1917] on The Complexity of the Chemical Elements 125 



which is laborious and difficult, and the radio-chemist had not then 

 fully appreciated the power of radioactive analysis in detecting a 

 very slight change in the proportions of two elements, one or both 

 of which were radioactive. The case is not at all like that of the 

 rare-earth group of elements, for example, in which the equivalent or 

 atomic weight is used as a guide to the progress of the separation. 

 Here the total difference in the equivalent of the completely 

 separated elements is only a very small percentage of the equivalent, 

 and the separation must already have proceeded a long way before 

 it can be ascertained. 



Human nature plays its part in scientific advances, and tlie 

 chemist is human like the rest. My own views on the matter 

 developed with some speed when, in 1010, I came across a new case 

 of this phenomenon. Trying to find out the chemical character of 

 mesothorium-I, which had been kept secret for technical reasons, I 

 found it to have precisely the same chemical character as radium, a 

 discovery which was made in the same year by Marckwald, and 

 actually first published by him. I delayed my publication some 

 months to complete a very careful fractional crystallisation of the 

 barium-radium-mesothorium-I chloride separated from thorianite. 

 Although a great number of fractionations were performed, and the 

 radium was enriched, with regard to the barium, several hundred 

 times, the ratio between the radium and mesothorium-I was, within 

 the very small margin of error possil^le in careful radioactive 

 measurements, not affected by the process. I felt justified in con- 

 cluding from this case, and its analogy with the several other similar 

 cases then known, that radium and mesothorium-I were non-separable 

 by chemical processes, and had a chemical character not merely like, 

 but identical. It followed that some of the common elements might 

 similarly be mixtures of chemically identical elements. In the cases 

 cited, the non-separable pairs differ in atomic weight from 2 to 4 

 units. Hence the lack of any regular numerical relationships 

 between the atomic weights would on this view follow naturally. 

 (Trans. Chem. Soc. 1911, xcix. 72.) This idea was elaborated in the 

 Chemical Society's Annual Report on Radioactivity for 1910, in 

 the concluding section summing up the position at that time. This 

 was I think the beginning of the conception of different elements 

 identical chemically, which later came to be termed "isotopes," 

 though it is sometimes attributed to K. Fajans, whose valuable con- 

 tributions to radioactivity had not at that date commenced, and 

 whose first contribution to this subject did not appear till 1918. 



In the six or seven years that have elapsed the view has received 

 complete vindication. Really, three distinct lines of advance con- 

 verged to a common conclusion, and, so far as is possible, these may 

 be disentangled. First, there has been the exact chemical charac- 

 terisation from the new point of view of every one of the members 

 of the three disintegration series, with lives over one minute. 



