August i i, 1923] 



NA TURE 



21 1 



that the expulsion of the a-particle caused the element 

 expelling it to move from the place it occupied in the 

 Periodic Table to the next place but one to it in the 

 direction of diminishing mass. 



At this time the chemistry of the post-emanation 

 members had scarcely been studied, though von Lerch, 

 from electrochemical researches, had put forward the 

 rule that the successive products are each electro- 

 chemically " nobler " than the last, a rule which 

 describes well enough the electrochemical behaviour 

 of the first three — the x\ to C members, as they are 

 called. Then, as a result of the experiments of Schrader 

 and Russell, it was found that their volatility was 

 much affected by chemical treatment and by the 

 atmosphere in which they were volatilised. Thus, in 

 hydrogen, radium C volatilises at as low a temperature 

 as 360° C, though, in air, a temperature of 1200° 

 is necessary. This clearly indicated the possibility 

 that even these excessively ephemeral elements have 

 a definite chemical character. Hevesy showed, by 

 electrochemical methods, that the three B-members 

 are identical in properties among themselves, and 

 also the three C-members. 



But the work which, more than anything else, 

 served to reveal, as in a flash, the simple and sweeping 

 generalisation which covers the evolution of the radio- 

 active elements was that of A. Fleck in my laboratory 

 in Glasgow. He studied the chemistry of the various 

 members, still uncharacterised, from the definite point 

 of view of ascertaining to which element each most 

 closely approximated in chemical character, and then 

 whether it was separable from that element or not. 

 In addition to confirming more rigorously many 

 conclusions already reached, he proved that meso- 

 thorium 2 was non-separable from actinium, the three 

 B-members from lead, like radium D, and the three 

 C-members and radium E from bismuth. 



Hevesy and Russell — the first with regard to the 

 valency of the radio-elements and the second with 

 regard to the positions they occupy in the Periodic 

 Table — published early in 1913 statements of the full 

 law underlying radioactive evolution, but only in 

 part correct. Within a month K. Fajans, in Carlsruhe, 

 published the scheme correct and complete, including 

 the complicated branchings that occur at the C-members. 

 In a paper, amplifying and amending Russell's scheme, 

 I arrived independently at the same scheme as Fajans. 

 Each a-ray expelled causes a shift of two places in 

 the Periodic Table in the direction of diminishing 

 mass, and each fS-ra,y a shift of one place in the 

 opposite direction. In its present form the scheme is 

 shown in Fig. i. The chief uncertainty remaining is 

 whether the actinium branch starts from uranium II, 

 as shown in the figure for convenience, or from 

 uranium I, or even from a third independent isotope 

 of uranium. So that the atomic weights shown for 

 the actinium series are purely provisional. 



By the consistent application of the two rules 

 mentioned, the members found to be non-separable 

 from one another fall in the same place in the Periodic 

 Table. The chemical character has nothing to do 

 with the radioactivity, nor with the series to which 

 the element belongs, nor with its atomic weight. It 

 depends upon a number, now called the atomic number, 

 shown at the top of the place in the figure. 



NO. 2806, VOL. 112] 



Before passing on to this, the chief practical con- 

 sequences of the generalisation may be briefly 

 enumerated. 



(i) Of the members still uncharacterised, the A and 

 C members must be the isotopes of polonium (radium F) 

 and radium Cg (now called radium C"), actinium D 

 and thorium D must be isotopes of thallium. Fleck 

 at once verified these predictions as regards radium A, 

 actinium D, and thorium D. 



(2) Uranium X, Hke mesothorium, must consist of 

 two successive ^-ray - giving products, intermediate 

 between the two uraniums. Fajans and Gohring at 

 once succeeded in separating from uranium X a 

 very short-lived product, uranium Xg, giving the more 

 penetrating of the two types of /3-ray expelled, the 

 uranium X^ giving the less penetrating ^-rays. 



(3) The parent of actinium in the IlIrd family 

 must be an isotope of radium, if actinium is formed 

 in a /^-ray change — a conclusion I at once experi- 

 mentally disproved — or it must be an isotope of 

 uranium Xg, in the Vth family, if actinium is formed 

 in an a-ray change. This was proved by Cranston 

 and myself, and the name " eka-tantalum " given to 

 the new element, and by Hahn and Meitner, who 

 named it protoactinium. It is linked to uranium 

 through uranium Y, a branch member discovered 

 by Antonoff in 191 1, and suspected to be in the 

 actinium series. 



Protoactinium, to give it Hahn and Meitner's name, 

 has been shown by them to give a-rays and to be 

 chemically so like tantalum that hitherto it has not 

 been separated from it. Its period is about 17,000 

 years, and from this it may be calculated that there 

 is about one-fifth as much of it by weight in minerals 

 as there is of radium. This may be sufficient to enable 

 it to be isolated, and for its spectrum, atomic weight, 

 and chemical character to be ascertained. The branch 



series runs 



u- 



VI 



-UY- 

 IV 



V 



►Ac— 

 III 



-RaAc- 

 IV 



"AcX- 

 II 



", etc. 



in which the figures in the second line refer to the 

 family in the Periodic Table to which the element 

 belongs. 



(4) All the ultimate products in all branches are 

 isotopes of lead. The atomic weight of the two 

 products of thorium are both 208, and of the major 

 branch of uranium 206. As is well known, this had 

 only to be tested to be proved correct. The atomic 

 weight of the lead from the purest thorium minerals 

 is as high as 207*9, ^"^ o^ ^^^^ irom the purest uranium 

 minerals 2o6'o. The spectra of these isotopes, but 

 for the infinitesimal difference already alluded to, are 

 identical. But the densities are proportional to their 

 atomic weights. This was a very simple prediction 

 I made, before testing it, from the theoretical views 

 about to be dealt with. 



The Theoretical Interpretation of Isotopes. 



The results on the theoretical side were no less 

 definite and important, and isotopes found a ready 

 explanation on the nuclear theory of atomic structure 

 put forward in a tentative form by Rutherford in 19 11. 

 This theory accounted for the large angles through 

 which occasional a-particles were deflected in their 



