and Surface- Colour of Nitroso-dimetliyl-aniUne, 101 



substances of low dispersion, the different colours are polarized 

 at nearly the same angle, i. e. very little colour-effect is ol)- 

 served when the reflected light is examined with the nicol. 



^^ 



3 



Fia-. 1. 





C 



In these cases the dark band to which I have referred is'so 

 broad as to occupy practically the entire visible spectrum. 

 In the case of a substance with as high a dispersion as that 

 of the nitroso, the angle of maximum polarization is quite 

 different for the different colours, so that the dispersion mav 

 be determined by observing the position of the centre of the 

 dark band and the angle of incidence. It was found that 

 values agreeing very closely with those obtained with the 

 prisms could be obtained by this method in the yellow, green, 

 and greenish-blue portions of the spectrum. On attempting 

 to drive the band throuo-h the orange and into the red, it 

 broadened so much that it was quite impossible to locate its 

 centre with any precision. This was of course due to the 

 fact that the dispersion in the red and orange is not sufficient 

 to make the method very accurate in this part of the spectrum. 

 The dispersion of selenium was also determined in this manner, 

 and found to agree very well with the results obtained with 

 prisms. To apply the method to the ultra-violet the observa- 

 tions were made photographically with a small quartz spectro- 

 graph made by Fuess. This instrument was furnished with 

 a Rochon prism mounted immediately behind the quartz 

 coUimation-lens. This prism, as furnished by the maker of 

 the instrument, refused to transmit the ultra-violet, and I 

 found that the two halves had been cemented together with 

 balsam, which I replaced with glycerine. Some preliminary 

 experiments were made with selenium mirrors, as they were 

 easier to handle than the fluid cell. The lio-ht from a cadmium 



