latter fell into the mill-race, and were much injured. The effect, however, did not stop here; for the fire communicated, strange as it may appear, to some of the other mills, although at some distance, and blew them up. Several explosions have happened at the same mills. An experiment, made at the same works, by the then proprietor, the father of the late commodore Decatur, by putting the nitre, charcoal, and sulphur, into a barrel, with iron balls covered with lead, which was turned upon its axis, terminated in the same way. It exploded, but no other injury or accident was sustained. On examining the balls, we found, that the lead was entirely worn off, and the explosion must have been owing to the iron. This experiment was performed, in order to find if the mixture could be made in this manner, a plan which was afterwards adopted in France, with success, but brass balls were used. In a series of essays, which I wrote for, and published in, the Aurora, in 1808, on the "_Application of Chemistry to the Arts and Manufactures_," as manufactures are vitally important to the _practical_ independence of this country, I mentioned the subject of gunpowder, and the different modes of preparing it, and among which, the various experiments on this subject. The machinery, required in gunpowder mills, is exceedingly simple. The power of the water, which may be given by an overshot, or undershot wheel, is communicated to the parts of the mill, which perform the work. Thus it is, that pounders, like the snuff, or plaster-paris mill, are put in motion, by a horizontal shaft, furnished, at different distances, with pieces of wood, which, by the revolution of the shaft, and meeting with the projecting pieces from the pounders, raises them in succession. They fall, then, in the same order of succession, in the respective mortars.