604 Dr. B. B. Boltwood on 



hydrogen sulphide. The precipitated sulphides were filtered 

 off, the filtrate was boiled to remove the excess of hydrogen 

 sulphide, and the salts in solution were oxidized with a small 

 quantity of nitric acid. About 0*1 gram o£ microcosmic salt 

 (NH 4 NaHP0 4 . 4H 2 0) was added to the solution, which was 

 allowed to cool somewhat, and was then poured into a solution 

 containing 10 grams of sodium carbonate (crystals) dissolved 

 in about 150 c.c. of warm water. The mixture was heated to 

 boiling and boiled gently for five minutes^ cooled to about 

 50° C, and 10 grams of ammonium chloride added. The 

 mixture was again heated, gently boiled for five minutes 

 longer, and allowed to stand until cold. The precipitated 

 phosphates were filtered off, the filtrate boiled in a large 

 beaker until the greater part of the ammonium carbonate 

 was driven off and the uranium phosphate was precipitated 

 from the solution. Just enough nitric acid was added to 

 completely dissolve the precipitate, and the solution was 

 boiled to expel the last trace of carbonic acid. About 

 O'l gram of microcosmic salt was added to the solution and 

 then 10 grams of pure crystallized sodium thiosulphate. 

 The solution was boiled for 15 minutes and the precipitate of 

 sulphur and uranium phosphate was filtered into a (platinum) 

 Grooch crucible, washed and ignited. The residue of phos- 

 phate was dissolved in the smallest possible quantity of hot, 

 concentrated nitric acid, the solution was separated from the 

 asbestos-felt, the excess of acid almost neutralized with 

 ammonia, and the clear solution poured into a solution of 

 3 grams of ammonium carbonate in 30 cc. of water. A slow 

 current of hydrogen sulphide was now passed into the 

 mixture for about two minutes and the liquid was heated 

 cautiously to boiling. The liquid was allowed to cool and 

 the precipitate, which in some cases was very slight, was 

 filtered off. The filtrate was boiled until the excess of 

 ammonium carbonate was expelled, acidified with nitric acid, 

 O'l gram of microcosmic salt added, and the uranium re- 

 precipitated as phosphate by adding 10 grams of sodium 

 thiosulphate and boiling for 15 minutes. The precipitate 

 was filtered into a Gooch crucible, washed thoroughly with 

 water, ignited gently until the sulphur was removed, and 

 then heated to a low red heat and allowed to cool in a 

 desiccator. The uranium contained in the green pyro- 

 phosphate of uranium thus obtained was calculated on the 

 assumption that the green salt contained 68*55 per cent, of 

 uranium *, an assumption which was frequently verified by 

 converting the green salt into the normal yellow modification 



