DISCHARGING OF ORDNANCE, 501 



a tube 15 inches deep, pierce a piece of flannel, and 

 fire gunpowder. And supposing the tube to re- 

 present the touch-hole of a gun, and the flannel and 

 gunpowder to be a substitute for the cartridge, then 

 we may conclude that the gun would be discharged, 

 although the cartridge were 15 inches distant from 

 the fulminating powder, which will never occur, 

 even in the largest pieces of ordnance. An objec- 

 tion, however, arises to the above conclusion ; that 

 it may be owing to the tying being very close, and 

 the flame having no room to spread, that the gun- 

 powder was inflamed in experiments first and se- 

 cond, and that when applied to the gun the flame 

 may be lost over the surface of the cartridge. It 

 becomes necessary to answer this objection, and 

 this is done by the two following experiments. 



Experiment 3. — A quantity of gunpowder was 

 scattered over the bottom of a circular tin canister. 

 The canister was 8 inches deep, and nearly S 

 in diameter. Over the powder was laid a piece of 

 cartridge flannel. The tube (A B) was made to 

 descend into the canister, to within 2 inches of the 

 flannel ; and then a grain of the fulminating pow- 

 der was exploded at the top. The result was quite 

 satisfactory, the flame pierced the loose flannel, and 

 fired the gunpowder. 



Experiment 4. — A tube 14 inches long, was 

 bent at the 10th inch from the top, so as to present 

 4 inches out of the straight line in which the flame 

 formerly proceeded, and through which it must now 



