ON POTASH AND SODA IN SOILS. 
319 
pie chloride of platinum, because an excess of this salt must be 
employed, which is easily decomposed by alcohol, in which, 
however, the chloride of platinum and sodium is soluble with- 
out decomposition." 
Again, when magnesia is likewise present, he sa)'s: 
" When a mineral contains magnesia, we always obtain 
this earth as a sulphate of magnesia, combined with sulphate 
of alkali. Its presence is discovered by pouring into the 
concentrated solution of the salt a large excess of strong caus- 
tic ammonia, which precipitates part of the magnesia. In 
this case, we evaporate to dryness, in order to volatilize the 
ammonia, and we treat the residual salt (as we have just said) 
with acetate of barytes. If, after the combustion of the acetic 
acid, the residue is treated with water, this liquid dissolves 
the alkali and leaves the magnesia mixed with carbonate of 
barytes. This earth may be separated by means of a sulphuric 
acid." 
The latter quotation refers solely to the mode of separating 
the magnesia : if the portion dissolved by the water should 
contain both alkalies, the operose process described in the 
first, will have to be performed. 
My method is founded upon the property of a saturated 
solution (at a given temperature) of sulphate of potassa being 
capable of dissolving sulphate of soda, while it is of course 
totally incapable (the temperature being the same) of dissolv- 
ing any more of the sulphate of potassa. This appears pretty 
evident, but it was necessary to prove, by actual experiment, 
whether the presence of a soda salt might not influence this 
insolubility, either by the formation of a double salt or from 
some other cause. I accordingly prepared some pure sul- 
phate of potassa by saturating exactly, crystallized bicarbo- 
nate of potassa with pure sulphuric acid, evaporating and ig- 
niting in a platinum crucible. I prepared also some pure sul- 
phate of soda, in a similar manner from pure carbonate of 
soda. 
I then mixed a known weight (nearly equal) of the two 
salts, by rubbing them to fine powder, and submitted the mix- 
ture for about an hour (taking care to keep the temperature 
