118 Ellen Gleditscli—Life of Radium. 



acid. The filtered solution contained no uranium. The resi- 

 due weighed only 213 grams and its activity was negligible. 



The cleveite solution was treated with hydrogen sulphide 

 and the sulphides were removed by filtration ; the hydrogen 

 sulphide was expelled from the solution by a current of air. 

 The solution was heated to boiling and oxalic acid was added; 

 the precipitate formed was not so large as I expected and I 

 believed that the precipitation was not at all complete. I now 

 made several other solutions of small quantities of cleveite 

 using different solvents and I tried various methods of precip- 

 itating; but I always found that the precipitation of the rare 

 earths as oxalates was an unsatisfactory one. The precipita- 

 tion as fluorides* seemed much better to me. I therefore 

 removed the oxalates formed in the main solution, evaporated 

 the liquid to dryness, f destroyed the oxalic acid and dissolved 

 the residue in water and hydrochloric acid. The solution was 

 boiled and an excess of hydrofluoric acid was added. 



The liquid was allowed to cool and the precipitate, which 

 was at first very voluminous, settled down as a heavy powder, 

 that could easily be filtered off. It is obvious that to work 

 with large quantities of hydrofluoric acid is very slow; the 

 platinum dishes available in an ordinary laboratory are not 

 large and it was necessary to precipitate only small portions 

 of the mineral solution at a time. But the precipitation is so 

 much better and more complete, that there can be no doubt 

 that this method ought to be preferred for minerals like cleve- 

 ite. The fluorides were filtered off and a solution containing 

 two grams of oxides of rare earths:}: were added. The earths 

 were again precipitated as fluorides and removed. The fluor- 

 ides were heated and by addition of sulphuric acid and ignition 

 converted into sulphates. All the sulphates were dissolved 

 and the hydroxides precipitated several times with ammonia. 

 The final precipitate was dissolved in dilute hydrochloric acid 

 and brought into a glass bulb; the results of the measurements 

 are given in Table 4. 



In the mineral solution a new quantity of rare earths was 

 added and precipitated with hydrofluoric acid ; the precipitate 

 was treated exactly as the main precipitate, and after solution 

 it was placed in a glass bulb : Test solution from cleveite. 



* See Keetman. Thesis. Berlin 1909. 



■f During this evaporation a slight accident occurred. A loss of ionium 

 will give a period of too high value. The result in the cleveite solution con- 

 sequently has to be considered as a maximum value. 



% The oxides of rare earths added here and in other solutions had all been 

 prepared earlier from the Norwegian mineral gadolinite ; they consisted 

 chiefly of yttrium earths with some cerium earths. The amount of thorium 

 in the gadolinite was very slight and it contained no uranium. 



