which depend on Weak Affinities. 
the formula. Mg C ^ + 6 Aq ; whence we infer that it con- 
tains, 
Magnesia, - - 29.583 
Carbonic acid, - - 31.503 
Water of combination, 38.914 
It still remained for me to determine the quantity of water, 
combined ¥/ith the crystals of the bicarbonate of potash. ^Vhen 
treated in the same apparatus as the salts akeady analyzed, 
those crystals yielded 9 per cent, of water, and 69 per cent, of 
carbonate of potash. This result agrees exactly with the for- 
mula, KC^ -{- 2 Aq *f*, which denotes that the carbonic acid 
contained 4 time^ the oxygen of the base, as well as of the 
water. 
These different points being adjusted, we again proceed to 
examine the result of our analysis of the double carbonate, in 
order to discover its true chemical composition. The 18.28 
parts of potash contain 3.0987 parts of oxygen ; and the 15.99 
parts of magnesia contain 6.1894, or twice as much. The 
34.49 parts of carbonic acid contain 25.57 parts of oxygen, or, 
with a slight error, 8 times the oxygen of the potash. If we 
seek to connect the acid with its two bases, we shall find, that it 
must be divided between them, in such a manner that they may 
contain equal quantities of it ; the potash, however, being at a 
higher point of saturation, or forming a bicarbonate, whilst the 
magnesia fnins only an ordinary carbonate. This becomes evi- 
dent, if we consider that the hicarhoiiate of potash was employ- 
ed in producing the salt ; and that, if the carbonic acid ^rere 
divided proportionally between the bases, it would give degrees 
* By applying the law stated in page 65. note 2. the quantities found by the 
experiment are sufficiently correct to indicate that, in conformity with this Ian', 
Mg C '2+ 6 Aq, is the exact formula ; or that (1 atom of magnesium with 2 of oxy- 
gen, or) 1 atom of rhagnesia joined with (2 atoms of carbonium, each containing 
2 of oxygen, or with) 2 atoms of carbonic acid, must be combined with 6 atoms of 
water, in order to form carbonate of magnesia.' But the weight of these several 
atoms being already ascertained, by numerous and varied experiments, the error 
of the analysis is rectified accordingly — Transl. 
t The meaning of this expression may be collected from the preceding Note^ 
that t© which it refers. — T raksi,., 
