chemical 'Properties attributed to Light. 46*5 
riment, and which had acquired a deep yellow colour, had lost 
that property by which it effected the reduction of the metallic 
oxide, I now poured an additional quantity of the aqueous so- 
lution of the oxide into the phial, and, shaking the phial, exposed 
it, with its contents, to the heat of boiling water. 
After it had been exposed to this heat about two hours, I 
examined it, and found, that though a considerable quantity of 
gold had been revived, yet the aqueous liquid still retained a 
faint yellow colour. 
The oil of turpentine had acquired a deeper and richer gold 
colour, approaching to orange. 
To the contents of the phial, I now added about half as much 
distilled water, and, mixing the whole by shaking, I exposed the 
phial again, during two hours, to the heat of boiling water; 
when the remainder of the oxide was reduced, and the aqueous 
liquid left perfectly colourless. 
On repeating this experiment with oil of turpentine, and va- 
rying it, by using a solution of the oxide of silvery (an aqueous 
solution of nitrate of silver,) instead of that of gold, the result 
was nearly the same: the metal was revived, and the oil of 
turpentine acquired a faint greenish-yellow colour. 
I also revived the oxides of gold and of silver with oil of olives , 
by a similar process, with the heat of boiling water. The oil of 
olives used in these experiments lost its transparency, and be- 
came deeply coloured ; that used in the reduction of the oxide 
of silver, taking a very deep dirty brown colour, approaching to 
black ; and that employed in reducing the oxide of gold, being 
changed to a yellowish-brown, with a purple hue. 
In the experiment with the oxide of silver, the inside of the 
phial, in the region where the oil reposed on the aqueous solu- 
MDCGXCVI II. 3 O 
