^45 
which depend on Weak Affinities, 
pitate contained. Magnesia, 43.16; carbonic acid, 36.47; wa- 
ter, 20.37. The superior quantity of magnesia obtained in this 
experiment, seemed likely to have proceeded from a little pot- 
ash carried off* by the precipitate. To ascertain this, I diges- 
ted the caustic magnesia produced by the analysis in boiling 
water. The liquid had a distinct alkaline taste, and restored 
the blue colour to turnsol-paper. After being saturated with 
muriatic acid, the liquid was then evaporated, and the remain- 
der heated to redness. This residuum was not dissolved by a 
few drops of water which were poured upon it ; but these drops 
when evaporated, left on the glass a white pellicle, which 
was sparingly soluble in water, and which, as it appeared com- 
pletely earthy, when viewed through a compound microscope, 
could be nothing but magnesia. From this fact, it seemed 
reasoiiable to conclude, that the magnesia produced by the ana- 
lysis, had contained no sensible trace of potash. My next 
portion of magnesia alba, was obtained in the form of a precipi- 
tate, by boiling a liquid carbonate of the earth. It gave. Mag- 
nesia, 43.2 ; carbonic acid, 36.4 ; water, 20.4. A second ex- 
periment, made upon another portion prepared in the same 
manner, at a different time, gave, Magnesia, 42.8 ; carbonic 
acid, 36.5 ; water, 20.7. All these analyses appear to agree in 
indicating, that magnesia alba contains about 43 per cent, of 
magnesia, and 20.5 per cent, of water. The result of M. Klapr 
roth’s analysis cannot, therefore, be regarded as correctly exhi- 
biting the composition of this substance. But, on the other 
hand, no one of my own analyses agrees with the laws of chemi- 
cal proportions, or gives a probable formula of composition. 
Yet, as no anomalous deviation from a general law can ever be 
admitted, it seemed probable that some circumstance had eluded 
my attention. Having, for this reason, resumed my experi- 
ments, I first examined the carbonate of magnesia, precipitated 
by the carbonate of potash in a cold state. It gave me. Mag- 
nesia, 36.40 ; carbonic acid, 30.25 ; water, 33.35. According 
to these proportions, the quantities of oxygen contained in the 
acid and water, are (with a slight error in excess) respectively 
equal to one and a half and two times the oxygen of the base. 
Though the liquid from which the precipitate had been deposi- 
ted still contained an excess of sulphate of magnesia, it exhibit- 
