of Light at Certain lletal-Ziquid Surfaces. 



95 



stand was also provided with a rack and pinion to give the 

 vertical adjustment necessary to bring the reflecting surface to 

 the center of the spectrometer table. A reservoir of mercury 

 hung at a higher level than that of the funnel was connected 

 with the tube (B) by a flexible tube provided with a suitable 

 pinchcock. In order to secure a mirror sufficiently free from 

 vibration, it was found necessary to mount the stand in a Julius 

 suspension. 



The operation of producing a clean surface was simply to 

 allow the mercury to overflow the edge of the funnel rapidly for 



rf rr _ h 





! \ A ' 



1 \ ' 

 \ ' 



! \ fl / 



1 i 



; : ; 



E 





1 i 

 i i 



i i 



i i 



>B< c 



i i 

 i i 







: ! 

 ii ii 





tm 



a few seconds. If care is taken to make the overflow uniform 

 all around the edge (accomplished by adjusting the levelling 

 screws), surfaces of extreme cleanliness can easily be obtained. 

 The test of the cleanliness is the behavior of a drop of water 

 placed on the surface. In making the test it is essential that 

 the drop itself be clean. This is most easily done in practice by 

 dipping a piece of platinum wire (cleaned by heating) beneath 

 the surface of a beaker of distilled water, and removing the ad- 

 hering drop quickly to the mercury surface. If the surface is 

 really clean, the drop is immediately sketched out to the edge 

 in a film of invisible thickness. The appearance of a mercury 

 surface to the eye is no test at all of its cleanliness ; it may 

 seem to be a perfect mirror and yet the drop remain in an hemi- 



