BAROMETER 329 necessarily very slow, aud several minutes may elapse before a sudden change of atmospheric pressure is indicated. The cause of the shift- ing pressure of the atmosphere is to be looked for in the operations of the winds which may be blowing in distant localities. By drawing the air away from any point, the pressure is here to some extent taken off, producing a partial vacuum which must soon be filled by a rush of air from other sources. Where the winds are equable, like the trade winds of the tropics, the movements of the barometer par- take of the same regularity. Humboldt, in his researches in the equatorial regions of South America, was greatly struck by the uniformity of the motion of the barometer in the different periods of the day. From 4 o'clock in the morning till 10 the mercury generally rises, and then falls until 4 in the afternoon. It then rises again till 10 at night, after which it falls till 4 in the morning. In temperate northern latitudes the barometer generally stands higher at 9 A. M. and 9 P. M. and lower at 3 A. M. and 3 P. M. than at other hours. Prof. Daniell recommends these hours as the best times for consulting the barometer as a weather glass. Its rise between 9 A. M. and 3 P. M. indicates fine weather. A fall from this time to 9 P. M. is likely to be followed by rain. Prof. Buys-Hal- lot of Utrecht occupied himself for many years in making with others simultaneous observa- tions in different localities of the changes in the barometer and in wind and weather. He de- termined positive numerical relations between the force of the wind and the height of the ba- rometer preceding it. He succeeded at last in finding the laws governing the forward motion of the centre of barometric depression, followed by storms, and induced the government of Hol- land to establish a weather bureau with public storm signals in 1860, which was followed by England in 1861, by France in 1863, and by the United States in 1870. These laws, as might be expected, differ in different localities. From this relation rules have been deduced by which the maximum force of the wind during the day may be predicted every morning, thus enabling outward-bound vessels to determine the safety of putting to sea. The BOILING POINT BAROM- ETER is an instrument whose action depends upon the variable temperature at which water boils at different elevations, or, what is the same thing, under different atmospheric pres- sures. It is constructed with a small cistern for the water, arranged in a cylindrical tin tube, which contains in the lower part an alcohol lamp for heating the fluid. The temperature is best noticed by suspending the bulb of the thermometer in the partially confined steam which rises from the boiling water. The dif- ference in the temperature observed at two different points, expressed in degrees of Fahren- heit's thermometer, being multiplied by 530, will give the approximate difference of eleva- tion between these two points. For greater accuracy correction should be made for the difference of the temperature of the air at the two places. Although the instrument is in a very portable and convenient form, it has not proved a favorite with scientific observers, from a want of confidence in its results. The ANE- ROID BAROMETER (Gr. a, vjip6f, and ddof, a form JPjy. 9. without fluid) is a modification of the vacuum case barometer, the earliest form of which was invented by M. Cont6, professor in the aeros- tatical school at Meudon, near Paris, and de- scribed by him in the Bulletin des sciences, Flo- real, year 6 (1798), p. 106. M. Cont6 in his balloon ascents found the reading of the mer- curial barometer subject to the same difficul- ties so much complained of on shipboard, aris- ing from the violent oscillations of the instru- ment. He therefore invented a watch-like, metallic, air-tight vacuum case, the lid of which, sustained by internal springs, rose and fell under the variable pressure of the atmosphere, an index showing the motion. M. Vidi sub- sequently devised a case of different form, with a flat corrugated top and bottom, flanged over and soldered to a rim, first pressed together at the centre by the withdrawal of the enclosed air, and then separated a certain distance by the introduction of a compensating spring. The instrument thus improved and constructed has come into extensive use. It is represented externally by fig. 9; fig. 10 shows the interior arrangement, while fig. 11 shows a cross sec- tion of the flexible air-tight box, which col- lapses when the air is withdrawn. (See fig. 12.) By means of a spring it is brought back
