Carbonic Acid and Ammonia. 25 
tic state, and the latter the more so of the 
two, being much the lighter, yet it is the 
acid rather than the alkali that escapes. But 
the reason is pretty obvious. ‘The acid has 
no affinity for water; but the alkali has a 
strong affinity for the water and is therefore 
retained by it. 
We now proceed to the subcarbonate of 
‘ammonia; or the salt with excess of base. 
_ Equal weights of the three samples of sub- 
carbonate mentioned above were dissolved in 
water and immediately saturated by test acid ; 
the first took 17 grains of acid, the second, 
15 grains, and the third, 12 grains. Hence 
we see that the most volatile of the salts was 
that which abounded most with ammonia, as 
indeed might be expected. Other portions 
of these samples were afterwards analysed in 
. the same manner as the carbonate above. 
The first had nearly two atoms of ammo- 
nia to one of acid; the second had less than 
two, and the third had rather more than one 
atom of ammonia for one of acid; so that the 
salts were evidently mixturesof carbonate and 
subcarbonate (the last being here understood 
to signify a compound of 1 atom of acid to 2 
atoms of base); it seems therefore most pro- 
bable that they were all originally subcarbo- 
nates 4n the last sense of the word, but that 
VOL. ILI, D 
