THE ARGON FAMILY OF GASES 657 



1 gram of radium, o'6 cu. mm. at N.T.P., is almost exactly that 

 calculated, 0*585 cu. mm., on the assumption that one whole 

 molecule of emanation results in the disintegration of one mole- 

 cule of radium chloride. 



There is, therefore, something like direct proof that in the 

 decomposition of a single molecule of radium chloride " the 

 disintegration of a single atom of radium of mass 226," as I 

 should prefer to say, one molecule of radium emanation of 

 molecular weight 222 and one molecule of helium of molecular 

 weight 4 is produced. Must it not be admitted that this is dead 

 against Prof. Armstrong's theory that " proto-helium," the hypo- 

 thetical single constituent atom of the known polyatomic helium 

 molecule, is the wondrous material at the root of radioactivity? 

 The radium atom is not decomposing into proto-helium and 

 something else, but into whole molecules of helium and 

 emanation. The view, therefore, that it is the intense affinity of 

 proto-helium and the single atoms of the complex molecules of 

 the inert gases generally, which is the cause of the peculiarities 

 and extraordinary energy of the radioactive changes, does not, 

 in itself, help towards the explanation of the facts. 



The logic of events has manoeuvred Prof. Armstrong out of 

 his formerly impregnable position that, since the argon gases do 

 not form compounds or enter into chemical changes, their true 

 atomic weight must remain unknown. For radioactive changes 

 now give precisely the same kind of evidence as chemists rely 

 on in the determination of atomic weights, and this evidence 

 shows that the smallest particle taking part in a radioactive 

 change is a whole molecule of helium or of emanation. In 

 deciding atomic weights the periodic law is the final court of 

 appeal. Can the elements be properly accommodated therein, 

 or must the multiple of the equivalent be altered before a vacant 

 place can be found for them ? I need only mention the periodic 

 law generalisation, that in an a-ray change the group number of 

 the element is reduced by two, whereas in a /3-ray change the 

 group number of the element is increased by one. The non- 

 valent emanation results from radium in one a-ray change, from 

 thorium in three a-ray and two /3-ray changes, from actinium in 

 two a-ray and one /3-ray change, in full accord with the accepted 

 group numbers II, IV, and III for radium, thorium, and actinium 

 respectively. Moreover, if radium emanation had a polyatomic 

 molecule, since its molecular weight is 220, its atomic weight 



