585

BAR

B A R

meuts of the domestic and useful arts, will be universally admitted.When Galileo , in the beginning of the seventeenth century, dis-covered that water could not ascend in a pump, unless the suckerreached within 33 feet of its surface in the well, he justly concludedthat the ascent of water in pumps, was effected by the pressure ofthe atmosphere, and not by the power ol suction; that a columnof water 33 leel high was a counterpoise to one of air of an equaldiameter and base, the height of w hich extended to the top of theatmosphere; and that consequently the water could not be attractedany larther by the sucker. This important discovery inducedhis great pupil Torricelli , to substitute a column of mercury forthat of water; because the former flu u being abouL 14 times heavierthan the latter, he wanted, according to th.it proportion, only about21) j inches of quicksilver to determine the accuracy of his experi-ment. He accordingly found Ural, alter having tilled a glass-tubewith mercury, and inverted it in a bason of the same fluid, it de-scended in the tube till it became stationary at about 29 J inchesabove the surface of that contained in tiie lower vessel. Manyyears, however, elapsed after this experiment, before any noticewas taken of the circumstance, that tiiis pressure of the air con-siderably varied at different limes, though the tube was uniformlykept in the same situation. Indeed, these variations in the mer-curial cc lumn, were too obvious to remain long unnoticed; andphilosophers began minutely to mark their degrees. As soon,therefore, as this point was properly attended to, they observedthat the changes in the vise and tall of the mercury were in genera!very speedily succeeded by variations in tile weather. Hence theinstrument obtained the name of weather-glass, lor which purposeit has, since that period, been generally employed.

Barometer, Common, invented by Torricelli , Plate XIX.Fig. 8. AB is a glass tube, of or or | inch wide, (the widertlie better,) and about 34 inches long, being close at the top A,and the open end 15 immersed in a bason of quicksilver CD,which is the better the wider it is. To till this, or any other baro-meter,—take a clean new glass-tube, of the dimensions as above,and pour into it well purified quicksilver, with a small funneleither of glass or paper, in a line continued stream, till it wan sabout half an inch or an inch of being full; then stopping it closewith the ringer, invert it slowly, and the air in the empty 1 part willascend gradually to the. other end, collecting into itself such othersmall air-bubbles as unavoidably get into the tube among themercury. In rilling it continue to invert it several times, turningthe two ends alternately upwards, till all the air-bubbles are col-lected, and brought up to the open end of the tube, and till thepart filled shall appear, without speck, like a fine polished steel-rod. This done, pour in a little more quicksilver, to fill theempty 7 part quite full, and so exclude all air from the tube ; thenstopping the orifice again with the ringer, invert the tube, andimmerse the finger and end, thus stopped, into a bason of purifiedquicksilver. In this position withdraw the finger, so shall themercury descend in the tube to some place as G, between 23 and31 inches above that in the bason at F, as these are the limits be-tween which it always stands in this country, on the common sur-face of the earth. Then measure, from the surface of the quick-silver in the bason at F, 28 inches to K, and 31 inches to I, dividingtire space between them into inches and tenths, which are markedon a scale placed against the side of the tube ; and the tenths aresubdivided into hundredth parts of an inch by a sliding index car-rying a vernier or nonius. The height of the mercury in the lubeabove the surface of the mercury in the bason is called the standardaltitude, and the difference between the greatest and least altitudesis called the limit or scale of variation. If each inch of the scaleof variation, AD, Fig. 10, of a barorncter-tube be divided into tenequal parts, marked with 1, 2, 3, &c. increasing upwards, and avernier or nonius, LM, whose length is -{.Jllis ot an inch, be di-vided into ten equal parts, marked with 1, 2, 3, &c. increasingdownwards, and so placed as to slide along the graduated scale ofthe barometer, (lie altitude of the mercury in the tube above thesurface of that in the bason may be found in inches and hundredthparts of an inch by this process. If the surface of the mercury inthe tube do not coincide with a division in the scale of variation,place the index of the vernier, M, even with the surface, and ob-serving where a division of tiie vernier coincides with one in tiiescale, the figure in the vernier will shew what hundredth parts of aninch are to be added to the tenths immediately below the index.Let, for instance, the surface of the mercury be between 7 and 8

