Experiments zLpon Gunpowder. 



19 



admit of a greater distance, and I was unwilling to ex- 

 pose any part of the apparatus in the open air. 



But the screen was found to answer perfectly well the 

 purpose for which it was designed, and it was continued 

 during the remainder of the experiments ; the paper 

 being replaced every third or fourth experiment. 



T^he Experiments continued. 



The bullets were now put naked into the piece, and 

 the powder was lighted by the short vent-tube {v, Fig. 

 3), and some little improvement was made in the steel 

 edges between which the ribbons passed that served to 

 measure the ascending arcs of the pendulum and of the 

 recoil, by which means the friction was lessened, and 

 the ribbon was prevented from twisting or entangling 

 itself as it was drawn out. 



Apparatus. 

 The barrel with its carriage as before. The pendu- 

 lum No. 3, and leaden bullets weighing 580 grains each. 



