HADDAM AND BRAZIL. 12 i 



By the fiftii treatment it was not diminished, and then pre- 

 sented itself in the form of a hght white powder, resemhUng 

 pure silica in appearance. 



C. The residue (B) was repeatedly strongly calcined 

 with six parts of uitrate of barytes, and subsequently boiled 

 with nitric acid. 



After the 1st treatment, there remained 0.4.3 grammes. 



Sd 0.15 



3d 0.00 



And by the 4th operation only o.oi gramme was dissolved. 



The remaining 0.05 grammes were essayed in the man- 

 ner related in the preliminary experiments, and thus proved 

 to be oxide of titanium. Hence we have 1. per 100 of that 

 oxide. 



D. The nitric solutions were united and evaporated to 

 dryness, to expel the excess of the acid. The saline mass 

 was dissolved in water, and after the barytes was separated 

 ■with sulphuric acid, an excess of subcarbonate of ammonia 

 was added to the solution. An abundant precipitate appear- 

 ed, which entirely rc-dissolved. The Glucirin was precipi- 

 tated by ebullition. After edulcuratiun and calcination, it 

 weighed 0.79 grammes, or 15.80 per 100. 



E. The several muriatic solutions (B) were united and 

 evaporated to a dry mass, which was treated with muriatic 

 acid, and there remained 0.33 grammes of silica, from 

 which deduct 0.13 grammes acquired from the at!;ate n»or- 

 tai ; and there will be 0.20 grammes, or 4. per lOO as a con- 

 stituent of the mineral. 



F. After tlie silica was separated from the liquid (E), 

 the alumina and oxide of iron were precipitated by means 

 of a great excess of subcarbonate of ammonia. Aftrr twenty- 

 four hours, the liquor was separated from the yellowish prc- 



VOL. II. Q 



