MR. \V. CROOKES OK RADIANT MATTER SPECTROSCOPY. 
705 
radiant matter tube a good orange-band spectrum when mixed with lime and treated 
as usual, although without lime it gave no spectrum. 
So long as the lanthana showed the didymium absorption bands I could not be 
certain whether the orange-band spectrum belonged to it or to didymium, therefore 
the tedious process of fractionation with very weak ammonia in cold dilute solutions 
was repeated for some weeks. The fisrt precipitates were lanthana containing most 
of the didymia, whilst the final precipitates were lanthana almost if not quite free 
from didymia, according to the quantity originally present. After several hundred 
fractional precipitations repeated over and over again, a little lanthana was got which 
failed to show the didymium absorption bands. As the purification progressed the 
phosphorescent orange-band spectrum became fainter, until finally a lanthana was 
obtained which, mixed with lime and treated in the usual manner, gave no orange- 
band spectrum whatever. This lanthana was snow-white, and had an atomic weight 
of 138*3. Marignac gives for lanthana 138*6, Brauner 138*28, Cleve 138*22. 
Purification of Didymia. 
126. The earth formerly called didymia is now known to be a mixture of didymia 
and samaria. The didymia which I prepared by the method described above, when 
mixed with lime and sulphuric acid, and tested in the radiant matter tube, gave the 
orange-band spectrum as brightly as I had ever seen it. It was not, however, quite 
free from the accompanying samaria, and systematic operations were now commenced 
with the object of obtaining the didymia and the samaria in a state of purity,—that 
is to say, in such a condition that one of them should show no orange-band spectrum 
at all, whilst the other should give the spectrum in its highest degree of intensity. 
I did not attempt the two purifications simultaneously on the same material. One 
earth only was taken in hand at a time, and by repeated fractionations and the most 
profuse sacrifice of material, I was at last enabled to obtain a little of the desired 
earth quite free from admixture. 
127. I took didymia first. About 1000 grins, of the earth, partially purified as 
described (120, 121), were dissolved in a large excess of strong nitric acid. To 
the nearly boiling liquid a hot saturated solution of oxalic acid was carefully added, 
and constantly stirred, until the precipitate, which at first rapidly disappeared, 
just refused to dissolve. A drop or two of nitric acid was now added to render 
the solution clear, and the liquid set aside to cool, when brilliant pink-coloured 
prisms of didymic oxalate (containing nitric acid) crystallised out. These crystals 
contained nearly all the didymium and samarium, whilst the mother-liquor con¬ 
tained the greater part of the lanthanum—reserved for the preparation of pure 
lanthana (125). 
The crystals of didymic oxalate were ignited and again converted into nitrate, and 
the above-described partial crystallisation as oxalate repeated five or six times, in each 
case rejecting the mother-liquor as contaminated with lanthanum. 
