We need only cite here, as a proof of the small magnitude of these inconveni ences, that which happens to divers, when they have been shut up within a bell which descended vertically in the water, and in which the air, pressed by the weight of the surrounding columns, contracts itself more and more, in pro portion as the vessel is found at a greater depth. The accidents which have oc curred to those who have continued for a certain time under the bell, have arisen in great part from the alteration produced in the air by respiration, and that which was most dangerous in this fluid was the defect of renewing it. See Divixa bell, BAROMETER, &c.
The elasticity of the air is verified by several well-known experiments. One of the most ordinary is that in which we employ the machine called the artificial fountain. It consists of a metallic vessel of a rounded form, its summit being pierced with an orifice, through which the vessel maybe filled with water to about two-thirds of its capacity. In this aperture a tube is then fixed, which descends into the vessel until it is within a little distance of the bottom, while its upper part, which projects from the orifice, is fur nished with a cock. To this same part a forcing pump is adapted, and the cock being opened, a great quantity of air is injected into the vessel : this air, be ing lighter than water, rises above it, and its elasticity augments with its den sity, in proportion as new strokes are riven to the piston. Then, after closing the cock, the pump is removed, and a kind of little hollow cone is substituted for it, open at its summit, which is turn ed upwards ; as soon as the cock is again opened, the condensed air exerts its force upon the surface of the wa ter, and drives it through the canal that is immersed into that liquid, whence it is seen to shoot out, under the form of a jet of more than twenty or thirty feet in height. An analogous effect may be obtained, solely by diminishing the natural elasticity of the air, by plac ing under the receiver of an air pump a little vessel, in which all is similar to what the artificial fountain presents at the moment when the cock is opened to give a free passage to the water, except that the air situated above tht liquid is in its ordinary state.
While the exhaustion is going on, the air included in the vessel, and whose pressure upon the water is no longer balanced by that of the exterior air, dilates itself, and gives birth to a jet which rises under the receiver: (See fig. 5.) But the most interesting ment relative to this object is that of Boyle, and of Mariotte, to show that the air contracts itself nearly in the ratio of the weights with which it is pressed. These kinds of experiments merit the prefer ence, since they are not confined to merely proving the existence of a phe nomenon, but make known also how it exists, by determining the law to which it is subject. Take a glass tube a b (Plate
Pneumatics, fig. 1), bent into two branches, the shortest of which is about twelve inches high ; it must be equally thick throughout, and hermetically seal ed at its extremity b. The other branch, which is open at a, should be at least five feet, but if it were eight feet in height, so much the better. The whole is fixed upon a plate which carries di visions adapted to the two tubes. First, let there be poured into the bent part a little mercury, to obtain a line of level, x 2., that we may estimate the number of degrees comprised between that line and the superior extremity of the shortest branch. In this state of things the air which occupies that branch maintains an equilibrium, by its elasticity, with the pressure of the column of atmo speric air gravitating in the other branch, and whose pressure is transmitted by means of the mercury comprised in the inferior curvature. This pressure, as we have seen in the article IhnomErmi, is equal to that of a mercurial column of about twenty-nine or thirty inches in height. Afterwards, let mercury be poured into the longest branch, and at the same time the air in the other branch will be condensed ; by the ex cess of the resulting pressure the mer cury will rise in the shorter branch un til an equilibrium is again produced. Then measure, on one part, the length of that column of compressed air, and on the other the excess of the column of mercury contained in the longest branch, above that which occupies the shortest, We - will suppose, for more simplicity, that this excess is equal to thirty inches ; in that case, we shall find that the column of compressed air is reduced to. the half of the height which it occupi- ed previously to the introduction of the fresh mercury. But that column is charged with a weight double of the former, since a pressure of thirty inches of mercury is added to an equal pressure exerted by the atmospheric air, and which is not con sidered as being diminished ; for we may neglect the small difference which results from this, that the thirty inches which terminate the atmospheric column at bot tom are actually occupied by the mercu ry. In general, if we take the ratio be tween the first pressure from the column of the atmosphere, and any other pres sure whatever exerted by that same co lumn, and by the mercury superadded, the corresponding spaces, occupied by the compressed air, will be respectively in the inverse ratio of the pressures ; whence it is obvious, that the air contracts itself, as we have stated, in proportion to the weights compressing it. If we after wards take out the mercury at several distinct times, the air will expand by rea son of its elasticity, and the spaces which it will successively occupy in a contra ry order will still conform to the inverse ratio of the pressures.