OF GUN-POWDER. IZ
44. The following experiment will prove that the degreeof heat to fire powder should be greater in proportion tothe rarefaction of the surrounding air, or to the density ofthe smoke. (35. N°2.)
Plate I. Fig. II. To the plate A B of the pneu-matic machine, let a thin thimble C of beaten brass, bejoined, with the convexity upwards. Let the iron Mexactly fill the concavity D ; and a moveable ring, F G,be fitted round the thimble, so that on covering the con-vexity C with powder, part may fall on it. Then ifM, made red-hot, be introduced into D, when F C Gis covered with powder, in the open air, the powder willinstantly explode; but if a glass receiver be placed upon theplate A B, and in part exhausted, and the hot iron be ap-plied, the explosion of the powder on C will not be sosudden as before, and part of that on F G will remainunaffected by the heat, in proportion to the quantity ofair left in the receiver. When the air is quite exhausted,the powder upon the convexity C will liquefy before ittakes fire, after having emitted several ebullitions; thenexplode at once, and cover with its flame the powderupon F G : this will, nevertheless, not liquefy; and a partof it will remain upon the ring and plate, where it has beenpushed by the blast.
45. This proves that the powder upon the ring, being ina very rarefied medium, does not burn, though surroundedby the flame of the fired powder. In this cafe, the findpowder spreading itself on all sides in the receiver, whichis much larger than the space occupied by the powder, thegrains that remain unfired are only surrounded by a part ofthe flame. If the receiver were filled with powder, the fireinclosed in a smaller space would be more intense, but thequantity of fired powder would be proportioned to therarefaction of the atmospheric air contained between thegrains. (Fig. III.) ABC is a funnel of bronze or iron,of which the part C C is made to screw on to V V ot a re-ceiver of beaten copper V X Z, containing about 70 cubicinches. The part D F of the funnel, which is to containthe powder for the experiment, may be diminished at plea-sure, by means of the rings G, of exact calibre, and of dif-ferent lengths, to be applied to the part F of the space.D F.A HI is a canal conical from A to H, and C) lindrical from
H ta