Page images

by heat, it occurred to Professor Daniell to unite a ring of platinum with the open end of the barometertube, so as to bring it into contact with the mercury, thus effectually preventing the ingress of air into the tube. The same distinguished philosopher also invented a new mode of filling barometer-tubes, which, with far less difficulty and danger, insures as much accuracy as by the old method. The improvement consists in pouring the mercury into the tube while both are under the exhausted receiver of a good air-pump. Care is required to prevent any splashing of the mercury as it descends into the tube, for this causes bubbles of highly rarefied air to become entangled between the mercury and the glass; this, however, is prevented by pouring the mercury through a long slender funnel extending to the bottom of the tube, and dipping into a small portion of mercury previously introduced, and boiled. By this means all agitation is confined to the tube of the funnel, which, on being removed, and filling up the barometer-tube, the only part containing bubbles is that last filled; and, as these bubbles are formed by the highly rarefied air of the exhausted space, they shrink into invisible points on exposure to the common pressure, and on inverting the tube the last portion containing these bubbles is expelled, and the tube left perfectly free of air.

In making the standard barometer for the Royal Society, Professor Daniell not only followed this process, but afterwards boiled the mercury still in vacuo; and he noticed, as a striking proof of the absence of air and the perfect contact of the mercury with the glass, that, although the bore of the tube was more than half an inch, yet, on inverting it, the fluid did not at once fall to its usual height, but remained suspended to the top of the tube as water would have



done in the same tube, until detached therefrom by a few concussions. Yet this fine instrument, having no platinum guard, as just described, was found after two years to grow dim (like an old looking-glass), from the insinuation of air-bubbles between the glass and the mercury, so that the tube had to be refilled as before; but, in doing so, a ring of platinum was added to the open end, which has since preserved the instrument from deterioration.

17. As the excellence of a barometer chiefly depends on the absence of all matter, except mercury from the tube, we may test its value by three indications. First, by the brightness of the mercurial column, and the absence of any flaw, speck, or dulness of surface; secondly, by the barometric light, as it is called, or flashes of electric light in the Torricellian vacuum, produced by the friction of the mercury against the glass, when the column is made to oscillate through an inch or two in the dark; thirdly, by a peculiar clicking sound, produced when the mercury is made to strike the top of the tube.

18. It has been shown, that the mercury in the tube of the barometer above the level of the mercury in the cup exactly balances the atmospheric pressure on the latter. The mean annual height of the column varies in this climate from 30.065 to 29.785 inches, and a scale is provided to measure this variation. Now, as this scale is always attached to the solid parts of the instrument, it can only measure the variations in one of the levels, namely, at the top of the column a, which is obviously less than the variations in the difference of level ab, Fig. 11, because every fall of the mercury in the tube produces a small rise of that in the cup, and vice versa. It must, therefore, be remembered, that the variations in atmospheric pressure,

or in the height of the mercurial column, are made up of both the variations of level at a and at b, and are, therefore, rather greater than those of a alone, as measured by the scale. Hence the divisions of the scale should not be exact inches, but diminished, as is usually done by the instrument-makers.

Fig. 11.

19. Various contrivances have been made for increasing the length of the scale, or for making it more convenient for use. The most popular form is the common wheel-barometer, or weatherglass, as it is called. In this instrument, the tube, instead of terminating at the bottom in a cistern, is recurved so as to form an inverted siphon, as in Fig. 12. As a rise of the mercury in the longer or closed limb is equivalent to a fall in the shorter limb, and vice versâ, a float is placed on the surface of the mercury in the shorter limb, and is connected with a string passing over a pulley, and very nearly balanced by another weight on the other side of the pulley. An index hand attached to the pulley moves over the surface of a dial-plate, graduated so as to indicate the oscillations of the mercurial column. With an increase of atmospheric pressure the mercury in the longer tube rises, and that in the short tube is depressed, together with the float, and this gives a small motion of revolution to the pulley, and also the attached index hand. A fall in the longer column causes the mercury with its float in the short limb to rise, and, consequently, moves the index hand in the contrary direction.

The siphon form of barometer, as commonly made,. is inferior to the cistern barometer, because a change


of pressure, such as would make a difference of nearly an inch in the upper level of the latter, would shew but half an inch in each level of the siphon; for, although the surfaces of the mercury in the longer and shorter limbs would be an inch farther apart, that inch would be compounded of a rise of half an inch at one surface, and a corresponding fall at the other. Our unit of measure, therefore, becomes only half as great, and necessarily diminishes in utility. In our figure, however, the upper end of the tube is expanded into a bulb, in order that, by enlarging the upper surface of the mercury, the difference of level may be made to depend almost entirely on the lower surface, giving


Fig. 12.


the same advantage as in a common barometer with a cistern of the same horizontal area as the bulb.

20. The desire to produce a more delicate measure of the atmospheric pressure has led to the construction at various times of a water-barometer, the lower surface of the water being protected from direct contact with the air by a layer of some more permanent liquid, such as oil, or some elastic solid, such as India rubber. Soon after the invention of the barometer, a waterbarometer was constructed by Otto Guericke (the inventor of the air-pump), and afterwards by Mariotte. It was supposed that the greater range of its oscillations would measure more minute changes of pressure. An instrument of this kind was constructed by Professor Daniell for the Royal Society. It consists of

one entire tube of glass, which was drawn out to the length of forty feet without much difficulty. Its diameter is about an inch, and the average height of the fluid column 400 inches. When originally put up in the year 1832, the water in the cistern was covered with a layer of castor oil; but, as that did not prevent the admission of the outer air, it was found necessary to refill the tube. This was done in January, 1845, and a solution of caoutchouc in naphtha was substituted for the castor oil. In windy weather this barometer appears to be constantly fluctuating, indicating numerous changes of pressure, which have no sensible effect on the most delicate mercurial barometer; the column appears to be in a state of perpetual motion, compared by Professor Daniell to the slow act of respiration. But the most important result is, that this instrument precedes by one hour the mercurial barometer of half an inch bore, as this does the mountain barometer of 0.15 inch bore, by the same interval in their horary oscillations; shewing that, while philosophers are disputing about the hours of the maxima and minima, much depends upon the construction of the instrument observed.

21. There are various other barometrical instruments of greater or less utility, which have been introduced at various times for measuring the absolute pressure or elasticity of any fluid in which they are placed. We must refer to larger treatises for an account of these instruments, but we may notice the principle upon which certain instruments, called differential barometers, are constructed. They consist essentially of a portion of liquid placed in the bend of a siphon tube shaped like the letter U, having its two ends open to the two fluids whose pressures are to be compared. The difference of these pressures causes a

« PreviousContinue »