168 RESULTS OF A SERIES OF EXPERIMENTS 



space already filled with common air. None of the high authorities we have mentioned 

 appear to doubt the truth of the fact which, from its paradoxical nature, and the contra- 

 dictory explanations which have been given of it, Avould seem to require a more sub- 

 stantial basis than that on which it appears to rest. The experiments of Robins were on 

 a small scale, with a musket barrel, and none of them appear to have been instituted 

 expressly for the purpose of settling the point in question. It is true that the opinion 

 would seem to be supported by common observation, for undoubtedly guns do burst more 

 frequently in cases where the ball is not down on the powder, than under other circum- 

 stances, but it should be recollected that the bursting, in every instance of this kind, may 

 be due to the fact that the ball is too large for the bore, and that in attempting to force it 

 down, it becomes wedged in the gun, and thus its resistance to dislodgement is mate- 

 rially increased. A comparatively small force applied to push down a ball too large for 

 the bore, would convert a portion of its spherical surface into the form of a cylinder, 

 which, by increasing the points of contact, would increase the resistance: we are aware 

 of the fact that ordinary friction does not increase with an increase of rubbing surface, 

 the total pressure remaining the same, but in the case of the ball the increase of the sur- 

 face of contact increases the wedging, which must increase the resistance in the same 

 proportion. From the foregoing considerations, the author of this paper was led to 

 doubt the truth of the proposition, and to institute a series of experiments which should 

 give it a more thorough test than any to which it before had been subjected. 



For this purpose, a number of guns, of a calibre two inches in diameter — thirty-five 

 inches long; of a uniform thickness of metal of one inch around the bore, and two inches 

 of solid metal behind the bottom of the cavity, were cast from the same fount of iron — 

 and were, consequently, nearly of the same strength. The weight of each was between 

 ninety-three and ninety-six pounds, and they were secured, in succession, to a large 

 block of timber, for the purpose of bursting. The charge consisted of cannon powder 

 done up in the usual way, in light flannel cartridges; no wads were used, and the ball 

 covered with a coating of felt, to lessen the windage, was placed in contact with the end 

 of the cartridge, or at different distances from the same, the intervening space being 

 occupied only with air. In all cases of a variation in the charge of powder, an increase 

 was made of one-fourth of an ounce each time, until the desired effect was produced. 

 The following are the particulars of the several experiments. 



1. In the first experiment, the bursting of the gun took place at the first dis- 

 charge, with seven ounces of powder, and the ball in contact with the cartridges. A 

 smaller charge would have been first used, had it not been supposed, from some previous 

 experiments with another set of guns, that this piece would have stood the quantity of 

 powder with which it was loaded. 



2. In another gun, the ball was constantly placed in contact with the cartridge. The 

 firing was commenced with five ounces of powder, and the bursting took place at the 

 fifth discharge, with six and a quarter ounces. 



3. The ball in the next gun was placed first at twelve inches from the powder, then at 

 four inches, and lastly in contact with the same. The experiment commenced with six- 

 ounces and a quarter of powder, and the explosion took place at the sixth discharge, with 



