Experiment 3.--Made at Woolwich Arsenal, with a Gomer mortar, the chamber being bored conically, so that the shell, when dropped in, fits closely all round, instead of being bored cylindrically, with a chamber in the centre. The mortar being laid at an angle of 45°, one drachm of powder was put into the bottom, and a 68-pounder iron shot over it. When fired, the ball was projected two feet clear of the mortar. A wooden ball, precisely the same diameter, but weighing only 5lbs., was scarcely moved by the same charge, and with two drachms of powder it was just lifted in the mortar, and fell into its place again. Here we find a weight of 68lbs. thrown to the distance of two feet by the same power which would not lift 5lbs., and the wooden ball scarcely moved by double the powder. This proves that the firing of gunpowder under such circumstances is not instantaneous. In the first instance, the small quantity of powder had a large space to fill below the ball, and a heavy weight to move; therefore, could not stir it at all until the whole was ignited, when the force was sufficient to throw it forward two feet. In the second case, the first portion of gas that was generated by ignition of the powder, was sufficient to lift the lighter weight, just enough to allow all the force to escape round it before it had time to accumulate. Experiment 4.--A cannon ball, weighing 24lbs., was placed exactly over the vent-hole of a loaded 32-pounder cannon, which was fired by a train of gunpowder, when the rush from the vent projected the 24-pounder ball to a very considerable height in the air, although the diameter of the hole was only two-tenths of an inch. Experiment 5.--A most ingenious method of ascertaining the relative quickness of ignition of different qualities of gunpowder.