72 



SCIENCE PROGRESS 



point on this line arc plotted the differences between the rota- 

 tions for other colours and the reference colour. It is then found 

 that the experimental points for the various colours of light 

 generally lie along straight lines which usually intersect one 

 another and the reference line in the same neighbourhood, but 

 never, as far as is known, at the same point. There thus 

 appears amongst optically active compounds a uniformity of 

 behaviour which was wholly unexpected, and wholly unpre- 

 dicted, a comparatively short time ago ; for it is found that 

 the data obtained for some active substances such as, for 

 example, methyl-Zc/Z-butyl-carbinol, in the homogeneous con- 



2" 4° 6" 8" 



Specific Rotation 



Fig. 7. 

 The upper inclined line is for mercuiy, violet ; the lower for sodium, yellow. 



dition at different temperatures, or in solution in different 

 solvents at various concentrations and temperatures, and — 

 what is even more remarkable — other derivatives of that sub- 

 stance, may all lie approximately upon the lines of one 

 characteristic diagram. 



Fig. 7 is the characteristic diagram for methyl-/rr/-butyl 

 carbinol, drawn from Pickard and Kenyon's data by the author's 

 method {J.C.S., 1014, 109, 1187). The points i, 2, 3, 4, and 6 

 represent the rotation values for 5 per cent, solutions in chloro- 

 form, carbon disulphide, pyridine, benzene, and ethyl alcohol 

 at the ordinary temperature, whilst 5, 7, 8, 9, and 10 are the data 

 for the homogeneous substance at the temperatures, 100°, 80°, 

 60°, 40°, and 20° respectively. The three lines along which these 



