OF ARTS AND SCIENCES. 443 



It is quite obvious that a single precipitation by the method used 

 does not yield titanic oxide free from alumina. The weighed precipi- 

 tate of each experiment was therefore fused with a small quantity of 

 sodic carbonate, and the product of fusion treated in the crucible with 

 strong sulphuric acid, which gradually with the aid of geutle heat dis- 

 solved everything completely. This solution after cooling was poured 

 directly into 100 cm.^ of cold water, and, the liquid remaining clear, 

 ammonia was added just to alkalinity, and then dilute sulphuric acid 

 corresponding to 2.5 grm. of pure acid. The precipitate thrown 

 down by ammonia was gradually, but after a few minutes entirely 

 redissolved, acetic acid wa.s added until it amounted in terms of abso- 

 lute acid to ten and a half per cent of the entire volume, 20 grm. of 

 sodic acetate introduced, the solution boiled, and the precipitate treated 

 as previously described. The results of this second treatment are 



In experiments (40) and (41), in which comparatively small amounts 

 of titanium were employed, the separation from alumina was reason- 

 ably good after the second precipitation, but in (42) to (46) the larger 

 quantities of titanic oxide still retained notable amounts of alumina. 

 The precipitates of (45) and (4G) were, therefore, treated again just 

 as before, to determine the effect of a third precipitation. 



