Page images
[blocks in formation]

filled, as soon as this is done, the atmospheric pressure on the surface of the liquid in D forces the liquid up the tube towards the highest point B, and if this point be not at a greater height than about 32 feet if water be employed in D, and not more than 30 inches if the vessel contain mercury, the fluid will pass beyond the highest point B and fill the whole of the tube to c. The vessel E can then be placed under the open end c, and the whole of the liquid in D situated above the open end a will be transferred into E.

The reader will understand why one limb of the siphon must be longer than the other by considering that, when the instrument is left to itself, the atmospheric pressure is acting as much at one extremity of the siphon as at the other. If, when the liquid column is raised to B, the mouth be withdrawn from c, the column will fall back into the vessel D. It will do the same if we get the liquid no further than F, which is the level of the liquid in D, because at that point the upward pressure of the atmosphere prevails over the downward pressure of the liquid; but beyond that point, in the direction F C, the downward pressure of the liquid prevails over the upward pressure of the atmosphere, and the liquid will flow out. Thus, the motion of the fluid is, as Mr. Webster remarks, similar to the motion of a chain hanging over a pulley. If the two parts of the chain be equal, the fluid remains at rest; and, if one end be longer than the other, it moves in the direction of the longer end. Fresh links, so to speak, are added continuously to the fluid chain by the atmospheric pressure on the surface of the fluid, so that the chain being continuous the motion is continuous also, and does not cease till one portion of the chain becomes equal to or less than the other.

A stream of water descending through the air tapers

downwards, and at a certain depth divides into drops, because each particle falls with accelerated velocity, and at length (when it has overcome their cohesion) leaves the other particles behind it. But, when the stream is inclosed in a tube, this separation of its parts is prevented by the atmospheric pressure above and below keeping them together and forcing the whole stream to flow with equal velocity; the lower part dragging the upper after it, while the upper (by its inertia) equally retards the lower, so that they move together with the mean of their natural velocities; and the discharge is, of course, more rapid than if there were no tube, and will be faster the longer the tube. Now, as the same is true of a stream of light fluid ascending through a heavier, this explains why the draught of a furnace depends on the height of the chimney.

16. As the barometer is by far the most important instrument connected with Pneumatics, it is necessary to enter into the construction of this instrument and the information to be gained from it at greater length than we have hitherto done in this or in the former introductory treatise.

The essential part of a barometer is a well-formed glass tube, 33 or 34 inches long, of equal bore, containing pure mercury only, and so arranged that this mercury may be supported by atmospheric pressure; all the other appendages being contrivances for protecting the tube and ascertaining the exact height of the mercurial column.

At first sight nothing appears more easy than to fill a tube with mercury and invert its open end into a cup of the same metal. There are, however, certain practical difficulties which render the construction of a good barometer a work of great nicety. In the first

[blocks in formation]

place mercury is very liable to contamination, from the facility with which it dissolves baser and cheaper metals, such as tin, lead, zinc, and bismuth; and, as the specific gravities of these metals are all much less than that of mercury, any admixture of them will cause the height of the barometer, as indicated by a column of the adulterated metal, to be greater than in a barometer containing pure metal. There are various methods of purifying mercury, for which we must refer to chemical works*; but, in addition to these sources of impurity, it was the opinion of Sir Humphry Davy, that mercury in its pure state, when exposed to the air, absorbs both air and moisture. If such really were the case, the construction of an accurate barometer would be impossible; but, according to the experiments of the late Professor Daniell, mercury is incapable of absorbing or retaining either air or moisture; and the air-bubbles seen to rise from it when heated or when relieved of atmospheric pressure are merely retained between the mercury and the glass vessel by their attraction for the latter.

On pouring mercury into the barometer-tube and inverting it, the air thus confined between the mercury and the inner surface of the tube will be relieved from atmospheric pressure and escape into the Torricellian vacuum, where it will oppose the pressure of the external air, and constantly maintain the mercurial column at a lower level than if the Torricellian vacuum were perfect; so that the observed height of the column would not indicate the true pressure of the atmosphere, but only the excess of the pressure above that within the tube. Now, in order to get rid of this air adhering to the tube, as well as any moisture which is sure to settle upon any thing left exposed to the air, it is * See Faraday's Chemical Manipulation.

necessary to introduce small portions of mercury into the tube, and boil them by the action of a small charcoal fire. This, of course, requires considerable care, not only to avoid the fracture of the tube, but also to prevent fumes of mercury from escaping, for these are very pernicious if inhaled. The tube is first gently warmed, so as to dry it thoroughly. A quantity of pure mercury is then poured in, so as to occupy two or three inches of the sealed end of the tube, which is held over the fire until the mercury boils, taking care to turn the tube round upon its axis, so that the heat may be equally applied. After boiling for a minute or two, the open end is closed by a cork to prevent the introduction of moist air, and the tube is then allowed to cool, in order that the cold mercury which is next to be poured in may not crack the tube. When a second portion of mercury, which is about equal to the first, has been poured in, the part of the tube containing this new portion is held over the fire until it boils. It is again removed from the fire, and corked up as before. A third portion of mercury is then introduced, and the heat again applied to that part of the tube containing the last addition of metal; and in this way the tube is at length filled, with the exception of a small portion from which the heat has expelled the mercury. This is filled up with mercury, and the finger is now placed over the open end so as carefully to exclude any air: the tube is then reversed into a cup of pure mercury; as the column sinks, it expels the last portion of mercury which had not been boiled; and, as there is neither air nor aqueous vapour above the mercurial column, its length exactly measures the atmospheric pressure whenever the temperature is too low for the mercury to give out any sensible quantity of vapour. It appears, however,



that mercury exists in the aeriform state whenever its temperature is above 60°, so that it must, in this state (though exceedingly rare), fill the so-called vacuum at the top of the tube, and, by its elasticity, depress the liquid and partly counter-balance the external pressure. But this source of error, which is unavoidable is fortunately extremely small.

With all the above precautions, however, the barometer is liable to continual deterioration from another cause. Capillary attraction, or that force by which solids attract fluids so as to be wetted by them, does not act equally between all solids and fluids. Most solids, with the exception of certain metals, display less attraction of this kind for mercury than for any other fluid, so that they are never wetted by it, and, when immersed in it, they always retain a film of the last fluid that touched them. Hence the precautions necessary for removing the film of air from the inner surface of the barometer-tube so as to bring the mercury into contact with the glass; but as this contact cannot be insured on the outside of the tube, where it is immersed in the cup, a film of air is always retained at this part of the tube, and also at its under edges, which film creeps by small portions at a time into the interior, and rises up in innumerable bubbles into the vacuum, the film being constantly renewed by the descent of more air between the outside of the tube and the mercury in the cup, and in this way the outer air slowly insinuates itself into the barometer. The only means of preventing this deterioration of the barometer is, by making the bottom of the tube of some substance which attracts mercury in preference to air, so as to be wetted by the mercury. Now, as the metal platinum possesses the rare property of welding with glass, that is, of uniting with glass when softened

« PreviousContinue »