536 Prof. R. Bunsen's Spectral- Analytical Researches. 



seen in Plate V. no. 15 e, faint lines in the neighbourhood of 

 70; but these lines are not sufficiently marked to serve as 

 a test for the presence of this substance. The absorption-spec- 

 trum of the solid or dissolved didymium salts, however, is so 

 characteristic as to afford a means for detecting the smallest 

 trace of the element in any liquid which is not very strongly 

 coloured by foreign subtstances. No. 15«, Plate V., gives the 

 absorption-spectrum of a crystal of didymium sulphate 0'4 

 millim. in thickness. 



Considerable doubt attaches itself to the purity of the 

 thorium oxide prepared according to the methods hitherto in 

 use. An inquiry into the behaviour of perfectly pure thorium 

 chloride in the spark-spectrum-apparatus was therefore neces- 

 sary. 



Orangite, from Brewig in Norway, was dissolved in hydro- 

 chloric acid ; the silica was removed by evaporation, and the 

 metals precipitable by sulphuretted hydrogen by the passage of 

 that gas through the liquid; the solution was then oxidized by 

 nitric acid, precipitated by ammonia, and the precipitate di- 

 gested with oxalic- acid solution until the thorium oxalate pro- 

 duced became perfectly white. The oxalate was washed, ignited, 

 and evaporated with concentrated sulphuric acid ; the resulting 

 sulphate, after dehydration at a high temperature, was dissolved 

 in the smallest possible quantity of water at 0° to 6° C. By 

 heating to 100° a precipitate was produced, which, when col- 

 lected on a niter, repeatedly washed with boiling water by 

 means of the water air-pump, and converted into chlorides, 

 showed the most prominent lanthanum-lines in the spark-spec- 

 trum. All the portions of the sulphate obtained by fractional 

 precipitation at 100° C, as w T ell as the mother liquor, showed 

 the lanthanum-lines ; the last portion exhibited also the didy- 

 mium absorption-bands and the spark-lines of cerium. Frac- 

 tional precipitation of the neutral sulphate by means of sodium 

 thiosulphate, yielded products which were not altogether free 

 from these impurities. A new way had therefore to be devised 

 for the removal of these foreign substances. The behaviour of 

 the oxalates of cerium, lanthanum, didymium, erbium, and 

 yttrium towards ammonium oxalate forms the basis of the 

 method. A concentrated boiling solution of the latter salt 

 dissolves small quantities of the oxalates, which are entirely 

 reprecipitated on dilution with water and cooling; oxalate of 

 thorium, on the other hand, is easily dissolved under the same 

 conditions, and is not reprecipitated either on coolings diluting, 

 or evaporating the liquid. Thorium oxide remains when the 

 residue, obtained by evaporating this solution, is ignited in a 

 platinum dish. By repeating this treatment with the sub- 



