» oe > 
NA TURE . 
461 

Aprit 7, 1923] 
° Letters to the Editor. 
[The Editor does not hold himself responsible for 
opinions expressed by his correspondents. Neither 
can he undertake to return, nor to correspond with 
the writers of, rejected manuscripts intended for 
_ this or any other part of Nature. No notice is 
taken of anonymous communications. | 


































On Urbain’s Celtium Lines. 
_ In a previous letter (NATURE, March 10, p. 322) 
ve have shown that the optical spectrum of the 
new element hafnium, of atomic number 72, which 
was discovered a short time ago by Coster and 
Hevesy, does not contain any of the lines belonging 
to the characteristic spectrum ascribed by Urbain 
(Comptes rendus, t. 152, 1911, p. 141) to an element 
celtium belonging to the family of rare earths, which 
element was assumed by Dauvillier and Urbain 
(Comptes rendus, t. 174, 1922, pp. 1347 and 1349; 
NATURE, February 17, 1923, p. 218) to possess the 
ytomic number 72. 
Through an examination of the very careful 
measurements of the spectra of the rare earths 
published during the last few years by Eder, we have 
learned in the meantime that the greater part of 
Urbain’s celtium lines have been observed by this 
author (Wiener Ber., Ila, vol. 124, I915) in the 
spectrum of a preparation of the rare earth element 
cassiopeium or lutetium. The discovery of this 
element (atomic number 71) was announced in 1905 
by Auer von Welsbach, to whom the former name 
due, while the name lutetium was proposed in 
1907 by Urbain, who published at the same time 
the first list of lines of its optical spectrum (Comptes 
rendus, t. 145, 1907, Pp. 759). : 
In the table below we give the wave-lengths of 
the celtium lines from Urbain’s paper, together with 
heir relative intensities (by use of the numbers 
4, 6, and 8 instead of Urbain’s notations : moyenne, 
ssez forte, forte, trés forte) and the corresponding 
wave-len and intensities from Eder’s measure- 
ments of the cassiopeium spectrum, corrected to 
Rowland’s scale. 

-Urbain’s Ct Seas Cp Lines 
* observed 
Lines. Hae Sa 
2459°4 2 | 2459°71 I 4 | 2885:23 3 
2469°3 2 he 4 A 
2481-6 4 | 2481-79 2 2 2931°56 I 
} 25369 4 | 253709 2 4 | 294982 3 
} 2677-77 2 | 2677:35 1 g | {3080-22* 4 
} 2685-2 8 | 2685-24 3 (3081-59 8 
27291 6 | 2729°08 3 8 3118-56* 5 
| 27379 2 + 6 | 3171°-49* 5 
2765°8 8 | 2765:88 3 8 | 3198-25* 8 
28343 4 | 283437 1 4 ‘is 
28373 4 . 6 | 3391-73 4 
2845-2 6 | 2845:23 2 2 x 
2870-2 2 es 
_ It is important to notice the very close accordance 
‘in the values of the relative intensities in the two 
tables, which we think justifies the identification of 
the lines also in the few cases where the difference 
between the wave-lengths is slightly greater than is 
_to be expected from the usual accuracy of Urbain’s 
measurements of wave-lengths of rare earths. Only 
a few of these lines, denoted by an asterisk, were 
‘included in Urbain’s original list of the lutetium lines. 
NO. 2788, vol. 111] 
In order to verify the origin of Eder’s lines and 
to endeavour to find out why most of these lines 
were not included in Urbain’s original list, we have 
examined the optical spectrum of a highly purified 
cassiopeium preparation kindly presented to this 
institute by Dr. Auer von Welsbach. At the same 
time Dr. Coster has photographed the X-ray spectrum 
of the same preparation and found it to contain no 
trace of an element with atomic number 72. Our 
exposures revealed in the spectral region investigated 
(2500-3500 A) all the lines observed by Eder and 
brought certain new features to light, which made 
it possible to understand why these lines were not 
observed in the original investigation of the lutetium 
spectrum published by Urbain in 1907. In an 
exposure in which the salt. was placed at the carbon 
anode of the arc, the celtium lines came out sharp 
and in about the same relative intensities as in Eder’s 
investigations, but when the salt was placed at the 
cathode, most of the celtium lines were much more 
intense and very diffuse and broad, especially in 
the part of the arc nearest to the cathode. The only 
¢eltium lines which also under these conditions came 
out as sharp as the rest of Eder’s cassiopeium lines 
were the four previously mentioned lines denoted in the 
table by an asterisk. On account of this behaviour 
of most of the celtium lines they will be very difficult 
to observe in less pure preparations. Some exposures 
taken with a less concentrated (10 per cent.) sample, 
formed by mixing Auer’s preparation with a scandium 
salt, did in fact show only the usual lines, whereas 
most of the celtium lines could be detected only by 
the presence of an increase of the continuous back- 
ground of the plate. Urbain’s results are therefore 
easy to understand, if we assume that the prepara- 
tions investigated by Urbain in 1907 contained a com- 
paratively small amount of the element with atomic 
number 71, and that only after treating the prepara- 
tions further a concentration was obtained sufficient 
for the production of the diffuse lines, which in 1911 
were ascribed to the presence of a new rare earth 
element, which was called celtium. 
As to the origin of the lines given by Urbain as 
celtium lines, and not present in Eder’s cassiopeium 
spectrum, we have found in the spectrum of Auer’s 
preparation a weak line with wave-length 2738-1 A, 
which may be identified with a celtium line (2737°9 A). 
The line 33260 occurs as a weak line in the mentioned 
mixture of scandium and cassiopeium, but could not 
be found in the spectrum of Auer’s pure preparation. 
As Urbain states that scandium was present as an 
impurity in his preparations, and as the scandium 
spectrum is not very well known, we have also taken 
a strong exposure of this spectrum, but could not 
find any of the remaining lines. Probably these 
lines have the same origin as the other celtium lines, 
but as they are weaker they will need a very strong 
exposure, especially if they also are diffuse. For 
such strong exposures we had not sufficient material. 
It is of interest to add that in a recent note (Comptes 
vendus, t. 176, 1923, p. 496), which first came to our 
notice after the above was written, dealing with the 
discovery by Coster and Hevesy of the element 
hafnium with atomic number 72, Urbain himself 
directs attention to the particular behaviour of the 
lines ascribed by him to celtium, and expresses the 
conjecture that these lines—the observation of which 
was the basis for his belief in the presence of a new 
element in his preparation—may actually constitute 
the spark spectrum of the element 71. 
H. M. HANSEN, 
S. WERNER. 
Universitetets Institut for teoretisk Fysik, 
Copenhagen, March 20. 
