280 Count Rumford's Experiments to determine 
the coefficient above found (=6.3744,), gives 101021 atmo- 
spheres for the measure of the initial force of the elastic fluid 
generated in the combustion of gunpowder. 
Enormous as this force appears, I do not think it over-rated ; 
for nothing much short of such an inconceivable force can, in 
my opinion, ever explain in a satisfactory manner the bursting 
of the barrel so often mentioned; and to this we may add, that, 
as in 7 different experiments, all made with charges of 12 grains 
of powder, there were no less than 5 in which the weight was 
raised with a report , and as the same weight was moved in 3 
different experiments in which the charge consisted of less than 
12 grains, there does not appear to be any reason whatever for 
doubt with regard to the principal fact on which the above 
computation is founded. 
There is an objection, however, that may be made to these 
decisions respecting the force of gunpowder, which, on the first 
view, appears of considerable importance; but on a more care- 
ful examination it will be found to have no weight. 
If the force of fired gunpowder is so very great, how does it 
happen that fire-arms and artillery of all kinds, which certainly 
are not calculated to withstand so enormous a force, are riot 
always burst when they are used ? I might answer this ques- 
tion by another, by asking how it happened that the barrel used 
in my experiments, and which was more than ten times stronger 
in proportion to the size of its bore than ever a piece of ordnance 
was formed, could be burst by the force of gunpowder, if its 
force is not in fact much greater than it has ever been supposed 
to be ? But it is not necessary to have recourse to such a shift 
to get out of this difficulty : there is nothing more to do than 
to show, which may easily be done, that the combustion of 
