148 W. F. Tlillebrand — Composition of Yttrialite. 



at 23i°C and 4-646 at 26° C. respectively. The two sizes 

 were then mixed, giving one sample of about 4*65 sp. gr. at 

 25° C. That of the unpurified material analyzed by Mackintosh 

 was 4*575. 



Qualitative tests showed the absence of zirconium, glucinum, 

 and aluminum and the presence of a little more than a trace of 

 fluorine, of a very little carbon dioxide, besides some other 

 gases which were set free by treatment with acids and by 

 fusion with an alkali carbonate. 



Silica was separated by two evaporations with hydrochloric 

 acid ; lead was then thrown out by hydrogen sulphide. 



The earths when finally collected together free from all iron, 

 manganese, uranium, titanium, phosphorus, calcium, and mag- 

 nesium, were ignited and weighed. No filtrate from a precip- 

 itate of earth-oxalates was ever regarded as free from earths till 

 after evaporating, igniting, and retesting. This is a needed 

 precaution in all similar analyses. 



The combined earths were dissolved in nitric acid and evapo- 

 rated to dryness. A saturated solution of potassium sulphate 

 was poured upon the dry mass, which wholly dissolved but 

 almost at once began to deposit double sulphates. Solid potas- 

 sium sulphate was then added in crystals. After twenty-four 

 hours the precipitate was collected on a filter and washed with 

 the precipitant solution. Twice was this precipitation repeated 

 after first dissolving the double salts in acid, precipitating by 

 potassium hydroxide, washing, redissolving in nitric acid, and 

 evaporating. Then only was the extraction of the yttrium- 

 erbium group practically complete. From the combined fil- 

 trates the soluble earths were thrown out, reconverted into 

 nitrates, and again treated with potassium sulphate, whereby a 

 little further insoluble matter was obtained. Yttrium-erbium 

 oxides obtained from the oxalates gave a light-colored mixture 

 of 265-6 Mol. W. (108-8 At. W.), which furnished a pink 

 nitrate solution showing the erbium absorption bands strongly 

 marked, with very faint indications of the strongest band of 

 the didymium components. 



The insoluble sulphates were converted into chlorides and 

 thrice treated with potassium hydroxide and chlorine to pre- 

 cipitate thorium and cerium. From the filtrates the soluble 

 earths were recovered and subjected to a repetition of this 

 treatment. Their oxides after purification were light dirty 

 brown when heated over the Bunsen flame, but grayish white 

 when blasted. Their Mol. W. was 335-5 (At. W. 143-7); their 

 nitrate solution was pink and gave the characteristic didymium 

 component absorption bands altogether free from those of the 

 erbium constituents. 



Cerium and thorium were separated by a combination of the 



