204 
IOWA ACADEMY OF SCIENCE 
oughly mixed with a platinum spatula. The covered crucible, al- 
ways of platinum, is heated for twelve to fifteen minutes with a 
Bunsen flame, and for about the same length of time with a blast 
lamp until there is no further escape of carbon dioxide, and the 
substance is in a state of quiet fusion. The flame should be di- 
rected obliquely against the sides of the crucible. This is a neces- 
sary precaution as it makes possible the oxidation of the substances 
rather than the reduction. The crucible containing the substance 
in a state of fusion is seized with the tongs and given a rotary 
motion to distribute the contents against the walls by which the 
melted mass is more easily removed. The mass is extracted with 
hot water, and hydrochloric acid is added, and the substance is 
evaporated, stirring until a fine, dry powder results. To hasten the 
removal of the fused mass from the crucible, hydrochloric, acid can 
be added to it directly. If the fused mass is green it denotes the 
presence of manganese, and hydrochloric acid would liberate chlo- 
rine which would attack the crucible. Under such conditions, nitric 
acid can be used instead of hydrochloric acid. 
Many of the authorities suggest equal weights or molecular 
weights of sodium and potassium carbonates as the flux by which 
fusion takes place at a much lower temperature. Hillebrand 1 says, 
“There is no advantage in using the much recommended, because 
more fusible, double carbonate of sodium and potassium, or the 
equi-molecular mixture of the normal carbonates or of sodium 
carbonate and potassium bicarbonate. As Dittrich 2 says potassium 
salts are more prone to pass into precipitates than sodium salts 
and it may be that the higher melting point of sodium carbonate is 
a distinct advantage. Certainly, for effective decomposition of 
some rock constituents, a far higher temperature than that of the 
double salt is required.” 
Substances which contain a considerable quantity of fluorine 
can not be treated with hydrochloric acid, as silicon tetrafluoride 
would be volatilized. Substantially the old method of Berzelius is 
employed. After the fusion with sodium carbonate, the mass is 
extracted with water, the residue pulverized as finely as possible, 
filtered, and the greater part of the silica in the filtrate is removed 
by precipitation with ammonium carbonate. The precipitate is 
filtered off, ignited and weighed. The silicic acid which remains 
in the filtrate is precipitated by ammoniacal zinc sulphate. The 
1. Bulletin 305, U. S. Geol. Surv. , page 72. 
2. Anleitung Zur Gesteinsanalyse 1905, page 5. 
