Experiments upon Gunpowder. 33 



sideration, I am now convinced that it is what ought to 

 happen, and that it may be accounted for very well, 

 upon principles that are clearly admissible. 



It is certain that a very small part only of the heat 

 that a piece of ordnance acquires in being fired is com- 

 municated to it by the flame of the powder, for the 

 time of its action is so short (not being, perhaps, in 

 general, longer than about 2-g-oth or y^oth part of a 

 second) that if its heat, instead of being 4 times, as 

 Mr. Robins supposes, was 400 times hotter than red- 

 hot iron, it could not sensibly warm so great a mass of 

 metal as goes to form one of our large pieces of cannon. 

 And besides, if the heat of the flame were sufficiently 

 intense to produce so great an effect in so short a time, 

 it would certainly be sufficient, not only to burn up all 

 inflammable bodies that it came near, but also to melt 

 the shot that it surrounded and impelled, especially 

 when they were small, and composed of lead, or any 

 other fusible metal ; but so far from this being the case, 

 we frequently see the finest paper come out of the 

 mouth of a piece uninflamed, after it has sustained the 

 action of the fire through the whole of the bore, and 

 the smallest lead shot is discharged without being 

 melted. 



But it may be objected here, that the bullets are 

 always found to be very hot, if they are taken up im- 

 mediately after they come out of a gun ; and that this 

 circumstance is a proof of the intensity of the heat of 

 the flame of powder, and of its great power of com- 

 municating heat to the densest bodies. But to this I 

 answer, I have always observed the same thing of bul- 

 lets discharged from wind-guns and cross-bows, especial- 

 ly when they have impinged against any hard body 



VOL. I. 3 



