224 Analyjts of the Arfemales 



arfenic acid. , The firft refiduum weighed 53. DifToIved, as 

 far as they could be, in muriatic acid, there remained 3, 

 which, upon examination, were found to be really filica. 

 Ammonia, poured in excefs into the muriatic folulion^ re- 

 difiolved 22,5, which were copper; and 27,5 of iron re- 

 mained behind. The proportions were. 



Silica - _ _ q 



Arfenic acid - - 33,^ 



Oxide of iron - - '^1^$ 



Oxide of copper - - 22,5 

 Water - - . \% 



9^>5 

 None of thefe experiments were fufficient to determine, 

 whether this ore is in the ftate of a triple fait, or merely a 

 mixture of two arfeniates. As, in a ternary combination, 

 the proportion of acid might vary, it cannot be juftly called 

 in to aid us in our inquiry. The folubility of one part of the 

 ore being much greater than that of the other, and in dif- 

 ferent quantities of each fait, incline more to the opinion that 

 it is but a mixture. 



Simple Arfeniale of Iron. 



This arfeniate, expofed to any degree of heat, gave but an 

 . unfatisfadory refult, with regard to the quantity of water. 

 The arfenic acid is volatilized from this ore with peculiar 

 facility, for which I fhall attempt hereafter to account. 



Some fubfcquent experiments, however, have induced mc 

 to fix the quantity of water at about 10,5. 



One hundred parts, boiled with potafh, left 58,5. The 

 liquor, treated, as ufual, by nitrate of lead, gave 31 of arfenic 

 acid. The 58,5 left four, which muriatic acid could not 

 diflblve, and which were filica. Ammonia diflblved 9; and 

 there remained 45,5 of iron. This analyfis prefcnts the fol- 

 lowing refult : 



Silica - _ _ _ 4 



Arfenic acid - - • 31 



O^ideofiron - - 45,5 



Oxide of copper - - 9 



Which will leave for water 10,5 



100,0 



This ore appears to be a pure arfeniate, mixed accidentalJy 



with a little copper ; as^ fome of the copper arfeniates cafually 



give traces of iron. This is the kind mentioned by Mr. 



Klaproth as au arfeniate of copper, and the firlt known 



midcr 



