354 
CHEMICAL PRINCIPLES. 
Oxidesofgold. quently does not contain any trace of the above-mentioned me- 
tallic alloy. All hundred parts of it, when dried and heated 
to a red heat, in a small glass retort, produced 7'6 parts of 
pure water without the extrication of any gaziform body. 
The residue in the retort has now the colour of brick-dust, 
exactly the same as is obtained by a mixture of the aurum 
fulminans with sulphate of potash or silica, when the gold has 
been reduced by heat. Whence it seems to follow, that the 
powder which remained was nothing but a mixture of metallic 
gold with the oxide of tin. The brick-dust coloured powder, 
when treated with the nitro-muriatic acid, gave a solution of 
gold, from which the murias ferrosus precipitated 25'5 p. c. 
of metallic gold. The oxide of tin, not dissolved, weighed 
66 5 parts. The loss of O’S parts mast be attributed to the 
oxide of gold dissolved with the gold. 
When I caused the purple not decomposed to be digested 
with heat, the oxide of tin was dissolved of a yellow’ colour; 
and the metallic gold remained. The solution seldom shewed 
any traces of gold, but it contained both the oxidum stanneum 
and the oxidum stannicum. These experiments prove, that 
the purple of Cassius is a triple combination of the oxide 
of tin (probably of the intermediate oxide) with the oxide 
of gold and water. The action of heat, together with that 
of the muriatic acid, produces a reduction of the gold, by 
means of the intermediate oxide of tin, (oxidum stanneum) 
which is converted into an oxide in waximum (oxidum stanni- 
cum*.) The reason why the purple is only formed in very 
dilute solutions, seems to be, that in a more concentrated solu- 
tion, the muriatic acid has a stronger tendency to preserve its 
neutral combination with the oxide of tin. For which reason 
it is, that only the metallic alloy is precipitated. But wheu 
the solution is very dilute, the combination of the two 
acids 
• It is necessary to observe here, that there is, in vol. XXXV, p. 
1G2, an error as to the nomenclature of the oxides of tin. The ter- 
mination in cww should be employed for the intermediate oxide, and 
the termination in icu7n for the oxide in maximum. 
