MR. W. CROOKES ON RADIANT MATTER SPECTROSCOPY. 
G99 
Italian coelestiue showed a good orange-band spectrum when crushed and examined 
in the tube without any chemical treatment. After getting the mineral into solution 
by fusion with sodic carbonate, &c., the x could be concentrated by fractionally 
precipitating with alkaline carbonates (coming down in the first fractions). The 
sulphate produced from this precipitate also showed the desired spectrum. 
This sulphate was digested for some time in warm amnionic carbonate, and now the 
old distressing anomalies re-commenced. On most occasions, when working roughly 
on a scale of a few grammes, all the x was found in the filtrate on evaporation and 
ignition. When, however, I took identically the same material, and worked it up 
more carefully, in pounds or hundredweights, it sometimes gave nothing at all, some¬ 
times only a ridiculus mus on the smallest sized filter, got from a mountain of raw 
material. This was at first accounted for by the want of homogeneity of the mineral. 
The real explanation, however, was not discovered till long after (115). 
A quantitative estimation was attempted of the amount of x substance got from 
Italian ccelestine. G30 grms. gave 1*525 grm., or 0'24 per cent. Analysis showed 
this to be chiefly strontic sulphate, and the atomic weight of the metal was close to 
that of strontium. 
Is “ x ” a Mixture ? 
109. For a considerable time strontium minerals and salts only were worked upon, 
these being considered the cheapest and most fruitful sources of x. A considerable 
quantity of material was thus accumulated, showing the desired spectrum with great 
brilliancy. When, however, attempts were made to separate the spectrum-forming 
body from the accompanying elements, as strontium, calcium, &c., all the foregoing 
anomalies were displayed. Ultimately two portions of substance were produced—a 
precipitate (113) containing the supposed new element, and a filtrate, containing the 
strontium, calcium, and other impurities. Neither the precipitate nor filtrate tested 
in the usual manner showed the orange band anything like so well as the material 
before such separation, and indeed at this stage of the experiments it frequently 
vanished altogether. 
Some of the filtrate and precipitate were now mixed together, treated with 
sulphuric acid, and tested as before : they gave the orange-band spectrum as 
brightly as did the original substance. The ammonia precipitate was too small 
for analysis, but judging from its origin it might contain any or all of the rare 
earths. Chemical analysis showed nothing but a calcium salt in the filtrate. 
110. Could it be that the union of two bodies was necessary to give this spectrum, 
and that calcium was one of these ? Could the other constituent be of the nature of 
an acid such as boric, or a halogen like fluorine ? 
Many experiments were tried to test this hypothesis. Pure Iceland spar was 
dissolved in acid, a little of the above described precipitate added, and the mixture 
tested in the usual way. The orange band appeared again. 
