Dr Daubeny on s^pamting Lime from Magnesia. 315 
water, used for waging off the sulphate of magnesia, may be es- 
timated and allowed for. 
If this process be considered too tedious, we may content 
ourselves with dissolving the two earths in muriatic acid, decom- 
posing the solution by sulphate of ammonia ; and after suiferiitg 
the sulphate of lime to subside, decanting off the supernatant 
liquor, and throwing down the magnesia from the latter by car- 
bonate of ammonium and phosphate of soda. In this way an 
approximation to the real composition of the earth may be rea- 
dily obtained. 
It now only remains for me to speak of the re-agents, which, 
where no other base is present, may be employed for the pur- 
pose of precipitating magnesia, in the state either of a pure earth 
or a carbonate. 
We are commonly told, in Elementary Works on Chemistry % 
that the alkalies and alkaline carbonates, throw down all earthy 
and metallic salts ; the only exception made, being as to the car- 
bonate of ammonia, which, it is well known, has no such effect 
upon magnesia. 
Accordingly, we are directed in our analyses, to throw down 
the lime by means of carbonate of ammonia, and the magnesia, 
by adding subsequently the pure alkali ; whereas the fact is, 
that neither process effectually answers the purpose designed. 
The tendency, indeed, which magnesia has to form triple 
salts, or, in other words, its affinity for the alkalies, interferes 
oonsiderably with the result in all these cases ; for I have reason 
to believe, that, owing to this circumstance, a part of the mag- 
nesia is often retained in the solution with the alkaline salt, and, 
perhaps, a trace of the alkali carried xiown along with the earth. 
I believe it to be in part owing to this circumstance, that:! 
have failed in obtaining the quantity of precipitate from a given 
weight of sulphate of magnesia, which I had expected from cal- 
culation, when the reagent employed was either ammonia, soda, 
or its subcarbonate f . 
Whenever these substances were made use of, I have 
* See Henry’s Elements, &c. &c. 
t When ammonia was employed, the quantity of precipitate varied considera- 
bly, as Berzelius has also found, (Spe his paper on Compounds, which depend on 
weak affinities, Edin. Phil. Jour. vol. i. p. 253.) With soda, the quantity obtain- 
