JiiStORV OF GOLD. 539^ 



ble than the oxide at a minimum, it is still very soluble in 

 the muriatic acid, and in aqua regia. For example, let fall a 

 single drop of a very acid muriatic solution at a minimum, 

 into a solution of gold, which is habitually surcharged with it, 

 and you will not fail to produce a purple compound ; a purple 

 that certainly has nothing similar to the coloured powders pro- 

 duced by the sulphureous or phosphorous acid, sulphate of 

 iron, &c. Thus, in the case we are examining, there is no 

 reason to suppose, that a few atoms of oxide, which have just 

 acquired a maximum, should thus quit a solvent, that attracts 

 them on all sides, to unite in preference with gold, if the gold 

 did not attract them by a particular affinity. Let me observe 

 too, that the acidity of the liquids increases, in proportion as 

 the gold and tin fall down. 



If, therefore, the gold and oxide unite, notwithstanding the 

 obstacle they must have to encounter in such acid menstrua, 

 it is obviously necessary, that a particular attraction should 

 intervene, to save the tin from its customary solubility, 



But a single fact will estabhsh incontrovertibly the particular The purple 

 state of combination, which unites the oxide of tin to the gold, fectiy soluble 

 Throw some of the purple powder of Cassius, recently precipi- inammonia. 

 tated into a phial full of ammonia, it will dissolve immediately, 

 and impart to the ammonia a vivid and intense purple colour. 

 The solution will pass through the filter, without losing anj 

 thing. Water does not decompose it, as it does most of the 

 ammoniacal solutions of metals, unless it be surcharged with 

 the purple powder, and then part of it may separate. Distil- 

 lation, too, occasions the powder to subside, by carrying off the . 

 ammonia ; but as long as the liquid contains any ammonia, it 

 continues to hold some of the purple powder in solution. Acids 

 precipitate it from the ammonia in the same manner. 



The metallic precipitates of gold are not soluble in ammonia. 

 The oxide of tin at a minimum is but very imperfectly soluble 

 in it, since the solution is always milky. If the purple powderj 

 therefore, dissolve thus copiously and perfectly in ammonia ; 

 if it have properties, which neither gold nor the oxide of tin 

 possesses, we must conclude, that these two substances form a 

 real combination with each other : for real combinations alone 

 can have properties strikingly different from those which 

 characterize their coj^ponent parts. The combination of a 



metal, 



