244 Count Rumford’s Experiments to determine 
corrosion of the hemisphere itself ; for as long as the bore re- 
tained its form and its dimensions, in that part to which the 
efforts of the elastic fluid were confined, that is, in that part of 
the bore immediately in contact with the lower part of the 
stopper, the experiment could not be affected by any imper- 
fection of the bore either above or below'. 
In the figures 2. and 4. this stopper is represented in its place, 
and fig. 6 . shews the plan, and fig. 7. the profile of one of these 
stoppers of its full size. Fig. 8. shews a small but very useful 
instrument, employed in introducing these stoppers into the 
bore, and more especially in occasionally extracting them : it re- 
sembles a common cork-screw, only it is much smaller. In the 
figure (where it is shewn in its full size), it is represented 
screwed into a stopper. Fig. 9. shews the plan, and fig. 10. a 
side view, of the hemisphere of hardened steel, by which the 
end of the barrel was closed. In the figures 2. and 4. the barrel 
is represented as being about half filled with powder. 
F*resuming that what has been already said, together with 
the assistance of the annexed drawings, will be sufficient to 
give a perfect idea of all the different parts of this apparatus, I 
shall now proceed to give an account of the experiments which 
from time to time have been made with it. And in order to 
render these details as intelligible as possible, and to shew the 
results of all these inquiries in a clear and satisfactory manner, 
I shall first give a brief account of the manner in which the 
experiments were made; of the various precautions used; 
and the particular appearances which were observed in the pro- 
secution of them. 
The powder made use of in these experiments was of the 
best quality, being that kind called poudre de chasse by the 
