'[ ] 
•fubftances may, however, be employed for the fame pur- 
pofe ; but none, perhaps, are fo unexceptionable as thofe 
I have named. In the other acids a proper degree 
of fixity is wanting, during the effervefcence ; the ni- 
trous and marine have fo much volatility that there is 
always a rilk of fome of the acid fumes palling the valve, 
and thus rendering the water acid, which it was intended 
to impregnate only with fixed air. To begin the pro- 
cefs, it is neceflary to fill the veflel (A) up to the dotted 
lines, with diluted oil of vitriol. By confining the height 
of the furface of the effervefcing mixture to the dotted 
lines in the glafs (A), none of the acid will be driven 
through the valve, during the intumefcence that at-^ 
tends the efcape of the fixed air. The glafs (B) is to be 
•totally filled with water, and the veflel (G) is to be put 
on it. Some powdered chalk is then to be thrown 
into the glafs (A), and the velTels are to be imme- 
diately placed as in fig-. 5 . except that the ftopper be- 
longing to (C) is to be left out. When the acid, in the 
dowermofl: veflel, a<51:s on the chalk, the extricated air 
pafles the valve in the middle glafs; and as the con- 
fl;ru<5tion of this valve allows the fixed air from the effer- 
vefcing fubftances to pafs, but denies a paflage to the wa- 
ter in a contrary dire 6 tion, the feparated air afcends to 
the upper part of the middle glafs, and at the fame time 
a portion of water, equal in bulk to the intruding air, 
pafles up the bent tube into the uj^permoftvefTel. As the 
effervefcence goes on, the fixed air continues to accumu- 
late in the middle veflel, and the uppermofl; one to be 
fiUed 
