64 BELL SYSTEM TECHNICAL JOURNAL 



of a radioactive atom does not proceed continually downward to lower 

 and ever lower atomic numbers, but is interrupted by occasional partial 

 recoveries. Whenever there are three consecutive transmutations of 

 which two involve the emission of beta-rays and one that of alpha-rays, 

 the element in which the third ends has the same atomic number as the 

 element from which the first originates. In each of the three lines of 

 descent made visible in Fig. 1 there are instances of this; the one 

 including radium for instance touches three times at atomic number 82, 

 the one commencing with thorium twice. In the sense in which I have 

 thus far used the word "element," the element 84 recurs three times in 

 one series and twice in the other. Here is an ambiguity which the 

 time has come to dispel. 



The ambiguity in the use of the term element is a question of words, 

 but not wholly a linguistic, much less a trivial one; it is such a question 

 as arises when a field of knowledge is expanded and enriched to such an 

 extent that its old vocabulary ceases to be adequate. This particular 

 question arose after the discovery that certain substances differing in 

 radioactivity are very much alike in their chemical properties — another 

 of the facts which the atom-model is especially adapted to explain. 



Consider, for an example, the three elements radium B and radium 

 D and radium G, which lie upon the same line of descent, the "radium 

 series," The first transmutes itself into the second, and the second 

 transmutes itself into the third, each in a three-stage process involving 

 the departures of two electrons and an alpha-particle (with the order 

 of their exits we are not now concerned) from the nucleus. The mass 

 of the third is four units less than that of the second and eight units less 

 than that of the first; in radioactivity also they differ. But all three 

 are alike in nuclear charge, and hence in the size and presumably in 

 the arrangement of their circumnuclear families of electrons; and 

 hence the presumption arises, that in their physical and chemical 

 properties apart from radioactivity and mass they should be quite 

 alike. 



This presumption about the chemical properties is confirmed by the 

 fact that RaB and RaD and RaG cannot be separated from one another 

 by chemical means after they are once mixed. In general, whenever 

 two of these "radioelements" coinciding in atomic number are sub- 

 jected to any of the very considerable variety of agents in the chemists' 

 armory, they respond in so nearly, if not exactly, the same way that 

 there is no method known for taking one and leaving the other. Crys- 

 tallization out of a mixture of salts of two such elements merely pro- 

 duces crystals containing the two salts in the same proportion as the 

 liquid; sublimation merely produces a deposit containing the elements 



