foration be covered with a valve, or little door, opening upward, as we press it down the elasticity will raise the valve and allow it to escape, and the piston will freely descend. If we attempt again to draw the piston up, we shall feel a strong opposition in the contrary direction. As the piston rises, an empty space is left between it and the bottom of the barrel; for the exterior air, in attempting to pass in, by its elastic force firmly closes the valve. If by strong muscular force we succeed in drawing the piston up, upon discontinuing the effort it will be forced down again to the bottom by the exterior pressure. Let us suppose now the bottom of the barrel to be fitted with the same kind of valve as the piston; as the latter ascends, the elasticity of the portion of air included in any vessel with which it may be connected will raise the valve, and the air will flow into the barrel: when the piston is again forced down, this valve will be closed by the force above it, and the included air will again pass off through the piston-valve, and we can repeat this operation, gradually withdrawing barrelful after barrelful of air from any vessel till the residual air becomes so reduced as not to have elastic force enough to raise the valve. Two such barrels so fitted with pistons, with a mechanical apparatus for alternately raising and depressing them, constitute the essential parts of the airpump.(4)

(4) a and b in the opposite figure represent the cylinders, in which the pistons, c and d, are accurately fitted; at the bottoms of the cylinder are the valves h, h, opening upward; and in each piston is a valve g, g, also opening upward; the bottoms of the cylinder are connected by a tube, e, with the pump-plate, upon which stands the bell-glass f; in the cylinder a, the piston is represented in the act of ascending, when the valve g is closed, and a vacuum would be formed beneath the piston but for the opening of the valve h, by the elasticity of the air in the receiver. In the cylinder b, the piston is in the act of descending when the valve h is closed, and the valve g open, by which all the air in the cylinder is forced out; and in this man

44. Now if we take a thin glass globe or flask, fitted with a stopcock, we may, by these means, exhaust it of the greatest part of the air which it contains, and equipoise it upon the balance. Upon opening the stopcock, air will rush into the empty vessel, and it will preponderate; and it will require a considerable weight in the opposite scale to restore the equilibrium. To ascertain the exact ner a portion of the air is withdrawn from the receiver f, at every stroke of the pump.

weight of any given volume of air, it will be necessary to measure it, and this we cannot do by measuring the capacity of the globe, for the best airpump will always have a residual quantity after exhaustion; but by connecting the exhausted vessel with an accurately graduated jar,(5) standing upon the water-bath, the air may be able to enter from the latter, and the rise of the water into the jar will indicate the exact quantity which has been thus abstracted. By careful experiments, conducted upon this principle, it has been found that 100 cubic inches of atmospheric air, under standard circumstances, to which we must hereafter advert more particularly, weigh 31 grains, or 815 times less than an equal bulk of water. We shall find that there are

many different kinds of aeriform matter, differing very greatly in their specific gravities; this is the mode by which they may be ascertained, and atmospheric air is the standard to which they are all referred, just as the specific gravities of solids and liquids are compared with water.

In the following table are included the weight of 100 cubic inches of the lightest and heaviest known forms of matter: of the same quantity of atmospheric air; and of water in its three physical states. The specific gravity of each, compared with air and water, is also shown.

(5) a represents the air-jar, graduated into cubic inches and parts of a cubic inch; b the glass balloon; each is fitted with a stopcock and conectingpiece.

§ 45. The weight or pressure of the atmosphere was first, however, demonstrated in a different way by a celebrated Italian philosopher, named Torricelli, in the year 1643. His attention was drawn to the subject by the attempt of a well-digger, at Florence, to raise water by a sucking-pump to a height exceeding 33 feet. The rise of water in a tube by these means had, up to that time, been ascribed by philosophers (who, as it is to be feared sometimes now happens, disguised their ignorance under the cloak of indefinite and specious expressions) to Nature's abhorrence of a vacuum. The well-digger failed in his enterprise, and applied to Torricelli for advice; who, seeing the absurdity of the conclusion that nature only abhorred a vacuum to the extent of 33 feet, suspected that the cause of the ascent of water in the pump-pipe might be the pressure of the atmosphere; and that a column of water about the height mentioned was sufficient to equipoise the air.(6)

P

M

ત

D N

(6) The annexed figure represents the construction of the common sucking-pump; A B is the barrel fitted with a piston, in which is a valve at A, opening upward; another valve E, at the bottom of the barrel, also opening upward, closes the communication with the pipe, B D, which descends

He concluded that, if this were the case, it would only support a shorter column of any denser fluid, and he immediately had recourse to experiment to confirm his conjecture. He filled a glass tube three feet long, and closed at one end, with quicksilver, and inverted it in a basin of the same fluid; it immediately sank about six inches from the top of the tube; proving that the pressure of the atmosphere which could support a column of water of about 33 feet in height, could only support a column of mercury of 30 inches, the height of the columns being in exact proportion to the specific gravities of the two liquids, or as 13 1-2 to 1.(7)

46. Thus was invented that useful instrument the barometer; for a tube filled with due precautions wholly to exclude the air, and accurately adjusted to a scale for the purpose of measuring the exact height of the column from the surface of the mercury in the cistern, constitutes the essential part of this simple but highly ingenious contri

below the level of the water, м N. As the piston rises, a partial vacuum is formed beneath it, and the superior elasticity of the exterior atmosphere pressing upon the water м N, forces it to ascend into the pipe from D to c. The descent of the piston closes the valve E, forces out another portion of the air through the valve A, and upon its return the valve E again opens, and the water is forced past it into the barrel. Another stroke of the pump drives the water past the valve A also, and lifts it into the reservoir L G, from which it flows through the aperture P.

(7) H B represents a tube, which, having been filled with mercury and closed with the finger, has been carefully inverted beneath the surface, D C, of the mercury in the glass basin, D C F E. Upon removal of the finger from B, the mercury has fallen in the tube from H to G, the column c of mercury, A G, being the exact equipoise of E the elasticity of the atmosphere at the time of the experiment