MAGNUSSON — ANOMALOUS DISPERSION OF CYANIN. 279 



appearing in figure 3, where no color screen was used, is entirely 

 absent in figures 1 and 2, while the form of the curve is apparent 

 in all. Cyanin has a very strong absorption band in the yellow, 

 and to let any light through in this part of the spectrum very 

 thin prisms must be used. In figure 4 of the same plate is seen 

 a photograph of a crossed prism spectrum when the cyanin 

 prism was thin enough to let light of all wave lengths pass 

 through. The refracting angle was only 24.7", so the deviation 

 is small, but the continuity of the curve is evident. "No expo- 

 sure was made by the direct path to avoid overlapping of the two 

 spectra. A photograph of a similar continuous, crossed prism 

 spectrum for a cyanin prism of 2' 51.6" refracting single, is 

 seen in figure 5 of the same plate, in which the deviations are 

 proportionally larger than in figure 4, and the form of the curve 

 can easily be followed through the absorption band. In figure 

 6 the results are seen of four successive exposures through a 

 cyanin prism of 28'30" refracting angle. The great disper- 

 sive power of the dye is apparent and although the yellow is ab- 

 sent the rest of the curve agrees very well with the dispersion 

 calculated by direct spectrometer readings. The haze in the 

 green is due to diffraction ; the angle of the prism was so large 

 that only a very narrow edge was transparent to the green and 

 this produced the ordinary narrow aperture effect. 



(d) The Interferometer Method. — The previous work has for 

 its basis the deviation of a beam of light by a cyanin prism. 

 The index of refraction can also be found by measuring the re- 

 tardation produced by a thin film. To make uniform films of 

 cyanin the crystals were dissolved in absolute alcohol and the 

 resulting solution filtered. This solution was then kept in an 

 air-bath at a temperature between 35° and 40° C. Pieces of 

 plate glass, after being carefully cleaned and heated to the same 

 temperature, were dipped into the solution, and placed on edge 

 to dry in the air-bath. The thickness of the film depends on 

 the concentration of the solution, and as the alcohol evaporates 

 rapidly in the air-bath, films of continually increasing thick- 

 ness are obtained by dipping in fresh plates from time to time. 



