340 Profs. H. da Bois and H. Rubens on Polarization of 



is similar throughout to our previous table viii. ; while, 

 however, the borrowed columns for 1'5 and 4*5 fi approxi- 

 mately prove the correctness of the above assertion, they 

 entirely fail to do so for the longer wave-lengths. 



On the whole the values here for n 2 at a given slit-width 

 b', in case of the finer grating Cu 1, are perceptibly larger 

 than those of the coarser Cu 2, so that, in the Hertzian sense, 

 the latter cet. par. polarizes more completely. 



§ 14. The coarser platinum grating Pt 2 (6 = 48 fi) fixed 

 normally gave, approximately, n 2 = 0*35 at 108 /jl, being 

 already a considerable though incomplete polarization. 



The iron grating Fe 1, w T hich was by no means perfect 

 even in 1893, has in the meantime, through rust, been com- 

 pletely spoiled, so that exact measurements w T ere impossible ; 

 this, however, also showed a strong Hertzian polarization, 

 wmicli at 108 fi is represented by a value of n 2 = 0'15. 



We finally investigated a very coarse silver grating 

 (a = 372/*, 6=186ya), which, by means of sulphuretted 

 hydrogen, was nearly completely blackened ; this gave the 

 following values for the long-waved radiations (about 314 yu,) 

 of the quartz-mercury lamp : — 



Q, = 0'30 

 Q p = 057 



Q m = 0-435 

 n 2 = 0-53, 



which are of the same order of magnitude as those given by 

 the copper grating Cu 2, which is about 3*5 times finer, at 

 \r=100fi (Table V.B). 



We previously mentioned (I. c. § 19) that a glass-fibre 

 grating show r s an exceptionally small effect in the visible 

 spectrum. We have, in our present contribution, abstained 

 from testing regular dielectric gratings. 



We have, on the other hand, tested a grating-like Auer 

 mantle with respect to its transmissibility (see § 16 below), 

 and found that in the Hertzian sense it polarizes with an 

 approximate value of it 2 = 0*55 at X= 100 //,. The material, 

 as is known, consists of a solid solution of 1 per cent, cerium 

 oxide in thorium oxide, and may indeed be considered, at 

 least at white heat, as a conductor. 



IV. Reflexion and Emission. 



§ 15. To determine the reflecting power of the silver 

 grating Ag 1 we carried out the following series of experi- 

 ments. The rays of the Invert mantle A were first of all 

 reflected by a plane silver surface at 45° ; it wns here 

 assumed that no loss occurred ; the reflected rays fell on the 



