8o Experiments ^lpon Gunpowder. 



the line M Q be an hyperbola, the area P M Q B would 

 always be the same when the charge was the same, and 

 consequently the velocities of the bullets would be as 

 the square roots of their weights inversely. But it ap- 

 pears, from the before-mentioned experiments, that when 

 the weight of the bullet was increased four times, the 

 action of the powder, or area P M Q B, was nearly 

 doubled; for in the 92d experiment, when four bullets 

 were discharged at once, the collective pressure was as i ; 

 but in the 89th experiment, when a single bullet was 

 made use of, the collective pressure was only as 0.5825 ; 

 and in the 85th, 86th, and 8yth experiments, when the 

 bullets were much lighter, the action of the charge on 

 them was still less. 



But though we can determine with great certainty, 

 from these experiments, the ratio in which the action of 

 the -powder upon the bullet was increased or diminished^ by 

 making use of bullets of greater or less weight, yet we 

 cannot from thence ascertain the relation of the elasticity 

 of the generated fluid to its density, nor the quantity of 

 powder that is inflamed at different periods before and 

 after the bullet begins to move in the bore. 



But assuming Mr. Robins's principles, as far as re- 

 lates to the elasticity of the fluid, and supposing that in 

 all the experiments except the 92d, a part only of the 

 charge took fire, and that that part was inflamed and 

 converted into an elastic fluid before the bullet began to 

 move, upon that supposition we can determine the 

 quantity of powder that took fire in each experiment, 

 for the quantity of powder in that case would be as the 

 collective pressure. 



Thus if the whole charge, = 145 grains in weight, is 

 supposed to have been inflamed in the 92d experiment, 



