Magneto-optic Phenomenon. 



301 



tube. The normals of the mirrors were nearly in the same 

 plane, and made angles of nearly 90° with each other *. 

 The polarizer was rotated by means of a long arm. In the 

 analyser a similar arm is fixed to a ring surrounding the 

 analyser-circle and capable of being clamped by a screw. 

 This ring carries the mirror used in reading the rotations. 

 Since only one scale and two telescopes, placed near together, 

 were used for these measurements, there was added, in case 

 the angle of incidence was not too great or too small, an 

 auxiliary mirror, in order to permit of the analyser-rotations 

 being measured in this manner. The scales were divided in 

 millimetres, and 1 minute of angle corresponded to 1*4 to 1*8 

 millim.f 



Fig. 3. 



The constancy of the results obtained by this method of 

 reading was also very satisfactory : it has been used since 1886. 



10. In these experiments sunlight or electric light was 

 used. The latter was furnished by a Siemens lamp which 



* This arrangement was chosen because the polarizer is only used in 

 the positions in which the plane of polarization is nearly parallel or 

 perpendicular to the plane of incidence. 



f In order to reduce the scale-readings to minutes, the readings are 

 taken with the Nicols adjusted to two consecutive degrees on the divided 

 circle, 



