PHOSPHORESCENT MOLECULES 281 



this has been found to be the case with selenium. 

 The substance wGuld acquire something of the nature 

 of a spongy structure, and the corpuscles would 

 diffuse through it as hydrogen does through 

 platinum. Similar effects are, as we know, pro- 

 duced, though in a smaller degree, in carbon and 

 tellurium. A substance will emit these corpuscles 

 in a greater or less degree according to the freedom 

 with which the molecular agglomerations are formed 

 in it and the size which they assume. Thus those 

 substances ought to be most radio-active in the slow 

 or easily absorbed (/3) rays whose molecules have 

 the greatest tendency to concentrate round nuclei 

 and to form into groups. Thin films of metals 

 ought to be more transparent to cathode rays when 

 they are exposed to light than when not, but this I 

 have not been able so far to detect. 



Dependence of Phosphorescence upon the Free-paths 

 and Velocity of the Corpuscles 



When it is borne in mind that the phosphorescence 

 in gases takes place only at low pressures, when 

 collisions seldom occur in comparison with those at 

 higher pressures, we have a strong argument in 

 favour of the existence of large aggregates of 

 numerous atoms or molecules closely packed together 

 in a single group. The charge of electricity carried 

 by such molecules can, on account of their size, be 

 likewise greater than the unit charge, so much so 

 that they would, if they carried their maximum 



