7-^ M. Berzelius on Compowids 
discover the exact quantity of the sulphuric acid. It is well 
known, that the oxides of gold and platina allow themselves to 
be entirely precipitated, if, to a mixture of their solutions and of 
sulphuric acid, v/e add muriate of barytes. I know not the na- 
ture of that aninity by which the sulphate of barytes carries these 
oxides along Vvith it. I dare not hazard any conjecture respect- 
ing the nature of their union ; but it does not seem improbable, 
that several substances, which in mineralogy are considered as 
foreign, may have been introduced by a similar affinity. 
But, to return from this digression, to examine the analysis of 
our double salt : — I have said that we ought likewise to know 
the composition of neutral carbonate of magnesia. I procured this 
salt by allowing a solution of magnesia in liquid carbonic acid to 
evaporate spontaneously. The carbonate deposited itself on the 
bottom and sides of the glass, in the form of small pellucid cry- 
stals, which I dried upon blotting paper. The dry salt w^as next 
introduced into a small apparatus, such as I have already de- 
scribed, and Iieated by the dame of a spirit-of-wine lamp. At 
the first application of the heat, the salt gave out a great quanti- 
ty of water, and became milk-W'hite, but preserved the form ol‘ 
its crystals. This salt has the property of efflorescing in drj 
air, where it loses its water of crystallization, without losing any 
part of its acid, as I have proved by a direct experiment. 1 
make the observation in tliis place, because it might be imagined 
that tlie efflorescence of the salt was in reality only a transforma- 
tion of it into magnesia alba. 
The salt contained in the small eornute w^as kept in the dame, 
till it had been red for a quarter of an hour. The tw’o recipi- 
ents had gamed 38.9 per cent, of winter. The glass bulb wffiich 
held the magnesia was anew exposed to a stronger heat, in a 
crucible of platina, among burning coals. There remained in 
it 29.6 per C{'Qt. of magnesia, entirely deprived of carbonic acid. 
The loss, 31.5 per cent., was therefore carbonic acid. The 
portions of oxygen contained in these quantities of magnesiaj 
carbonic acid, and v/ater, are 11.457, 22.89, and 54.33 ; or pro- 
portional to 1, 2, and 3. Consequently, the acid contained 2, 
and the water 3 times the oxygen of the base. The composition 
neinral carbonate of magnesia may therefore be expressed by 