tenths above 30 incb.es, ai d the index of the vernier being placedeven with it, and the tiguie 3 upon the vernier being observed tocoincide with a division upon the scale, the altitude of the baro-meter will be 30 inches -fa and -rj^ths of an inch: or 30,75 inches.The 3 inches, between 38 ami 3 l, so divided, will answer for allthe ordinary purposes of a statemarv or chamber barometer; butfor experiments on altitudes and depths, it is proper to have thedivisions carried on a little higher up, and a great deal lower. Inthe proper filling and otherwise fitting up of the barometer, severalcircumstance, are to be carefully uo'ted: as, that the bore of thetube be pretty wide, to allow the freer motion of the quicksilver,without being impeded by an adhesion to the sides ; that the basonbelow it be also pretty large, in order that the surface of the mer-cury at F, may not sensibly rise or fall with that in the tube ; thattiie bottom oi the tube be cut off rather obliquely, that when itrests on the bottom of the bason there may be a free passage forthe quicksilver; and that, to have the quicksilver very pure, it isbest to boil it in the tube, which will expel all the air from it. Hav-ing carefully filled the tube, hold it with the sealed end lowest, in aninclined position, over a chafing dish of burning charcoal, placednear the edge of a table, in order that ail parts of the tube may beexposed successively to'the action of the fire, by moving it ob-liquely over the c'ballng-dish. The sealed .end is to be firstgradually presented to the lire. As soon as the mercury becomeshot, the internal surface of the tube will be studded with an in-finite number ot air-bubbles, giving the mercury a kind of greycolour: these increase in size by running into one another, andascend towards the higher parts of the tube, where, meeting witha cooler part of the lluid, they are condensed, and nearly disappear.In consequence, however, of successive emigrations towards theupper parts of the tube, which are successively heated, they finallyacquire a bulk, which enables them in their united form entirelyto escape. When the first part of the tube is sufficiently boiled,move it onward, by little and little, through its whole length.W’lien the Mercury boils, its parts strike against each other, andagainst the sides of the tube, with such violence that a person un-acquainted with the operation, naturally apprehends the destruc-tion ot his tube. By tiiis process the mercury is entirely deprivedof the air which adhered to it.

The barometer is commonly fitted up in a neat mahogany case,,together with a thermometer and hygrometer, as represented inPlate XIX. Fig. 9. As the scale of variation is small, being in-cluded within three inches in the common barometer, several con-trivances have been devised to enlarge the scale, or to render themotion of the quicksilver more sensible, among which is the in-vention of the diagonal barometer, by Sir Samuel Moreland, fig.11. which enlarge the natural scale of three inches perpendicular,CD, by extending it to any length BC in an oblique direction.Tiiis is liable in some degree to the inconvenience of friction andbreaking; and hence it is found that the diagonal part BC cannotproperly be bent from the perpendicular more than in an angle of45°, which only increases the scale nearly in the proportion of sevento five.

Barometer, Wheel , fig. 12. This was invented by Dr. Ilookabout 1668, and is meant to render the alterations in the air moresensible. Here the baroineter-tube has a large ball AB at top,and is bent up at the lower or open end, where an iron ball Gfloats on the top of the mercury in the tube, to which is connectedanother ball II by a cord, hanging freely over a pulley, turning anindex KL about its centre. When the mercury rises in the partFG, it raises the bull, and the other ball descends and turns thepulley with the index round a graduated circle from X towardsM and P; and the contrary way when the quicksilver and the ballsink in the bent part of the tube. Hence the scale is easily en-larged 10 or 12 fold, being increased in proportion of the axis ofthe pulley to the length of the index KL. But then the frictionof the. pulley and axis is some obstruction to the free motion of thequicksilver, Contrivances to lessen the friction, &c. mav also beteen in the Phil. 'Frans, vol. 52. art. 29. and vol. 60. art 10.

Barometer, Horizontal, or 'Rectangular , was invented byJ. Bernouilli and Cassini; and is represented in Fig. 13. AD is a.pretty wide cylindrical part at the top of the tube, which is bent atright angles at F, the lower part of it FG being turned into tiiehorizontal direction, and dose at A, but open at the lower end C,where however the mercury cannot run out, being'there opposedby the pressure cf the atmosphere. But on account of the attrition