COMPOSITION OF N.S.W. LABRADORITE AND TOPAZES. 201 



dissolved in water and filtered. Ammonium carbonate was added 

 to the hot solution and a gelatinous precipitate, consisting chiefly 

 of aluminium hydroxide, formed. After filtering and cooling a 

 more granular but still gelatinous precipitate separated out and 

 collected at the bottom of the beaker. This was filtered off and 

 examined. Only a small amount was obtained which on drying 

 became powdery and fused at a red heat. The sodium and 

 aluminium were determined by conversion into sulphates, the 

 fluorine by fusion with silica and alkaline carbonates. '0210 gram 

 heated to redness gave 23 8 per cent, sodium, 26*2 per cent, 

 aluminium, and 0*200 gram gave 3 33 per cent, of fluorine, leaving 

 oxygen by difference 167 per cent. Assuming some of the 

 aluminium to be combined with the oxygen as alumina due to 

 imperfect separation of the fluorine precipitate from aluminium 

 hydroxide, we have for the composition of the former fluorine 

 precipitate sodium 36*8, aluminium 11*7 and fluorine 51*5 per cent. 

 Cryolite requires 32-9, 12*8, and 54-3 per cent. 



In another experiment on a different precipitate -0130 gram, 

 gave sodium 208, aluminium 29*2 per cent., and 0-201 gram gave 

 30*8 per cent, of fluorine, hence oxygen by difference 19*2 per 

 cent. Allowing for alumina this gives sodium 35*1, aluminium 

 13*0, fluorine 51*9 per cent. 



It seems probable therefore that the fluorine is precipitated as 

 a double fluoride of sodium and aluminium. Like cryolite this 

 precipitate also fuses readily. But as the amounts analysed were 

 so small it is proposed to confirm the results by working with 

 larger quantities. 



Comparison of the various methods for fluorine on the topazes. — 

 1 2 3 



Fusion with alkaline Tetra- Fusion with alkaline 



carbonates. fluoride. carbonates and silica. 



N "fL W Mud^ } 5 ' 1101 1293 17-04, 17-60, 17-10 



Brazil \n 3 2\!l' 2 n!n-k 1090 "-*U-« 



New England topaz 16-30,15-92 



A table is given above shewing the results obtained by the 

 various methods for fluorine, on the topazes. While obtaining 



