284 
NATURE 
[Fuly 23, 1885 
spectrum at all, whilst the other should give the spectrum in its 
highest degree of intensity. 
I commenced the purification of didymia in the latter part of 
the year 1883, and the operations have been going on since 
almost daily in my laboratory. At intervals of some weeks the 
didymia in the then stage of purification was tested in the radiant 
matter tube, a little lime having previously been added to bring 
out the discontinuous phosphorescence. During the first month 
or two the intensity of the orange band spectrum scarcely dimin- 
ished. After this it began to fade, but the last traces of orange 
band were very stubborn, and not till the last few weeks could 
I obtain a didymia to show no trace of the orange band spec- 
trum ; and this result has not been accomplished without sacri- 
fice. My 1000 grammes have dwindled away bit by bit, till 
now less than half 2 gramme represents all my store. 
Samaria.—The foregoing experiments left little doubt that ~, 
the orange-band-forming body, was samarium ; the last problem 
was, therefore, to get this earth in a pure state. The general 
plan of operations was the same as I adopted in getting didymium 
free from samarium, only attention was now directed to the por- 
tions richest in samarium which had been formerly set aside. 
The colour of samaria, as pure as I have been able to prepare it, 
is white with the faintest possible tinge of yellow. The absorp- 
tion spectrum of samarium salts is much more feeble than the 
spectrum of didymium. 
The Phosphorescent Spectrum of *Samarium.—Pure samaric 
sulphate by itself gives a very feeble phosphorescent spectrum. 
When, however, the samaria is mixed with lime before examina- 
tion in the radiant matter tube, the spectrum is, if anything, 
more beautiful than that of yttrium. The bands are not so 
numerous, but the contrasts are sharper. Examined with asome- 
what broad slit, and disregarding the fainter bands, which 
require care to bring them out, the spectrum is seen to consist of 
three bright bands—red, orange, and green—nearly equidistant, 
the orange being the brightest. With a narrower slit the orange 
and green bands are seen to be double, and on closer examina- 
tion faint wings are seen, like shadows to the orange and green 
bands. 
Preliminary experiments had shown me that lime was one of 
the best materials to mix with samaria in order to bring out its 
phosphorescent spectrum, but it was by no means the only body 
which would have the desired effect. 
The samarium spectra, modified by other metals, may be 
divided into three groups. The first group comprises the spectra 
given when glucinum, magnesium, zinc, cadmium, lanthanum, 
bismuth, or antimony is mixed with the samarium. It consists 
simply of three coloured bands—red, orange, and green ; as a 
typical illustration I will select the lanthanum-samarium spectra 
(Fig. 1). 
The second type of spectrum gives a single red and~range 
and a double green band. This is produced when barium, 
strontium, thorium, or lead are mixed with samarium. The 
lead-samarium spectrum (Fig. 2) illustrates this type. 
The third kind of spectrum is given by calcium mixed with 
samarium. Here the red and green are single, and the orange 
double. Aluminium would also fall into this class were it not 
that the broad, ill-defined green band is also doubled. The 
calcium-samarium spectrum (Fig. 3) is a good illustration of this 
type. 
Mixed Samarium and Vitrium Spectra.—It was interesting 
to ascertain what spectrum a mixture of samarium and yttrium 
would give. A mixture of 90 parts of samaria to Io of yttria 
was treated with sulphuric acid and then ignited, and afterwards 
examined in the radiant matter tube. The result was as remark- 
able as it was unexpected. Not a trace of the yttrium spectrum 
could be detected. The powder phosphoresced with moderate 
intensity, but the spectrum was almost the facsimile of that given 
by pure samaric sulphate, except that the sharp orange line, 
which in the spectrum of pure samaric sulphate is only just 
visible, had gained sufficiently in intensity to be measurable, 
and was found to lie at 2693, on the a scale, 
A large number 
of samaria and 
are given in full 
of experiments were next tried on mixtures 
yttria in different proportions, and the results 
in the paper. 
Up to mixtures of 43 parts samaria and 57 parts yttria the 
spectrum nearly resembled the lead-samarium spectrum. Not 
a band of the yttria spectrum could be detected, and the brilliant 
orange line stood out sharply in the whole series. This spectrum 
is represented in Fig. 4. 
After that proportion had been reached a change rapidly came 
over the spectra, and in the next trial mixture—samaria 35, 
yttria 65—the only indication of the samarium spectrum that 
could now be found was seen in the two faint green bands next 
to the citron line of yttria, and the new orange line, which shone 
out as brightly and sharply as ever. 
Fic. 1. 
It will be remarked that a sudden change of spectrum occurs 
between very narrow limits of mixture. 
The spectrum of a mixture of 44 parts samaria and 56 parts 
yttria, except for the orange line, is the pure samarium spectrum. 
The spectrum of 42 samaria and 58 yttria is built up of some of 
the component bands of the spectrum of each earth ; whilst the 
spectrum of 39 samaria and 61 yttria is almost a pure yttria spec- 
trum, the sharp orange line running across them all. 
The Delicacy of the Spectrum Test for Samarium.—Experi- 
ments were now commenced with the object of getting some 
approach to a quantitative estimate of how small a quantity of 
samarium could be detected. 
A mixture was first made in the proportion of 1 part sama- 
Fic. 3. 
rium to 100 parts of calcium. The spectrum is very brilliant, 
and but little inferior in sharpness to,the spectrum given by a 
50 per cent. mixture. 
A mixture was now prepared containing I part of samarium 
to 1000 parts of calcium. Very little difference can be detected 
between the spectrum of this mixture and that of the last. The 
bands are, however, a little less sharp. 
FiG. 4. 
A mixture containing I part of samarium to 10,000 parts of 
calcium was now tested. The bands are now getting fainter, 
the second green band is fading out, and the continuous spectrum 
of calcic sulphate is getting brighter. 
The next mixture tried contained one part of samarium in 
100,000 parts of calcium. Here the green is almost gone, being 
| overshadowed by the continuous spectrum of calcium which has 
