Dr Daubeny m separMing Lime from Magnesia,- II'l 
.operator, in applying heat, is placed in the dilemma of either 
not driving off* the whole of the water contained in the sulphate 
of magnesia, and I may add, of not subliming all the muriate 
and sulphate of ammonia present, or of separating a part of the 
acid combined with the magnesia. 
In the former case, he v/ill of course overrate the quantity of 
magnesia present, by regarding the sulphate as dry, whereas it 
will contain, according to M. Longchamp, nearly J^th of its water 
of crystallisation, and in the latter he is in danger of estimating it 
too low, not merely from considering the sulphate as though it 
were a neutral instead of a subsalty but also from confounding a 
part of it with the sulphate of lime left behind, owing to the 
very sparing solubility of the magnesian^ salt, after exposure to 
a considerable heat. 
In order to ascertain the correctness of this statement, I be- 
gan by submitting a given quantity of crystallised Epsom salt 
to a temperature not exceeding 212°, and found that it was re- 
duced from 20 grains to 14.44, and that the residuum dissolv- 
ed readily in water, 
I then exposed the same quantity of the salt to a strong red 
heat for about three hours, at the end of which time it weighed 
only 8.3 grains., and when treated with water, left an insoluble 
powder, amounting to 0.7. It was evident from this, that the 
salt had undergone some change from the heat which had been 
applied ; and that this was in part at least connected with a loss 
of acid, I proved by comparing the quantity of precipitate that 
could be obtained, by adding muriate of barytes to the solution 
of the salt which had been submitted to a temperature of 212°' 
only, and that which had been exposed to a red heat. 
From the former I obtained no less than 20.15 grs. of sul- 
phate of barytes, whilst the latter yielded only 15.7 grs. Great 
care was taken that both precipitates should be dried under pre- 
cisely similar circumstances, so that the ratio which they bear 
to each other may be relied upon as correct ; but it seems pro- 
bable that neither of them was reduced to the extreme point of 
dryness, as 20.15 grs. of sulphate of barytes would indicate 6.8 
gr. of sulphuric acid, whereas 20. gr. of crystallised sulphate 
of magnesia contain only 6.5, and ought therefore to form riQ 
more than 19 M gr. of the barytic salt. 
