BAROMETER (Greek, (weight-measure)), an instrument invented by the Italian physicist Torricelli, and used for determining the sure of the atmosphere. (For an account of its early history see ATMOSPHERE). ID ItS plest form the mercurial barometer consists tially of a vertical glass tube about a yard in length, closed. at the top and open at the bottom, and partially filled with memury, into a vessel of which its lower end also dips. In preparing the strument for use, the tube is first completely filled with mercury; but as soon as it is free to do so the column of mercury in the tube sinks (leaving a vacuous space at the top of the tube) until it stands at a height (usually about 30 inches) such that the pressure of the umn exactly balances that of the atmosphere. A graduated scale of metal or glass is .provided, by means of which the ence in level between the top of the column and the surface of the mercury in the open vessel (called the (cistern() at the bottom can be measured with precision. In the Fortin instrument (the design commonly adopted for all but the most refined work) the cistern is closed below by a piece of flexible leather, which can be raised or lowered by means of a screw, in order to bring the face of the mercury in the cistern to a certain fuced level, before the reading is taken. A pointed index, k, preferably of ivory, projects downward into the cistern from the upper cover, the position of its tip, with respect to the scale on the barometer tube above, being known. The mercury in the cistern being first brought accurately into contact with the tremity of k, the position of the upper end of the barometric column is read from the scale. The (apparent)) height of the barometer is then known; but in order to deduce the (true( height, certain corrections must be applied. The most important of these is the correction for temperature. The scale from which the height of the column is read is longer when the perature is high than when the temperature is low; and the mercury in the column is also less dense at higher temperatures than at lower ones. These two sources of error partially sate each other; for at a high tempemture the reduced density of the rnercury tends to make the column stand too high, while the greater length of the scale at such a temperature tends to make the reading too small. The com pensation is not perfect, however, and when the coefficient of expansion of the scale is lmown, a table of temperature corrections must be cal culated, to reduce the direct reading to what it would have been if it had been taken at some fixed standard temperature. The temperature of melting ice is adopted, by universal consent, as the standard to which ate (apparent° read ing is to be reduced. Another important cor rection must be applied in order to allow for the variations of gravity with the latitude and elevation of the place of observation. Where gravity is relatively weak, a longer column of mercury will be required to balance a given at mospheric pressure than would be required to balance the same pressure in a region where gravity is stronger. All the barometric read
ings taken at the International Bureau of Weights and Measures, near Paris, are reduced to the values they would have if made at the level of the sea, in latitude 45° ; and this prac tice is growing in favor among physicists gen erally. To reduce a barometric reading to sea level and to latitude 45°, it is merely necessary to multiply the observed height of the column (after applying the correction for temperature) •• the expression (1—.00259 cos L) (1 — .0101111. H), where L is the latitude of the place of observation, and H is its height above the sea, in feet. Several secondary corrections have also to be considered, when great refine ment is desired. Prominent among these is the correction for (capillarity,( which is made necessary by the fact that the mercury does not stand as high in a small barometric tube as it does in a larger one, on account of the surface tension (q.v.) of the liquid. No simple formula for this correction can be given, and it varies somewhat according as the barometer is rising or falling at the time of the observation. Tables for finding the capillary correction are given in Guyot's meteorological and physical tables, published under the direction of the Smith sonian Institution at Washington. An excellent table is also given in Guillaume's (Thermome trie de Precision,' where the elaborate precau tions taken in filling the modern precision barometer are also described.
The barometer is a simple instrument, and of the greatest use in all kinds of scientific work. The greatest fault of the mercurial in strument is the difficulty of transporting it with out brealcage and without destroying the vacuum in the upper part of the tube by the admission of air bubbles. Instruments like the Fortin type may be transported by screwing up the leather bottom until both the cistern and the tube are completely filled with mercury, then revising the barometer, and carrying it to its destination bottom side up. The aaneroicP barometer, although not nearly so accurate as the mercurial instrument, possesses the advan tage of portability, since, as its name signifies, it does not contain any liquid; and it is there fore used to a considerable extent in the deter mination of the heights of places above the sea. (See HYPSOMETRY). Various forms of the aneroid exist. One of these consists of a cylindrical metal box, exhausted of air, and having a lid of thin, corrugated metal. The lid, Ihhich is highly elastic, yields to every change of atmospheric pressure, and delicate multiply ing levers transmit its motions to an index that moves over a graduated scale, whose divisions are marked on the dial empirically, by compari son with a mercurial barometer. For further information concerning the barometer and its use, consult Stewart and Gee, 'Elementary Prac tical Physics) ; Glazebrook and Shaw, 'Pkactical Physics) ; Abbe, 'Meteorological Apparatus and Methods); Guillaume, 'Thermometric de Pre cision.) See also METEOROLOGY.