280 



BULLETIN OF THE UNIVERSITY OP WISCONSIN. 



To secure good films it is important to use "absolute" alcohol 

 and to keep .the temperature inside the given limits throughout 

 the operation. The retardation produced by the film was meas- 

 ured with a Michelson interferometer (F of figure 7), photo- 

 graphs being made of the displaced fringes. The interfer- 

 ometer w r as illuminated with monochromatic light obtained from 

 the direct vision spectroscope, C, as described in the first part 

 of this paper. From one-half of one of the coated glass plates 

 the cyanin film was carefully removed, the remainder of the film 

 having a straight edge down the center of the plate. This plate, 

 H, was placed between the movable mirror, M, of the interfer- 

 ometer and the half silvered plate, P, with the straight edge of 

 the cyanin film vertical, the instrument having been previously 

 adjusted for horizontal fringes. A plate of clear glass, K, of 

 thickness equal to that of the coated plate was placed in front of 

 the other mirror, L, of the interferometer as a compensator. 



M 

 H 



CO" • 



\M 



Fig. 7. 



The fringes formed by the interference of rays coming through 

 the cyanin were displaced relatively to those formed by rays 

 coming through the clear portion of the plate. In order to fa- 

 cilitate the determination of the displacement the, fringes were 

 photographed by means of a camera at Q. The camera was so 

 arranged that by moving the back portion slightly, successive ex- 

 posures could be made on the same plate, showing the relative 

 displacements of the fringes for all values of /* from the ex- 

 treme red to the ultra violet. 



Since the retardation is a function of the wave length, the dis- 



