ClILOllIDES AND OXIDES OF MERCURY. 
253 
mixture of ammonia and nitrogen, and, at the same time, 
calomel is sublimed, generally blackened' by a little ammonia, 
from which it may easily be separated. The precipitate is 
readily soluble in the nitric and muriatic acids. Mixed 
with potassa, soda, lime or baryta, ammonia is liberated; the 
mass becomes yellow, but the decomposition is never com- 
plete. 
The nature of the products will be given farther on; it 
must be remarked that an alkali, even in excess, never libe- 
rates all the ammonia. 
The iodide of potassium produces a red precipitate of deuto- 
iodide of mercury, all the ammonia is disengaged, and the 
liquid remaining is a solution of caustic potassa. The sul- 
phuret of barium acts in the same way, bisulphuret of mercury 
being precipitated. 
To obtain a perfectly pure precipitate, some precaution is 
necessary. A cold solution of deutochloride of mercury 
must be treated with a slight excess of ammonia, the pre- 
cipitate thrown on a filter and well drained before being 
washed. It may then be washed with a quantity of water 
sufficient to displace the first liquid, and this even with great 
care, for cold water decomposes a small portion of it and its 
milky colour is lost. Having thus obtained a precipitate 
which I considered pure, I analyzed it, as the chemists who 
have studied this compound difier as to its composition. I 
repeated my analysis several times; and even changed my 
method of analysis. 
1. When a solution of sublimate is precipitated by ammonia, 
the mercury is contained in the precipitate, and a portion of 
chlorine remains in solution as sal ammoniac. One hundred 
grains of sublimate were dissolved in cold water, this solution 
was decomposed by a slight excess of ammonia, and the 
precipitate collected and washed on a previously weighed 
filter; it was dried and weighed with care; the washings were 
added to the first liquid, the whole acidulated with nitric acid 
and precipitated by nitrate of silver; the chloride was collected 
VOL. III. — ^NO. III. 32 
