derable degree, it will be difficult to remove the hand without violence, or without re-admitting the air. 3. Placea small receiver, O fig. 39, over the hole of the pump-plate, and, upon exhausting the air, the receiver will be fixed down to the plate by the pressure on its outside; then, by turning the cock of the pump, and re-admitting the air, the receiver will become loose. In order to prove that the receiver O is held down by the pressure of the air, suspend it on the hook of the wire PP passing through the collar of leathers at the top of the receiver M, by which it is covered, and thus let it down on the plate of the pump; and when the air is exhausted from both receivers, the large receiver M will be fixed to the plate by the pressure of the external air; but the small one will be loose and easily removed; then again, in letting in the air, the lesser one will be fixed to the plate, and the other will be released. 4. The same may be otherwise shown by the glass employed in experiment 2, viz. by tying over the hole at top a piece of bladder; then, as the air is exhausted, the bladder will be pressed into the receiver in a concave hemispherical form, and by carrying the exhaustion to a sufficient extent, the bladder will at length burst with a loud report; or, instead of the bladder, a piece of common windowglass may be employed, resting on a rim of leather; when it will be found that the pressure of the external air will break the glass as soon as the air is exhausted. 5. Join together the two hollow brass hemispheres A and B, fig. 40, the edges of which must be made very perfect, or otherwise they must have a circle of wet leather placed between them; screw the end D into the plate of the pump, and open the stop-cock E of the pipe CD communicating with the hemispheres; and having exhausted the air, turn the cock again so as to stop the pipe; remove the ball from the plate, and screw on at the end D. the handle FH. Two strong men now, one taking hold of each handle, and pulling straight against each other, will find considerable difficulty, or will be, perhaps, wholly unable to separate the two hemispheres. The degree of strength that will be requisite depends upon the diameter of the sphere and the nature of the exhaustion. If the latter be nearly complete, and the diameter be four inches, the area of the section will be

42 x .7854 12.5664 inches;

the pores of wood, and made to pass through it by the pressure of the atmosphere. 7. Immerse the neck c d of the hollow glass ball e b, fig. 41, in the water of the vessel a a; place it on the plate of the pump, and cover it and the hole of the plate by the receiver A; exhaust this receiver, and the air will escape by its spring, from the ball e b, through the neck c d, rise in bubbles through the water, and pass off into the external air. When the water has done bubbling, turn the cock of the pump, and the air that is admitted will, by its pressure on the surface of the water, force it up in a jet into the ball e b, and nearly fill it; the small quantity of remaining air which occupied the whole ball, and which is now reduced to a small space by condensation, is all that prevents the water from filling the whole cavity of the ball. This experiment may be varied by screwing the end A of the brass pipe ABF, fig. 42, into the hole of the pump-plate, and placing it, by means of a circle of leather, upon the plate cd, a tall receiver GH, close at top, exhausting the receiver of air, and stopping the pipe by the cock e: when this is done, remove the apparatus from the pump, set its end A in a basin of water, and open the pipe by turning the cock e, when it will be found that the pressure of the air on the water will force it up through the pipe, so that it will ascend in a jet to the top of the receiver

Again, place the tall open receiver, A B, fig. 43, on the pump-plate over the jar D. containing quicksilver; the latter being placed near the hole or pipe communicating with the barrels. Into the plate C, placed upon the upper end of this receiver, introduce the open glass tube gf, immersed at its lower extremity in the quicksilver of the jar D, and screwed by a brass top annexed to it at h, to the syringe II, which is itself screwed to the plate C. By means of the ring I, draw up the piston of the syringe, and thus exhaust the tube of its air; and the quicksilver in the basin, pressed by the undilated air of the receiver AB, will ascend in the tube. That this ascent is owing to the pressure of the air, and not to what is vulgarly called suction, may be shown by exhausting the receiver of its air, which will cause the quicksilver to descend into the jar, and, by re-admitting the air, it will rise again in the tube, although the piston of the syringe be not moved. If the tube be thirtytwo or thirty-three inches in length, the quick

12.5664 x 15 188lbs.

for the force or strength requisite to produce the separation. 6. The pressure of the air may be also shown as follows: Set a square phial of thin glass upon the pump-plate, and, to prevent accidents, cover it with a wire cage, and place both under a close receiver. The phial must be supplied at top with a small valve opening upwards, so as to be exhausted with the receiver, but which shuts and prevents the air from afterwards entering. Let the air be exhausted from the receiver and phial, and it will be found that, upon re-admitting the air into the former, the latter will be crushed into a number of small pieces. Quicksilver may also be forced into

stands at that time in the barometer; and if the syringe have a small hole at m, and the piston be drawn up above that hole, the air will pass through it into the syringe and tube, and the quicksilver will immediately fall down into the jar.

