t *5 6 j 
grain meafures : therefore the weight of the inflam- 
mable air was i 3*. grains i. e. of the weight of the 
tin, and its denfity 8918 times lefs than that of water. 
The quantity of moifture condenfed was about three 
grains. 
It is evident, that the truth of thefe determinations 
depend on a fuppofition, that none of the inflammable 
air is abforbed by the pearl-afhes. In order to fee 
whether this was the cafe or no, Idiffolved 86 grains of 
•zinc in diluted acid of vitriol, and received the air in a 
meafuring bottle in the common way. Immediately 
after, 1 dififolved the fame quantity of zinc in the fame 
kind of add, and made the air to pafs into the fame 
meafuring bottle, through a cylinder filled with 
dry pearl- afhes, in the manner reprefented in Fig. 5. 
I could not perceive any difference in their bulks. 
It appears from thefe experiments, that there is but 
little, if any, difference in point of denfity between the 
different forts of inflam mable air. Whether the 
difference of denfity obferved between the air pro- 
cured from zinc, by the vitriolic and that by the ma- 
rine acid is real, or whether it is only owing to the 
error of the experiment, I cannot pretend to fay. By 
a medium of the experiments, inflammable air comes 
out 8760 times lighter than water, or eleven times 
lighter than common air. 
In order to fee whether inflammable air, in the 
date in which it is, when contained in the inverted 
bottles, where it is in contad with water, contains any 
confiderable quantity of moifture ftiffolvedin it, 1 forced 
192 ounce meafures of inflammable air, through a 
cylinder filled with dry pearl-afhes, by means of the 
fame apparatus, which I ufed for filling the bladders with 
inflam- 
