1917] on The Complexity of the Chemical Elements 127 



pairs or groups of elements suffers perhaps from being in a negative 

 form. It looks too much like a mere negative result, a failure, but 

 in reality it is one of the most sweeping positive generalisations that 

 oould be made. Ionium we say is non-separable from thorium, but 

 every chemist knows thorium is readily separated from every other 

 known element. Hence, one now knows every detail of the chemistry 

 of the vast majority of these new radio-elements by proxy, even when 

 their life is to be measured in minutes or seconds, as completely as if 

 they were obtainable, like thorium is, by the ton. The difference it 

 makes can only be appreciated by those who have lived through 

 earlier days, when, in some cases dealing with the separation of 

 radio-consUtuents from complex minerals, after every chemical 

 separation one took the separated parts to the electroscope to find 

 out where the desired constituent was. 



As the evidence accumulated that we had to deal here with some- 

 thing new and fundamental, the question naturally arose whether the 

 spectrum of isotopes would be the same. The spectrum is known, 

 like the chemical character, to be an electronic rather than mass 

 phenomenon, and it was to be expected that the identity should 

 extend to the spectrum. The question has been tested very 

 thoroughly by A. S. Russell and R. Rossi in this country, and by 

 the Austrian workers at the Radium Institnt of Vienna, for ionium 

 and thorium, and by various workers for the various isotopes of lead. 

 Xo certain difference has been found, and it may be concluded that 

 the spectra of isotopes are identical. This identity probably extends 

 to the X-ray spectra, Rutherford and Andrada having shown that 

 the spectrum of the y-rays of radium-B is identical with the X-ray 

 spectrum of its isotope, lead. 



The Periodic Law axd Radioactive Chaxge. 



The second line of advance interprets the Periodic Law. It 

 began in 1911 with the observation that the product of an a-ray 

 change always occupied a place in the Periodic Table two places 

 removed from the parent in the direction of diminishing mass, and 

 that in subsequent changes where a-rays are not expelled the product 

 frequently reverts in chemical character to that of the parent, though 

 its atomic weight is reduced 1 units by the loss of the a-particle, 

 making the passage across the table curiously alternating. Thus the 

 product of radium (Group II) by an a-ray change is the emanation 

 in the zero group, of ionium (Group lY), radium, and so on, while, 

 in the thorium series, thorium (Group IV) produces by an a-ray 

 change mesothorium-I (Group II), which, in subsequent changes in 

 which no a-rays are expelled, yields radio-thorium, back in Group IV 

 again. (Chemistry of the Radio-Elements, p. 29, 1st Edition, 1911.) 

 Xothing at that time could be said about ^-ray changes. The 

 products were for the most part very short-lived and imperfectly 



