Brass. 
37 
88 parts of lead. This method therefore of analysing 
brass cannot be depended on, unless the cupreous preci* 
pitate be afterwards separately treated to separate the lead^ 
which would render the analysis very complicated,, 
The following methods are given by Vauquelim 
3. Dissolve a known weight of brass in nitric acid; put 
it into a well closed bottle and add caustic potash to ex- 
cess, so that there shall be a very sensible alkaline taste 
in the liquor, shake the mixture well, and keep it some 
time in digestion. By this process the oxyds of copper 
and zinc are first both precipitated by the alkali, and af- 
terwards the zinc alone redissolved in the excess of potash., 
so that the clear solution is oxyd of zinc in potash, and 
the sediment left undissolved is the oxyd of copper. This 
oxyd is brown and nearly of the colour of metallic cop- 
per, but when thoroughly washed and gently dried, it 
only contains 65 per cent, of the metah If a previous as- 
say has shewn that the specimen of brass contained only 
copper and zinc, when the weight of the former is known, 
that of the other may readily be inferred ; or else, the al- 
kaline solution of zinc may be supersaturated with sul- 
phuric acid, so as first to precipitate and afterwards to re- 
dissolve the metal, when the sulphuric solution may be 
decomposed by a carbonated alkali. A very trilling 
quantity of copper passes into the alkaline solution of zinc, 
occasioned by the action of a small quantity of ammonia, 
generated by the nitrated metals when caustic alkali is 
added to them. If necessary this might be again sepa- 
rated by a heat cautiously kept below boiling, which 
would expel the ammonia, the cause of this error, but if 
brought fully to boil some of the zinc would separate 
from the alkali and cause a much greater error than be- 
fore. 
4. Dissolve brass in sulphuric acid, dilute with twenty 
times as much water, and immerse a stick of zinc exact- 
