t *°5 1

matter as water will easily extract, a considerable part ofit will still dissolve in pure volatile spirits, but not somuch as before the ablution : I have not observed thewhole of the gold to be taken up in either case, thoughsome report that they have sound it to be so in both.Pure fixt alkalies, added in large quantity after the preci-pitation, do not appear to redissolve any of the gold.

If the aqua regia has been prepared with sal ammoniac,or if the precipitation is performed with a volatile alkali,the unwashed precipitate explodes, on being heated, witha bright flalh and a smart noise; whence its name aurumfulminans. If the aqua regia has been made without salammoniac, and the precipitation is performed with a sixtalkali, the precipitated gold makes no explosion : gradu-ally heated, it changes its dull yellowish to a bright pur-ple or purple-violet colour, and on further increasing theheat resumes its metallic aspect. A volatile alkaline salt,either in the dissolvent or in the precipitant, seems to beessentially necessary to the fulmination.

Aurum fulminans weighs about one fourth part morethan the gold employed, three parts of gold yielding fourof the fulminating powder : this I relate on the authorityof Lemery, Kunckel, and other practical writers, forthough I have often made the preparation myself, I havenever examined the increase of its weight. Part of theincrease proceeds from the volatile alkali; for on addingto the aurum fulminans a little vitriolic acid, the volatilefait rises in sublimation, satiated with the acid : theremaining powder is found to be divested of its fulmina-ting power. From the coalition of the volatile alkaliwith the nitrous acid in the menstruum results an ammo-niacal nitre, a salt which of itself detonates on being heated :by what power or mechanism its detonating quality is Ibremarkably increased in the aurum fulminans, is unknown.

P The