The Interference of Natural Light on a Magnetic Field 7 



Upon emerging its plane has been rotated 45 °. The greater part 

 of that entering tube NS is unaffected until by its final transmis- 

 sion and reflection at ab a portion of it is polarized in the same 

 plane as that entering EW. If, however, the natural light in 

 traversing the coil NS was rotated through 45 ° the portion of it 

 which suffered polarization at ab would be resolvable into two 

 components at right angles, one of which would be at an angle of 

 90 to the rotated polarized light emerging from the tube EW 

 and so incapable of interference. The remaining portions of the 

 light entering the observing telescope composed of natural light 

 would have been rotated 45 ° in each coil and would also be in- 

 capable of interference. The- criterion, then, for the rotation of 

 natural light is, in the present form of experiment, the disappear- 

 ance of all the fringes previously observable. 



The question now arises as to whether or not the fringes 

 formed by the natural light are of sufficient intensity to be visible 

 if no rotation of natural light should occur. If the portion of 

 natural light entering the field of the telescope is sufficiently large 

 for the fringes formed by it to be observable when the other 

 portion of the light fails to interfere, the method used seems suf- 

 ficiently rigorous. The amount of unaffected light as found by 

 the formulae of Fresnel was some .26 of all the light capable of 

 forming interference bands. To find the minimum value that 

 this quantity could have and still be observable, a series of read- 

 ings was taken for plane polarized light in the following man- 

 ner : A field was produced in only one coil. The current was 

 then increased until the fringes vanished. The current was 

 then varied until they reappeared ; at this point the reading was 

 taken. The angle as calculated for the rotation corresponding to 

 the difference between the value of the current at which the 

 fringes were visible and that at which they were entirely extinct 

 gave a measure of the sensitiveness of the present arrangement. 

 It was found as a mean of several readings that fringes formed 

 by 0.06 of the total light -could be recognized. The criterion 

 then, as stated above, for the rotation of natural light was the 

 disappearance of all the fringes previously observable. 



151 



