62 BELL SYSTEM TECHNICAL JOURNAL 



due to the difference between the two types of emission. Alpha- 

 particle emission is violent and unique; positively-charged particles 

 moving with a speed like theirs are not produced in any other way 

 known to man. Beta-particle emission is considerably less violent, 

 and there are so many known processes for producing fast-flying 

 electrons that one must always keep in mind the possibility that some 

 of the electrons proceeding from radioelements may be due to one or 

 another of these; in fact, many certainly are. Perhaps the best way to 

 state the evidence is this: every radioelement which does not emit 

 alpha-particles transmutes itself into an element lying one step farther 

 up the procession,'' and all but one (actinium) of these elements is 

 known to emit electrons, all of which agrees with the assumption that 

 in each of these transmutations one electron is extruded from each 

 participating nucleus. Stated thus, it may not sound very convincing ; 

 but if the second part of the Fajans-Soddy law were not true, we should 

 hardly have failed thus far to find something definitely inconsistent 

 with it. 



Were gamma-rays without an accompanying beta-particle or alpha- 

 particle to be emitted from a nucleus we could scarcely call the result 

 a transmutation, since it would not affect the nuclear charge nor the 

 electron-family of the atom. There is no reason for denying that this 

 might happen; but I am not aware that it is known ever to happen, 

 except in cases of nuclei which have just previously undergone a 

 transmutation — cases which we shall eventually examine. 



If now each radioelement is passing over into another element, one 

 step before it or two steps behind it in the procession according as it 

 emits beta-rays or alpha-rays — then it must be possible to draw up 

 genealogies of radioelements, series of elements of which each member 

 is transmuted out of the foregoing and transmutes itself into the follow- 

 ing one. All of the known radioelements fall into one or another of 

 several such series. To represent all these relations, and one more, it is 

 convenient and suitable to draw a graph in which the atomic numbers 

 of the elements are laid off horizontally, and their atomic weights are 

 laid off vertically. Each element is represented by a point upon this 

 graph ; when the element transmutes itself it moves to another point, 

 two units to the left if an alpha-particle is emitted and one to the 

 right if the change is a beta-ray change. Now the emission of an 

 alpha-particle involves the departure of four units of mass from the 

 nucleus which it leaves; the loss of an electron however involves a loss 



^ More precisely, into an element having the chemical features distinguishing the 

 column of the Periodic Table containing the element lying one step farther up the pro- 

 cession than the original one. 



