Analyfis of the Air . \6<y 
the other leg of it out of the bolthead, by 
which means the water continued up to 
and could not fubfide. Then I placed un- 
der the bolthead, while it was in the water, 
the vefiel x x> which done, I lifted the vef- 
fel x x y with the bolthead in it, out of the 
water, and tied a waxed thread at z> to mark 
the height of the water : And then ap- 
proached the retort gradually to the fire, 
taking care to fereen the whole bolthead 
from the heat of the fire. 
The defeent of the water in the bolthead 
{hewed the fums of the expanfion of the 
Air in the retort, and of the matter which 
was diftilling : The expanfion of the Air a- 
lone, when the lower part of the retort 
was beginning to be red hot, was at a me- 
dium, nearly equal to the capacity of the 
retorts, fo that it then took up a double 
fpace; and in a white and almoft melting 
heat, the Air took up a tripple fpace, or 
fomething more: for which reafon the leaft 
retorts are beft for thefe Experiments. The 
expanfion of the diftilling bodies was fome- 
times very little, and fometimes many 
times greater than that of the Air in the 
retort, according to their different natures. 
M 3 When 
