184 Experiments with Cannon, ^c. 



likely to be wanting in strength than larger cannon of a 

 similar form ; but I never could see any reason for sup- 

 posing that a small piece of iron ordnance would be 

 wanting in strength, if means were employed for press- 

 ing the metal with a considerable weight while it remains 

 in the mould in a fluid state ; and for prolonging the 

 time of its cooling. 



When cannon are cast of gun-metal, or brass, the 

 solid cylinder, or rather cone, which, when bored, con- 

 stitutes the gun, is cast at least 1 feet longer than it is 

 intended that the gun should be. The additional piece 

 of metal beyond the end of the muzzle, which is cut oiF 

 before the gun is bored (and thrown away, or rather pre- 

 served and melted again), serves to compress the fluid 

 mass in the lower part of the mould, and to force up- 

 wards those bubbles of air that are frequently carried 

 down into the mould with the descending current of 

 melted metal ; and which, if they were not expelled, 

 would form what are called honeycombs in the metal ; and 

 would render the gun unsound and unserviceable. 



The mould is always placed in a vertical position, 

 with that end upwards which is to be the muzzle of the 

 gun, in order that the soundest part of the metal — 

 namely, that which, having sustained the greatest press- 

 ure while the metal remained in a fluid state, is most 

 effectually freed of air-bubbles — may be in the neigh- 

 bourhood of the chamber of the piece, where strength 

 and soundness are most wanted. ^ 



The utility of these precautions being evident, they 

 were not neglected in casting the iron guns that were 

 constructed for my experiments ; and means were used 

 for prolonging the time of the cooling of the metal in 

 the mould. • 



