462 a 
On Celtium and Hafnium. 
In our letter of February 9, which appeared in 
Nature of February 24, p. 252, we have shown that 
the element hafnium, of atomic number 72, detected 
by us in zirconium minerals, possesses physical and 
chemical properties quite different from those ascribed 
to a rare earth element celtium, the discovery of which 
was announced by Urbain in t911, and which recently 
was believed by Dauvillier and Urbain also to possess 
the atomic number 72. In a communication of 
February 19 to the Paris Academy of Sciences 
(Comptes vendus, vol. 176, p. 496, 1923) Urbain dis- 
cusses the same problem and still claims the identity 
of his celtium with our hafnium and by a claim of 
priority rejects the latter name. In the meantime, 
through the investigation of Hansen and Werner (see 
Nature of March Io, p. 322, and the above letter in 
this issue) on the optical spectrum of hafnium and on 
the spectrum ascribed by Urbain to celtium, new data 
have been brought to light, and we should therefore 
be glad to take the opportunity to complete our 
arguments as regards the questions discussed by 
Urbain. 
To put the matter clearly we must go back to the 
time preceding the announcement of the discovery of 
celtium. As is well known, Marignac succeeded in 
1878 in isolating from a mineral from Ytterby a 
substance which was considered to be a new element 
and called ytterbium. In 1905 Auer von Welsbach 
(Wiener Anzeiger, x., 1905; see also Sitzungsberichte 
115, July 1906, and Lieb. Ann. 351, p. 464, 1907) 
announced the discovery that this substance was a 
mixture of two elements, for which he later proposed 
the names aldebaranium and cassiopeium. Detailed 
information with regard to the spectra of these ele- 
ments and their atomic weights was first published by 
him in 1907 (Wiener Sitzungsberichte, 116, December 
1907), shortly after a similar announcement had been 
made by Urbain (Comptes rendus, 145, Pp. 759, 1907), 
who was the first to publish lists of lines for the 
separate spectra of the two new elements, and pro- 
posed the names neo-ytterbium and lutetium. By 
continued purification of his preparations, Urbain 
observed in the following years a gradual change in 
the spectrum and magnetic properties and announced 
in 1911 (Comptes rendus, 152, p. 141) the detection 
of a further element, which was called celtium, for 
which a separate list of spectral lines was published. 
Through the work of Hansen and Werner referred 
to above, it is clear, however, that the latter lines are 
due not to a new element but to the element which 
was called lutetium by Urbain and cassiopeium by 
Auer von Welsbach, in the spectrum of preparations 
of which the same lines were also observed by Eder 
in 1915. To this we may add that the same view is 
supported in a striking way by investigations on the 
magnetic properties of the various preparations. 
The circumstance that the paramagnetism of Urbain’s 
preparations of 1911 was three or four times smaller 
than that of his former preparations need not be 
explained as due to the presence of a new element, but 
may be considered as a consequence of the gradual 
concentration of the element cassiopeium or lutetium 
in his preparations. Thus Stephan Meyer (Wiener 
Sitzungsberichte, 117, Pp. 955, 1908) in his investiga- 
tions of the magnetic properties of the rare earths had 
already found in 1908 for the paramagnetism of a 
cassiopeium preparation a value almost as small as 
that measured in 1911 by Urbain for the preparation 
which was believed to contain the largest percentage 
of celtium. 
In view of the conclusion drawn by Hansen and 
Werner as regards the optical spectrum, this cireum- 
NO. 2788, VoL. 111 | 
NATURE 
stance may be taken as a proof that the original. 

[ArRIL 7, 1923 

preparations of Urbain from 1907 (Comptes vendus, 
145, P- 759) contained only a rather small fraction of 
the element cassiopeium or lutetium and presumably 
much less than the preparations of Auer von Welsbach 
of the same time. In this connexion it is of interest 
to mention that, according to the quantum theory of 
atomic structure, the element of atomic number 71 
must be assumed to be diamagnetic in its trivalent 
chemical compounds. In fact the absence of para- 
magnetism of such compounds is a necessary conse- 
quence of the theoretical conclusion that in the triply 
charged ions of the element 71 we first meet with a 
completed electronic configuration of four quantum 
orbits. (See N. Bohr, “Theory of Spectra and 
Atomic Constitution,’’ pp. 106, 114, Cambridge 
University Press, 1922.) It was this conclusion which 
led to the anticipation verified by our discovery of 
hafnium, that the element 72 should not have proper- 
ties analogous to the rare earths, but be a homologue 
of zirconium. 
As will appear from the above, the existence of an 
element with the properties ascribed to celtium cannot 
be maintained, and we think ourselves justified in 
concluding that the important problem of the nature 
of the element 72 may be considered as settled by the 
discovery of hafnium and the investigation of its 
properties. While thus the general conclusions of 
Dauvillier and Urbain must be rejected, there remains 
the question, of secondary importance, whether the 
two extremely faint X-ray lines observed by Dauvillier 
in Urbain’s preparation, which was believed to con- 
tain celtium, can have been due to a contamination 
of this preparation by a trace of hafnium. In the 
discussion of this point it must be emphasised, in the 
first place, that Urbain, in the course of his purifica- 
tion, made all possible precautions to remove elements 
other than the rare earths from his preparations. In 
fact, in his note in Comptes rendus, 1911, quoted above, 
M. Urbain states that ‘“‘des impuretés de toutes 
sortes, provenant soit des vases, soit des réactifs, 
s’accumulérent nécessairement dans des eaux méres 
successives. J'ai fait différents traitements de ces 
eaux méres par l’hydrogéne sulfuré, l’ammoniaque 
et l’acide oxalique, de maniére a éliminer tout ce qui 
dans cette substance n’appartenait pas au groupe des 
terres rares. J'ai examiné ensuite la terre purifiée et 
parfaitement blanche.’’ Now our investigations of 
the chemical properties of hafnium have shown that 
this element, just like zirconium, can be separated 
easily from the rare earth elements by a treatment 
with oxalic acid. 
Only two lines of the element 72 were claimed to 
have been detected by Dauvillier, and even in the case 
of the most intense of these lines we meet with the 
difficulty that it falls in the same place in the spectrum 
as the strongest zirconium line in the second order. 
As an argument against ascribing this line to zircon- 
ium, Urbain states that the optical spectrum of his 
preparations did not show any zirconium line. An 
investigation of Urbain’s spectrum of the “ celtium ” 
preparation, however, does not show any line of the 
hafnium spectrum (see Hansen and Werner, NATURE, 
March 10, 1923) either. If the possibility of the 
presence of one of these elements in Urbain’s prepara- 
tion can be taken seriously into consideration, it 
should be expected that zirconium would be present 
in greater amount. In fact, zirconium was likely to 
be more abundant in the original mineral than 
hafnium, and a purification of rare earth preparations 
from zirconium and not simultaneously from hafnium, 
by treating with oxalic acid or any other method 
mentioned by Urbain, is scarcely imaginable in view 
of the close similarity of the chemical properties of 
ag i ee 
