In Paris I saw a fireplace so ingeniously contrived as to serve conveniently two rooms, a bedchamber and a study. The funnel over the fire was round. The fireplace was of cast iron (Plate, Fig. 13,) having an upright back A, and two horizontal semicircular plates B C, the whole so ordered as to turn on the pivots DE. The plate B always stopped that part of the round funnel that was next to the room without fire, while the other half of the funnel over the fire was always open. By this means a servant in the morning could make a fire on the hearth C, then in the study, without disturbing the master by going into his chamber; and the master, when he rose, could, with a touch of his foot, turn the chimney on its pivots, and bring the fire into his chamber, keep it there as long as he wanted it, and turn it again, when he went out, into his study. The room which had no fire in it was also warmed by the heat coming through the back plate, and spreading in the room, as it could not go up the chimney.

Description of a new Stove for burning Pitcoal, and consuming all its Smoke.

READ AT A MEETING OF THE AMERICAN PHILOSOPHICAL SOCIETY, JANUARY 28TH, 1786.

TOWARDS the end of the last century an ingenious French philosopher, whose name I am sorry I cannot recollect, exhibited an experiment to show, that very offensive things might be burnt in the middle of a chamber, such as woollen rags, feathers, &c., without creating the least smoke or smell. The machine in which it was made, if I remember right, was of this form, (Plate XV. Fig. 1,) made of plate iron. Some clear burning charcoals were put into the opening of the short tube A, and supported there by the grate B. The air, as soon as the tubes grew warm, would ascend in the longer leg C and go out at D, consequently air must enter at A descending to B. In this course, it must be heated by the burning coals through which it passed, and rise more forcibly in the longer tube, in proportion to its degree of heat or rarefaction, and length of that tube. For such a machine is a kind of inverted siphon; and, as the greater weight of water in the longer leg of a common siphon in descending is accompanied by an ascent of the same fluid in the shorter; so, in this inverted siphon, the greater quantity of levity of air in the longer leg, in rising is accompanied by the descent of air in the shorter. The things to be burned being laid on the hot coals at A, the smoke must descend through those coals, and be converted into flame, which, after destroying the offensive smell, came out at the end of the longer tube as mere heated air.

Whoever would repeat this experiment with success, must take care that the part A B, of the short tube, be

quite full of burning coals, so that no part of the smoke may descend and pass by them without going through them, and being converted into flame; and that the longer tube be so heated as that the current of ascending hot air is established in it before the things to be burnt are laid on the coals; otherwise there will be a disappointment.

It does not appear, either in the Memoirs of the Academy of Sciences, or Philosophical Transactions of the English Royal Society, that any improvement was ever made of this ingenious experiment, by applying it to useful purposes. But there is a German book, entitled Vulcanus Famulans, by John George Leutmann, P. D., printed at Wirtemberg in 1723, which describes, among a great variety of other stoves for warming rooms, one, which seems to have been formed on the same principle, and probably from the hint thereby given, though the French experiment is not mentioned. This book being scarce, I have translated the chapter describing the stove, viz.

"CHAPTER VII.

"On a Stove, which draws downwards.

"Here follows the description of a sort of stove, which can easily be removed, and again replaced at pleasure. This drives the fire down under itself, and gives no smoke, but, however, a very unwholesome vapor.

"In the figure, A is an iron vessel like a funnel, (Plate XV. Fig. 20,) in diameter at the top about twelve inches, at the bottom near the grate about five inches; its height twelve inches. This is set on the barrel C, which is ten inches diameter and two feet long, closed at each end E E. From one end rises a