II. — On the Effect of a Magnetic Field on the Interference of 



Natural Light 



BY JOHN MILLS 



The fundamental principles stated by Fresnel and Arago of the 

 interference phenomena of plane polarized light are well enough 

 known. In the case of natural light our conception is that it is 

 an elliptically polarized vibration, where the major axes of the 

 ellipse assume in a very short space of time all the possible azi- 

 muths, and the eccentricity of the ellipse is at the same time under- 

 going continuous changes. We further conceive of this elliptical 

 vibration as capable of resolution into two straight line compo- 

 nents at right angles to each other and in no way correlated. In 

 the interference of natural light under ordinary conditions we 

 may look upon the phenomenon as produced by the superposition 

 upon an ether particle of two sets of the component parts of an 

 ellipse, where the two components of the same kind are acting in 

 directions parallel to each other. The rotation of the ellipse as 

 a whole through 90 would bring like components into a position 

 at right angles to each other, and interference phenomena should 

 then disappear. 



In an article by L. Sohncke 1 the statement is made that Abbe 

 had noticed in his investigations that natural light was capable of 

 rotation by natural rotatory substances in a way entirely analogous 

 to the rotation of plane polarized light. In the path of one ray 

 of an interfering system he had placed a piece of right-handed 

 quartz capable of producing a rotation in the case of plane po- 

 larized light of 45 , in the path of the other ray a similar piece 

 of left-handed. The rotation thus brought about of the two 

 beams through an angle of 90 with respect to each other caused 



Sohncke, Uber electroma%netische Drehung naturalischen Lichts, Wied. 

 Ann. 27, p. 206, 1886. 



University Studies, Vol. IV, No. 2, April, 1904. 



145 



