chapter T if teen 



THE EVAPORATION, CONDENSATION AND 



REFLECTION OF MOLECULES AND THE 



MECHANISM OF ADSORPTION 



In a recent paper R. W. Wood ^ describes some remarkable experiments 

 in which a stream of mercury atoms is caused to impinge upon a plate of 

 glass held at a definite temperature. With the plate at liquid air temperature 

 all the mercury atoms are condensed on the plate, whereas with the plate 

 at room temperature all the atoms appear to be dififusely reflected. Wood 

 seems to consider that this is a real case of reflection. 



There is, however, another way of interpreting the experiments, which, 

 I think, is more in harmony with other facts than that suggested by Wood. 



We may, for example, consider that all the atoms of mercury which 

 strike the plate condense no matter what the temperature of the plate. 

 When the plate is at the higher temperatures the condensed atoms may 

 reevaporate again so rapidly that the surface remains practically free from 

 mercury. At first sight it might appear that there is no essential difi^erence 

 between this reevaporation and a true reflection, but more careful con- 

 sideration shows that the two phenomena are quite distinct. The difiference 

 would be manifest if the so-called reflection were studied at intermediate 

 temperatures. In the case of reflection the number of atoms reflected would 

 always be proportional to the number striking the surface, whereas, accord- 

 ing to the reevaporation theory, the number leaving the surface can never 

 exceed the normal rate at which mercury evaporates into a perfect vacuum. 



Wood states that he intends to undertake experiments at intermediate 

 temperatures, and in view of this I should be loath to take up a discussion 

 of the matter at present if it were not for the fact that much of my work 

 during the last few years has had a very direct bearing on the question of 

 the condensation versus the reflection of gas molecules.^ 



* Phil. Mag., 30, 300, 1915. 



^ In this connection, for the sake of historical accuracy, attention may be called 

 to the fact that the rectiHnear propagation of molecules in vacuo, which Wood and 

 Dunoyer (Comptes Rendus, 153, 593, 191 1) have described, has long been familiar to 

 incandescent lamp manufacturers not only in the case of tungsten lamps, but also with 

 carbon lamps. Professor W. A. Anthony, in a paper read before the American Insti- 



295 



