170 Misceilciiieous Intel/igeuce. 



circumstance that every time an alkaline glass is heated very 

 highly with fixed bases in quantity sufficient to combine with the 

 silica a portion of the alkali is volatilized, amounting in some 

 cases, even to one half. Boracic acid complicates the analysis, 

 and especially that part of it which relates to the alkali. 



Here M. Berthier has been induced to suggest, in place of cither 

 those substances, the use of oxide of lead, and has found it so 

 convenient in his own experiments as to have employed it con- 

 stantly for more than a year at I'Ecole des Mines. At first, car- 

 bonate of lead in the proportion of three to one of the stone was 

 used, but on fusing this in platinum or silver vessels, though a 

 mass was obtained perfectly soluble in acids, yet the crucibles 

 were frequently acted upon, because a portion of lead was some- 

 times reduced by something in the stone, or by the fumes of 

 the fuel. 



To obviate this inconvenience, nitrate of lead was used, and 

 the following is the process ; The mineral is to be pulverized and 

 then exactly mixed with two parts of nitrate of lead and one 

 part of ceruse also pulverized. The mixture is to be placed in 

 a small covered platinum crucible, and this placed in another 

 crucible also to be covered ; they are then to be heated red for 

 a quarter of an hour without being uncovered : the fusion takes 

 place tranquilly and without swelling, and a yellow or brown 

 transparent liquid glass is obtained. The crucible is then taken 

 out of the furnace, and its contents poured into water, as much 

 of it being taken out with a spatula as may be, and then the 

 crucible itself plunged into the water. The glass splits into 

 small portions and is readily acted on by acids. It is then bbiled 

 with nitric acid, and from time to time the portions of silica are 

 crushed with an agate or porcelain pestle. The silica remains 

 in a gelatinous state. The lead is precipitated by sulphuric acid, 

 the liquor being tested ultimately by sulphuretted hydrogen that 

 no portion may remain. The solution is then boiled with car- 

 bonate of ammonia, and the precipitate analyzed by the or- 

 dinary method. Finally, the liquid is to be evaporated to dry- 

 ness, the salts calcined in a platinum crucible, the residuum 

 collected and weighed. This residuum contains the alkali in the 

 state of sulphate, and almost always mixed with sulphate of 

 magnesia. It may be treated as follows : 



1. Precipitate the sulphuric acid by acetate of barytes, collect 

 the precipitate and weigh it ; precipitate the barytes by carbonate 

 or oxalate of ammonia. Evaporate the clear solution to dryness, and 

 collect the salt left : it will be a mixture of alkaline subcarbonate 

 and caustic magnesia; weigh it, and expose it to the air. If it con- 

 tains potassa it will deliquesce ; wash off the alkali from the mag- 

 nesia, and weigh the latter, then ascertain whether the alkaline 

 carbonate be of potassa, soda or lithia. When the sulphates 

 first obtained arc neutral, it is sufficient to determine very 



