Pottox— On the Extraction of Glucnum from Beryl. 141 
Zimmerman* investigated the original method of separating aluminia and glucina, 
by diluting and boiling the potash solution of the oxides. Atkinson and Smith,+ 
in 1895, published a method for the separation of iron and glucinum by nitroso 
8 naphthol after the method of yon Knorre. Havens? shows that aluminium and 
elucinum may be accurately and quantitatively separated, by adding ether to the 
mixed chlorides, and saturating with hydrochloric acid gas; all the aluminium 
chloride precipitating, and none of the glucinum chloride. Hartley§ has investi- 
gated the spectrum of glucinum. Lebeau|| has lately described an excellent 
method of making glucinum, by the electrolitic decomposition of the fused double 
fluoride of glucinum and sodium; also the reduction of the oxide, when mixed 
with another metal, or metallic oxide, in the electric furnace.4| The same author** 
describes the formation of the carbide and iodide of glucinum. 
Petterssont} gives the heat of neutralization of glucinum fluoride as Gl(OH),, 
2HF, Aq=19,683 ecal.; and Carnellyt{ has given the melting-point of glucinum 
chloride as between 585°—617°, and maintains the accuracy of these figures, though 
they are called in question by Nilson and Pettersson, who point out that glucinum 
chloride boils at 520°. Many other authors have contributed to the hterature of 
glucinum ; but their papers have no direct bearing on the subject of the present 
inquiry; and even the shortest reference to each would entail a long dissertation. 
Ill. Awnanysis oF BERYL. 
Ten pounds weight of beryl, from Limoges, was procured from Messrs. 
Harrington Bros., of London. It was of a pale green colour, opaque, and in plates 
or sheets, not showing any signs of hexagonal crystals; it seemed of uniform 
character, and not contaminated with rock, or other minerals. The whole 
quantity was ground up, mixed, sampled, and a fine sample carefully ground for 
analysis. ‘The analysis was performed by fusion with alkaline carbonates, 
separation of silica, precipitation with sulphuretted hydrogen, ammonia, sulphide 
of ammonia, carbonate of ammonia, and phosphate of soda; and the alkalies 
were estimated by treatment with hydrofluoric acid; all after the usual manner. 
The iron was separated from the alumina and glucina by caustic soda, redissolved 
in hydrochloric acid, and reprecipitated by ammonia. The caustic soda solution 
was acidified with hydrochloric acid ; and the glucina and alumina precipitated by 
* Zeitsch. Anal. Chem., vol. xxvii., p. 61. { Compt. Rend., vol. cxxy., 1897, p. 1172. 
| Journ. Amer. Chem. Soc., vol. xvii., 1895, p.688. ** Compt. Rend., vol. cxxi., 1895, p. 496. 
{ Amer. Journ. Sci., [4], vol. iv., 1897, p. 111. tt Zeitsch. Phys. Chem., vol. v., 1890, p. 259. 
§ Chem. Soe. Journ., vol. xliii., 1883, p. 316. tt Ber. Deutsch. Chem. Gesell., Jahrg. xyii., 1884, 
|| Compt. Rend., vol. exxvi, 1898, p. 744. p: 1357. 
2B2 
