l28 Vibration of Polarized Light in the Plane of Polarization. 



section is vertical, we find that the image which was hitherto 

 the feebler, remains, and that the image which was previously 

 the stronger disappears ; that is, a Nicol's prism, whose principal 

 section stands vertical, transmits only horizontal vibrations, or the 

 vibrations of light occur in the plane of polarization. 



4. Stokes found, from measurements of the angles y and y,, 

 the opposite result. I endeavoured therefore to measure also 

 the inclination of the plane of polarization towards the vertical 

 with light diffracted by the lampblack grating; but I found 

 everywhere and decidedly that this angle of the plane of polari- 

 zation of the diffracted light with the vertical is the smaller, the 

 larger is the angle of diffraction ; and that this angle is always 

 smaller than in the incident light ; so that it follows from this, 

 that light vibrates in the plane of polarization, and not at right 

 angles to it. The results of my measurements, compared with 

 the formula (1), are contained in the following Table : — 



We see that the angles which I obtained for the plane of po- 

 larization in diffracted light are decidedly smaller than those in 

 incident light, so that there can be no doubt about the directions 

 of the vibrations of the light, whether in, or at right angles to, 

 the plane of polax'ization. But we see that the formula gives by 

 no means with accuracy the angle of the direction of vibration in 

 the incident light, so that there is still something to be investi- 

 gated here. At first I believed that I could explain these devia- 

 tions, from the fact that longitudinal waves, which the incident 

 Nicol transmits to the diffraction grating, would give a transverse 

 component in diffracted light. But I convinced myself by expe- 

 riments, that I could find nothing perceptible, at least, of such 

 light resulting from longitudinal vibrations by diffraction, at 

 any rate with my apparatus; from which it follows, that here 

 also no such longitudinal vibrations are transmitted through 

 the Nicol. This has been hitherto generally assumed, but, as 

 far as 1 know, never been proved. To this point I hope to return 

 at another time. 



Stokes arrived at his erroneoua results from investigations 



« 



