Dispersion and Selective Absorption of Fuchsin. 217 



was made using various spectroscopes. The conclusion reached 

 was that a spectroscope of very low dispersive power having 

 a low-power eyei)iece was the most satisfactory, and therefore 

 that kind was used in this work. 



The bands as viewed with the system just described pre- 

 sented the appearance represented diagrammatically in PL IX. 

 fig. 3, and showed the anomalous dispersion of fuchsin in a 

 general w^ay at a glance. 



In discussing these bands ^ve will, for the sake o£ clearness, 

 call those bands affected by the fuchsin the fuchsin -bands 

 and the others the air-bands. The bands as shown in fig. 3 

 correspond to a film too thin to produce a w^hole band retar- 

 dation, so at the two points wdiere the tw^o sets of bands coincide 

 the index of refraction is unity. To determine the index of 

 refraction in any other part of the spectrum one of the 

 mirrors is moved parallel to itself by means of a screw till a 

 fuchsin-band appears in the required portion of the spectrum, 

 unless indeed one is already there. Now by means of a com- 

 pensator an air-band is brought into coincidence with the 

 fuchsin-band, and the amount of retardation introduced by 

 the compensator is equivalent to the retardation produced by 

 the fuchsin. In the case of a film thick enough to produce 

 more than one band displacement, the fraction of a band is 

 measured by means of the compensator, and the w4iole number 

 of bands added to this. 



It was found that when an attempt was made to produce 

 an arbitrary shift of a definite fraction of a band, either by 

 moving one of the mirrors or rotating a compensator, the 

 bands were disturbed by the mere touching of the mirror or 

 compensator. A very thin mica compensator, so thin that 

 it had to be rotated through about 20 degrees to produce 

 a shift of one band, was therefore used instead of the usual 

 thick glass ones, and with this no such disturbing effect was 

 observed since an infinitesimal movement of this would not 

 visibly affect the bands. To overcome the difficulty of mica 

 having three different indices of refraction, the piece of mica 

 w^as cut so that one of its axes of elasticity was the axis of 

 rotation., By using plane-polarized light, and making the 

 plane of polarization parallel to the axis of rotation of the 

 mica, the retardation introduced by the mica when at different 

 angles was always proportional to just the one index of re- 

 fraction. It was found incidentally that the introduction of 

 the nicol prism N, fig. 1^ made the bands much more distinct. 

 On rotating the nicol prism through a right angle, the bands 

 became more confused than with the nicol out. 



The compensator (C, fig. 1) was made by cutting a thin 



