204 
observation that a substance is present in commercial 
uranium salts capable of generating radium and not 
removed from it by the barium sulphate method used 
first for separating the radium, but separated, at 
least mainly, by the ether method. 
In the meantime a cognate discovery of first im- | 
portance was made by Boltwood, in America, who 
120 
ROM SIMPURIMIES 
SEPARATED FROM 
eo +. 
= 
—> QUANTITY OF RADIUM 
ae Us (YEARS 
Oo 2 ae 
Fic. x. 
proved that actinium preparations obtained from 
uranium minerals, and initially free from radium, 
grow a fresh crop of radium with lapse of time. 
The growth is not by any means a very minute one 
as in my experiments, in which the growth can 
only be put beyond all doubt after the lapse of years. | 
The growth of radium from constituents 
| 
NATURE 
PRODUCTION OF RADIUM 
[APRIL 25, 1912 
similarity with known elements is one of the features 
of the chemistry of radio-elements. 
Returning to the experiments with the uranium 
solutions purified by ether, Fig. 3 shows the growth 
of radium therein. The three curves labelled Hs JUL. 
III. refer to these preparations. No. III. was the 
last prepared, after experience with the others, and 
contained. both the greatest quantity of 
uranium and the least radium initially. No. 
IV. refers to a much later experiment with 
no less than 6 kilograms of uranyl nitrate, 
purified by repeated crystallisation in the 
course of other work. In all, there has been 
a distinct growth of radium, but it is so 
small, and the period over which the 
measurements extend is so prolonged, that 
the errors of the individual measurements are 
relatively great. The general scope of the 
curves, as indicated in the figure, is, how- 
ever, probably not far wrong. A conservative 
view to. take is that in all cases the curves 
are straight lines. There is some indication 
in No. I. of an increasing slope, but it is 
negatived by the evidence of Nos. II. and 
IT. 
The quantity of uranium in the four pre- 
parations differs widely. In Fig. 4 the curves 
are replotted in a different way to eliminate 
this difference. The ordinates represent the 
quantities of radium formed in terms of the 
amounts of radium in equilibrium with the 
uranium. The equilibrium amount is the 
amount that theoretically should be formed 
after the lapse of sufficient time, if uranium 
is the ultimate parent of radium. It will be 
seen that the slopes of the four curves are 
all different and diminish in order, the 
growth in the first being the greatest, 
and in the last, after all the experience in methods 
of purification, the least. This is additional evidence 
that, so far, the radium formed is derived, not from 
the uranium, but from varying infinitesimal quanti- 
ties of ionium still unremoved by the purification 
processes. 
separated from minerals can be readily de- 
tected and measured in a relatively short 
space of time. The curve shown (Fig. 2) is 
taken from a paper by Keetman (Jahr. 
Radioact. Elektronik, 1909, vi., 270), who 
has worked upon this parent of radium in 
Germany. Although the total quantity of 
radium represented by this curve is only 
nine millionths of a milligram, it is 
enormous compared with that shown by the 
other diagram (Fig. 1), in which the quan- 
tity of radium produced in a period about 
eight times longer is nearly a hundred times 
less. Further work on this parent of radium 
proved that it was not actinium, but a new 
radio-element admixed with it, which Bolt- 
wood called ionium. It is radio-active, and 
its radiation consists entirely of a-rays of 
very low range. The chemical nature of this 
ionium is absolutely identical, so far as is 
known, with that of thorium, and it cannot 
be separated from it. On the other hand, it 
is easily separated from any mixture, how- 
ever complex, by adding a trace of thorium 
Menge Radium (< 107 © mg) 
and senarating and purifying the latter. It is 
interesting to note that no fewer than three at 
least of the known radio-elements—ionium, radio- 
thorium, and uranium X—are absolutely identical in 
chemical properties with thorium. This complete | 
NO, 2217, VOL. 89] 
Fic. 2. 
Taking No. III. as the best of the first batch of 
preparations, the growth of radium therein is only 
about 1/30,000 part of what would have occurred if 
uranium were the direct parent of radium. Some 
idea of the minuteness of the quantities of radium 
