the Origin of Radium . 605 



(containing 66*81 per cent.), by solution in nitric acid, 

 evaporation of the solution, and general ignition of the 

 residue. It is important that the precipitate, obtained after 

 treating with hydrogen sulphide the ammonium-carbonate 

 solution of the first uranium precipitate, be tested for uranium, 

 by dissolving it in hydrochloric acid, nearly neutralizing the 

 free acid with ammonia, and adding potassium ferrocyanide. 

 If a reddish precipitate, indicating the presence of uranium. 

 i- obtained, it is be<t to repeat the entire analytical operation, 

 although fairly satisfactory results can usually be obtained 

 by the following procedure : — The precipitate of ferro- 

 cyanides is filtered off and treated on the filter-paper with a 

 solution of ammonium carbonate. The filtrate is boiled with 

 a small quantity of pure sodium hydroxide until it no longer 

 has the odour of ammonia, and the precipitated sodium 

 uraninate is filtered off, dissolved in nitric acid and added to 

 the filtrate from the original precipitate. The precipitate 

 never contains appreciable quantities of uranium unless (1) 

 the solution contains too much tree nitric acid, (2) the 

 hydrogen sulphide is passed too long into the solution, or 

 (3) the solution in ammonium carbonate is heated too 

 strongly. 



Many of the minerals used in this investigation were not 

 soluble in dilute acids and required the action of hot, con- 

 centrated sulphuric acid to effect their complete decomposition. 

 These minerals were treated with a large excess of concen- 

 trate 1 sulphuric acid and a little hydrofluoric acid in a 

 platinum dish, and heated until sulphuric-acid fumes were 

 copiously evolved and the mineral was completely decdm- 

 posed. The resulting mixture was allowed to cool and was 

 then treated with about 100 c.c. of cold water. The whole 

 ^ a- heated to boiling, oxidized with bromine water, and 

 filtered. The filtrate was diluted to a volume of about 

 2.*>(> c.c, heated to boiling, and an excess of ammonia added. 

 The precipitate was filtered off, treated with dilute hydro- 

 chloric acid, the excess of acid partially neutralized with 

 sodium hydroxide, and the solution was poured into a solution 

 containing 15 grams of sodium carbonate. The mixture was 

 boiled for five minutes, allowed to cool somewhat, 10 grains 

 of ammonium chloride added, and the rest of the operation 

 carried out as previously described. The quantities of the 

 minerals taken for the analysis varied from 0*1 gram to 

 1 grain, depending on the percentage of uranium present, the 

 larger quantities being of course taken of those minerals 

 containing the least uranium. 



Like many other analytical operations, this one requires a 



