Light at certain Metal- Liquid Surfaces. 239 



It was feared at first that the attainment of the second 

 ohject would present formidable if not insuperable difficulties. 

 Drude has suggested the possibility of removing the film of 

 gas by heating *, but this would involve obvious complica- 

 tions and difficulties. As mentioned in the introduction, it 

 seems probable to expect that the gas film is introduced by 

 entrapment in the act of immersion. Hence it was decided 

 to produce the clean mercury surface while the mirror was 

 immersed. The method of Rontgen for producing the clean 

 surface lends itself very readily to this purpose ; and it 

 would seem from the results obtained that complete success 

 in this important objeet has been attained. The form which 

 the apparatus for producing the mirrors finally took is shown 

 in the figure. 



A glass tube of. about 4 cm. diameter was drawn down to 

 form a funnel-shaped vessel (A), into the neck of which was 











r , ■ 





\ A ! 



E 





i i 



; i 







1 i 

 i i 



b! c 







1 i 

 i 





WE 



sealed a small tube (B) having its end drawn out to a 

 capillary point and projecting into the funnel. This tube, 

 passing through a cork at the bottom of the cylindrical iron 

 vessel (C) which surrounded the funnel, was supported so 

 that the mouth of the funnel (ground to a plane) came 

 slightly above the top of the cylinder. Another tube (D) 

 * TVied. Ann. xxxix. 1891, p. 545. 