Once more: the jar A, fig. 44, being filled with quicksilver, and placed on the pump-plate, cover it with the receiver B, and push the open end of the glass tube de through the collar of leather in the brass neck C, almost down to the quicksilver in the jar; then exhaust the receiver В of its air, and the tube d e, which is close to the top f, will at the same time be exhausted. When the exhaustion has been carried on to a sufficient extent, push the open end of the tube, into the quicksilver of the jar, and it will be

found, that although the tube is exhausted, the mercury will not rise in it, because there is no pressure on the surface of that in the jar; but upon admitting the air into the receiver, the quicksilver will immediately rise, and stand as high as it did, in consequence of the action of the syringe, in the preceding experiment. These latter experiments not only exhibit the weight and pressure of the atmosphere, but they also show that they are increased or diminished in proportion to the increase or decrease of the density of the air.

The elasticity of the air may be shown in the most satisfactory manner, by placing under the receiver of an air-pump a flaccid bladder, well tied and secured at its neck, so as to prevent the air within it from escaping. Exhaust the air out of the receiver, and the bladder, as the process proceeds, will continue to expand, til at length it will appear like one full-blown, and even burst if the exhaustion be carried on to a sufficient degree; but upon re-admitting the air, it will quickly return to its original flaccid state. In experiment 6, we supposed the glass ball, fig. 41, to be filled with water, with the exception of a small bubble of air; if, in this state, it be placed with its neck downwards into the empty jar a a, and covered with a close receiver, and the air of the latter be exhausted, the air-bubble will expand itself, and by its elastic force protrude the water out of the globe into the jar. The same may be shown by screwing the pipe A B, fig. 42, into the pump-plate, and by placing the tall receiver G H upon the plate cd; for now, exhausting the receiver, remove the apparatus, and screw it into the copper vessel C C, fig. 45, half filled with water: then turn the cock e, fig. 42, and the air confined in this vessel will, by its spring, force the water through the pipe A B, and cause it to form a jet into the exhausted receiver equal to that which was produced by the pressure of the air in the former experiment. The little amusing experiment called the Cartesian devil, depends upon principles nearly the same as the above. The figure of a man, made of glass, or enamel, is so constructed as to have the same specific gravity as water, and will therefore remain suspended in a mass of that fluid. A bubble, similar to that in the last experiment, communicating with the water, is placed in some part of the figure, sometimes in a small globe, as shown at m, fig. 46. At the Dottom B of the vessel is a diaphragm of bladder which can be pressed upwards by applying the finger to the extremity e of lever e o, moving about a centre o. The pressure applied to e is communicated through the water to the bubble of air which is thus compressed. The specific gravity of the figure is thereby increased, and it consequently sinks to the bottom; but by removing the pressure, the figure again rises, so that it may be made to oscilate or dance in the vessel without any visible cause. Fishes made of glass are sometimes substituted for the human shape, and when a common jar is used for the experiment, the pressure is applied to the upper surface of it at A, which is, in this case, a piece of bladder, instead of being placed at the bottom, as shown in the figure. The same effects

will be produced by placing the jar under the receiver of an air-pump, and varying the pressure by rarefaction; but, in this case, the specific gravity of the figure ought a little to exceed that of the water in the jar.

Bodies cannot move about in the atmosphere without displacing it, which requires force, and therefore the resistance of the air always diminishes the velocity of moving bodies; but this diminution will be greater or less according to the density of the falling or moving mass; for which reason those bodies which we call light, fall much slower than more dense or heavy substances. To show that this is actually the case, we need only let fall a feather and a piece of metal out of our hands at the same instant, and it will be found that the metal has reached the floor while the feather is yet falling; but place them both under the receiver of an air-pump and exhaust the air, and then let them fall, and they will both be found to reach the bottom in the same time: an apparatus for performing this experiment is shown in fig. 49; the two bodies being laid together on the brass flap d or e, which may be at any time let down, by simply turning the wire f, which passes through a collar of leathers g, placed in the head of the receiver A B. Figure 50 represents another apparatus for showing the same thing. It consists of two sets of brass vanes put on separate axles, in the manner of windmill-sails. One set has the edges placed in the direction of their whirling motion, that is, in a plane to which the axis is perpendicular. The planes of the other set pass through the axis and they are, therefore, trimmed so as exactly to front the air through which they move. Two springs act upon pins projecting from the axis, and their strengths or tensions are so adjusted, that when they are disengaged in a vacuum, the two sets continue in motion equally long; but if they are disengaged in air, those vanes which oppose the air by their planes will stop long before those that cut it edgewise.

In the preceding account of the air-pump we have seen that the most modern of these machines may be made use of either for the purpose of rarefaction or condensation; but in many cases the latter operation is required to be performed when no pump is at hand: and moreover, it does not always require the same delicacy of process as that of rarefaction, and more simple apparatus are, therefore, frequently employed.

Take, for example, a prismatic tube AB, fig. 51, shut at one end, and fit it with a piston, or plug C, so that no air can pass by its sides, which, in a cylindrical tube, is best done with a turned stopper, covered with oiled leather, and fitted with a long handle CD. When this is thrust down, the air, which formerly occupied the whole capacity of the tube, is condensed into less room, and the force necessary to produce any degree of condensation, may be concluded from the weight necessary for pushing down the plug to any depth; but it is obvious that the instrument in this form is not sufficiently accurate for us to deduce, from experiments upon it, any very accurate conclusions; the following is, therefore, more commonly the form of the condenser. The end of the tube, instead of