106 On the Composition of the Topaz, [Fun. 
new analysis of the topaz, directing my principal attention to the 
extraction of potash from it. 
1 pulverized in an agate mortar a fine crystal of yellow topaz 
from Brazil, and exposed toa red heat a mixture of three grammes 
(462 grains) of it with 12 grammes of carbonate of barytes. The 
mass was then put into a platinum cup, and treated with muriatie 
acid, which dissolved it without the disengagement of any carbonic 
acid, leaving no other residue but pure silica, Into the solution I 
poured sulphuric acid as long as any precipitate fell, and even added 
an excess of that acid, and then evaporated the liquid till a portion 
of this excess was driven off. The residue was digested with water 
for 24 hours, and the liguid then precipitated by carbonate of am- 
monia added in excess, The filtrated liquid was evaporated to dry- 
ness. ‘The sulphate of ammonia thus obtained was put into a pla- 
tinum crucible, exactly weighed, and exposed to the heat of a spirit 
of wine lamp. The ammoniacal salt was volatilized, and left upon 
the sides of the crucible reddish stains, besides a trace of a saline 
matter at the bottom, The crucible had gained only six milli- 
grammes (0°0924 grain). I poured water into it. The red stains 
were not attacked; but the saline matter was dissolved. ‘The small 
quantity of liquid thus obtained was divided into two portions. ‘The 
one was evaporated to dryness, and the saline residue examined by 
the microscope. The crystals were irregular, and confounded 
together; and none of them bore any resemblance to sulphate of 
potash. When dissolved in water, and mixed with tartaric acid, I 
could perceive no trace of supertartrate of potash. ‘The other por- 
tion, being mixed with ammonia, let fall alumina. Hence it is 
evident that the salt under examination, at least the greatest part of 
it, was sulphate of alumina. I shall not determine whether it 
might contain any trace of alum, as it is evident that this can have 
no influence. 
I consider this experiment as a decisive proof that potash does not 
belong to the chemical constitution of the topaz, and that, if we 
find traces of that alkali in topazes of certain districts, it can only 
be considered as one of those foreign substances with which minerals 
are so frequently mixed, Jt can scarcely be doubted that topazes 
exist mixed with traces of felspar, just as there are crystals of nitre 
mixed with common salt, without our drawing as a conclusion from 
this that muriatic acid isa constituent part of crystallized saltpetre. 
You will permit me, perhaps, to add an observation with respect 
to the existence of these traces of sulphate of alumina, although 
the liquid had been precipitated by an excess of carbonate of am- 
monia. The cause of it is, that alumina is not absolutely insoluble 
in ammonia, Caustic ammonia dissolves a considerable quantity of 
it. A gros (72 grains) of caustic ammonia dissolves a large piece 
of alum without leaving any residue, and the alumina dissolved by 
ammonia is not precipitated till after a long continued ebullition. 
Carbonate of ammonia, on the other hand, dissolves so little alu- 
mina, that, unless we add a great excess of it, the dissolved portion 
