﻿[ 1107 ] 



OIL Polarized Resonance Radiation of Mercury- 

 Vapour. ByR^Y. Wood*. 



[Plate IX.] 



IN my earlier papers on this subject I expressed the opinion 

 that the resonance radiation of mercury vapour showed no 

 traces of polarization, a somewhat surprising circumstance,, 

 in view of the fact that the resonance spectra of sodium and 

 iodine are strongly polarized. 



Lord Rayleigh published a short letter in 'Nature' several 

 years ago stating that strong polarization could be observed 

 in that portion of the excited vapour at some distance from 

 the window through which the stimulating radiation entered, 

 the percentage of polarization falling off' as the window was 

 approached. This indicated that the polarization was pro- 

 duced only by radiation not quite in synchronism with the 

 molecule, of which we have a very marked example in the 

 case of the light scattered by air and other gases, in which 

 the wave-length of the radiation is very far removed from 

 that of the absorption bands of the gas. 



During the past winter I have made a further examin- 

 ation of the subject, and have found that the radiation is 

 strongly polarized, but that the percentage of polarization 

 does not appear to depend upon the distance to which the 

 exciting radiation has penetrated. The mercury vapour 

 was contained in a highly exhausted quartz tube with 

 windows of the same substance ground and polished. The 

 exciting radiation was furnished by a quartz mercury arc, 

 the lower half of the tube being covered with cotton, over 

 which a stream of water flowed continuously, this arrange- 

 ment being necessary to prevent the reversal of the 2536 

 line which excites the vapour. The radiation from a wide 

 slit, placed close to the arc, was passed in turn through a 

 large quartz lens, a quartz prism of about 40°, and a second 

 lens : the prism was cut parallel to the optic axis of the 

 quartz, consequently two polarized spectra were obtained, 

 and a polarized monochromatic radiation of wave-length 2536 

 could be obtained from a second slit suitably located. The 

 plate in which this slit was cut was coated with barium platino- 

 cyanide to facilitate its adjustment in the ultra-violet regions 



* Communicated bv the Author. 



