Before being introduced into the tube, the mercu ry ought to be well heated, or eveu boiled in a gla zed earthen pipkin, in order to drive off any mois ture which may adhere to it ; but this will be unne cessary, if the mercury has been recently revived.
The mercury ought likewise to be boiled in the tube, to expel any air or moisture which may still remain attached to it, or to the inside of the tube. This is done in the following manner: Pour as much mercury into the tube as will make it stand to the height of three or four inelya; and introduce a long wire of iron to stir it during al act of boiling. Ex pose the mercury in the tube gradually to the heat of a chafing dish of burning charcoal ; and when it begins to boil, stir it gently with the iron wire to facilitate the disengagement of the bubbles of the air. When the first portion of the mercury has been sufficiently boiled, and all the air extricated, remove the tube from the chafing dish, and allow the whole to cool, taking care not to bring it into contact with any cold substance. Introduce an equal quantity of mercury, and treat it in the same manner, withdraw ing the wire a little, so that it may not reach below the upper part of the mercury already freed from air. The chafing dish must also be placed immediately under the mercury which has been last poured in. Repeat the same process with each successive por tion of mercury till the tube is filled, always apply ing the heat very cautiously ; and be equally careful in allowing it to cool, before a fresh portion of mer cury is poured in.
It sometimes happens, when the tube is carefully inverted, as in the Torricellian experiment, that the mercury, after being completely freed from air, in the way we have described, remains suspended in the upper part of the tube, and does not assume its pro per level, with respect to the pressure of the atmo sphere, till the tube has been gently shaken. This fact, which seems to have given great difficulty to Huygens, is certainly owing to the capillary attrac tion of the tube, and the mutual attraction of the particles of mercury, as it takes place only in tubes of a small bore. To say, that it is owing to the in fluence of an invisible ethereal fluid more subtile than air, is no less unphilosophical than the semicircular hypothesis of Linus, or the principle of the horror of a vacuum, particularly when we can assign a cause for it, of whose operation we have many simple and obvious proofs.
The Torricellian experiment exhibits the barome ter under the simplest, though not the most conve nient form. The tube AB, ( Fig. 2. Plate LII I.) which
• is hermetically sealed at A, ought to be about 33 ' inches in length, and 94 or 3 lines in diameter of ' bore. Since height cf the mercury in the tube must be reckoned from the surface of the mercury in the bason, the width of the latter ought to be such, that the elevation or depression of the mercury in the tube may have little effect in changing the level of the mercury in the bason. In the more im proved barometers, contrivances, which we shall af terwards explain, have been adopted for preserving the lower surface of the mercury at the some level ; but this is obtained sufficiently well for common pur poses, by giving the bason a considerable diameter with respect to that of the tube.
The common barometer, represented in Fig. 3, • differs but little from the Torricellian tube. Instead of a bason, a small reservoir is usually attached to the lower extremity of the tube, or rather the tube itself is swelled out into a bulby form, as represented at Fig. 4. In this form, however, the instrument is by no means accurate, particularly when the diame ter of the bulb, as is usually the case, differs but lit tle from the diameter of the bore of the tube. In order to keep the surface of the mercury in the ba son always at the same level, the father of the late Mr George Adams constructed the bason wholly, or in part, of leather, and by means of a screw at the bottom, adjusted the surface of the mercury in it, so as to have it always at the place from which the di-, visions on the scale commence.
In this country, the lowest station of the mercury is observed to be about 28 inches, and its highest 31 inches above the level of the mercury in the bason : and when the instrument is to remain in a fixed posi tion near the surface of the earth, we may consider the interval between these two points, as the range of the barometrical scale. The scale which, conse quently, will embrace three inches, may be subdivided into smaller divisions, according to the degree of nicety required. Each inch is commonly divided in to ten equal parts; and these are subdivided into hundredths of an inch, by a contrivance called a ver nier scale. (See VERNIER.) By this means the height of the mercury is ascertained, by inspection, to the nth of an inch. For nice purposes, the vernier may be made to indicate the , of an inch.