S3G Profs. H. du Bois and H. Rubens on Polarization oj 



choosing of b' as single independent variable appeared, with 

 oar previous conditions of experiment, to be valid ; more 

 variables must, however, at present be taken into account. 



Fisr. 5. 



This also explains, among other things, why previously a 

 small inclination of the grating had scarcely any significance, 

 while now for values of /3=±5° considerable differences 

 appear in some cases in comparison with the normal position 

 £ = 0°. 



§ 12. Four types of curves may be distinguished : — 



Type I. (Cu 2, Ag 1 , Au 1, at 24 /*) : Q s shows one maximum, 

 Q^ two. In the case of Cu 2 the maxima certainly appear 

 only feebly indicated. 



Type II. (Cu 1, Pt 1, at 24/*; Cu 2 at 52/*) : Q, falls off 

 gradually, Q^, shows two maxima. 



Type III. (Agl, Au 1, at 52/*) : Q s decreases gradually, 

 Q p exhibits one well pronounced maximum. It is to be 

 remarked that Au 1 polarizes X with an uninclined grating, 

 while at about 5° a point of inversion exists and 1 1 polariza- 

 tion in the Hertzian sense first appears at larger inclinations; 

 this behaviour was checked several times. It has already 

 been remarked that the grating is made up of gilded silver 



wire. 



