486 M. Verdet on the Optical Properties developed in 



a screw allowed the telescope to be turned very slowly. The 

 precision which it has been possible to give to the experiments 

 has shown that none of those precautions were superfluous. 

 The telescope was also capable of two rotatory movements, a 

 vertical and a horizontal one, so that it was always possible to 

 bring the axis into the direction of the pencil of light. The 

 analyser was sometimes a birefracting prism of Iceland spar 

 achromatized for the ordinary ray, sometimes one of Rochon's 

 prisms. The telescope being adjusted so as to perceive clearly 

 the image on the diaphragm, two images could be seen through 

 the spar prism, and four through the Rochon prism, that is to 

 say, two principal images, and two secondary images due to the 

 imperfection of the construction. Having conveniently chosen 

 the distance of the telescope and the diameter of the diaphragm, 

 it could be so arranged that only the image whose variations 

 were to be examined remained in the field of vision, an indispen- 

 sable condition for the exactitude of the experiment. I took a 

 diaphragm 3 millims. in diameter, and placed the lunette at a 

 distance of m, 80. 



I used, besides, two methods which have given me results 

 perfectly accordant ; sometimes I employed homogeneous light, 

 and determined the position of the plane of polarization by ob- 

 serving the extinction of the extraordinary image ; sometimes I 

 have employed white light, and have had recourse to the obser- 

 vation of the tint of passage. 



In experimenting on homogeneous light, I neither made use 

 of red glass, which would have given me too small deviations*, 

 nor of the monochromatic lamp, which would have furnished too 

 feeble a light. I employed a solution of sulphate of copper in 

 carbonate of ammonia, which, when its thickness amounts to 

 some centimetres, only allows passage to the violet rays near to 

 the ray G. If it be requisite that the light employed should 

 have a sufficient intensity, the use of solar light is indispensable. 

 Rotations of the plane of polarization almost twice as great as 

 when operating with the tint of passage, are obtained by this 

 mode of proceeding ; but the estimation of the extinction is not 

 always made with much exactitude, and depends in a singular 

 manner upon the sensitiveness of the eye. I determined the 

 two positions of the analyser which caused the image of the dia- 

 phragm to disappear by two movements in opposite directions ; 

 but as between the two disappearances it was necessary to illu- 

 minate the graduation in order to read off the position of the 



* We know, in fact, that the rotation of the plane of polarization due to 

 the magnetic action varies with the length of an undulation, almost in the 

 same manner as the rotation produced by quartz and by organic liquids, 

 It is hence the smallest possible for red rays. 



