012 \ was obtained. 



Observations. 



In the following tables and diagrams, the number of revolu- 

 tions of the cylinder per second is given in the column V. 

 AX, is the relative retardation of the one ray: the positive 

 sign indicates a right-handed rotation of the plane of polariza- 

 tion, and a negative sign the opposite effect. T is the 

 temperature. 



The first three tables show the effect of speed on the double 

 refraction, where the fluid used was a solution of gum-arabic, 

 * Ber. d. Akad, Wiss. Berfin, 1894. 



