and Reflexion of Gas Molecules. 



365 



During the past winter the work has been continued, 

 cadmium being used in place of the mercury. Permanent 

 films can be obtained in this way, and measurements made 

 wiLh the photometer. The chief point of interest in con- 

 nexion with the work, however, has been the discovery of the 

 fact that there is a sort of critical temperature for each 

 substance, below which condensation appears to take place 

 at the first collision, in other words the chance that reflexion 

 takes place is zero. This is true, however, only in case the 

 wall is of some substance other than the metal in question. 

 The critical temperature is surprisingly low. For mercury 

 it is in the vicinity of —140°, for cadmium about — 90°, and 

 for iodine about —60°. 



We will take up first, however, the law of reflexion. 



In my first communication, I drew attention to the sup- 

 posed circumstance that there was practically no reflexion for 

 directions which made small angles with the reflecting surface; 

 in other words, there was a zone of clear glass just above the 

 circle cut by the plane of the reflecting surface in the wall 

 of the bulb. Above this clear zone the gradually increasing 

 thickness of the deposit indicated that the law of reflexion 

 approximated the cosine law, and was independent of the 

 angle of incidence. 



As soon as it was found that permanent films of cadmium 

 could be obtained, quantitative experiments were commenced. 



Fig. 1. 



% Pump ► 



The form of tube used is indicated by fig. 1 (a) and (b). 

 (a) is the type used for demonstrating the one-dimensional 

 motion of the cadmium molecules, to which attention was 

 first drawn by L. Dunoyer in the case of sodium. The tube 

 must be kept in communication with a Gaede pump during 

 the experiment as an extremely high vacuum is necessary, 

 and the cadmium is heated by a gas-flame 3 or 4 mm. high 

 burning at the tip of a glass tube drawn down to a fine 

 capillary. 



