Whatever be the number of divisions in the scale of variation, and in the ver nier, the height of the mercury in the ba rometer is easily discovered by a process similar to that already mentioned.
`There are several other kinds of baro meters, of which it will be sufficient to give a short description.
1. In the portable barometer, the lower part of the tube is bent upwards, and wider than the rest of the tube : and in this recurvated part the mercury is ex posed to the pressure of the atmosphere ; or the mercury in the basin is contained in a flexible leathern bag, exposed to the same pressure. In this last, the mercury is forced into the tube so as to fill it, by a screw fixed in the bottom of a wooden box containing the bag, lest the motion of the mercury should break the tube.
2. In the diagonal barometer, (fig. 11.) the scale of variation is bent into the di rection D R, making an obtuse angle with the vertical part B D. The scale of vari ation is by this barometer increased ill the ratio of DR:DA; but this increase does not compensate for the friction and attraction of cohesion upon the lower side of D R. And when the angle R D A is greater than the instrument is ren dered useless by the separation of glo bules of mercury from the column.
3. The wheel-barometer, (fig. 12.) is a compound tube, SERB D, open at D and closed at E, the diameter of the highest part, S E R, being much greater than that of the rest, and filled with mercury from D to S R, and above that vacuous. Upon the surface of the mercury, in the recurv ed leg, there is an iron ball in equilibrio with another, H, by a string passing over a pulley, P. As the ball at D rises and falls with the mercury, the string turns the pulley, and an index, I N, fixed to it, which points to different parts of a gradu ated circle. It is clear, that by increasing the diameter of the circle, this contrivance will shew the minutest variations of the air, provided the friction be inconsidera ble, which is seldom true.
4. The pendent barometer, (fig. 13) is composed of a tube of a very small bore, a little conical or tapering, closed at the smaller orifice, A, and filled with prepared mercury from A to B, whose distance is equal to the greatest altitude, or about 31 inches. Let the tube be suspended verti cally, and the mercury will subside, and be quiescent in that part whose length is equal to the standard altitude at that time: and supposing that to be the least, it will occupy a space F E equal to 28 inches ; and consequently A F is the scale of vari ation. A E = 60 inches, then A F = 32, when in the common barometer it is only 3 inches. The diameter of th is baro meter tube is very small, and consequent ly the attraction of cohesion considerable, which prevents the freedom of motion necessary to ascertain minute variations of the air's pressure.
5. In the horizontal rectangular baro meter, (fig. 14.) the highest part of the tube, opposite to the scale of variation, is wider than the rest of the tube ; and the mercury descending three inches,tfrom A to D, will describe a much longer space in the horizontal leg F G, these spaces being to each other inversely as the squares of the diameters of the tubes, and that F G being very small, its mo tion will be extremely sensible. But the free motion of the mercury in F G is im peded by friction, and the attraction of co hesion, which, from the smallness of the tube, is considerable ; and besides this globules of mercury are apt to be sepa rated from NI, and flow out at G.
By the above, and other expedients, as using water, or water and mercury, the scale of variation is enlar: ed ; but the common barometer is the best, being sub ject to the fewest inconveniences. In the construction and use of it, the following particulars are to be observed. 1. The diameter of the tube should be one-third or one-fourth of an inch, to prevent the effects of the attraction of cohesion ; the length of the tube 33 or 34 inches, with a bulb upon the top, into which the air may be diffused, should any remain in the mercury. 2. The diameter of the cistern, containing the mercury, should be large, (at least ten times greater than that of the tube) that the addition or subtraction of the mercury, contained between the greatest and least altitudes, may not sen sibly affect its depth ; for the numbers, marked upon the side of the tube, shew their distance from a fixed point, and can not shew the height of the column above the mercury in the cistern, unless its sur face coincide with this point, and be im moveable. 3. The mercury should be free from any mixture of other metals, and purged of air, by being boiled in a glazed earthen vessel, closely covered, and pour ed, when hot, through a glass funnel, with a long capillary tube, into the barometer tube, washed with a rectified spirit, and cleaned with a piston of shammy leather, if both ends were not hermetically sealed when it was made, and heated and ren dered electrical by rubbing. 4. Unless the temperature of the air remain the same, the dimensions of a given quantity of mercury will be variable, and the alti tude of the mercury is an uncertain mea sure of the weight of the atmosphere, be cause it is dilated by heat, and contracted by cold, when perhaps the weight of the atmosphere is unaltered. If very great exactness be therefore required, the dif ference of temperature, at the different times of observation, and the depression or elevation of the mercury produced by it, must be ascertained before the height of the column, raised by the weight of the atmosphere, can be discovered. See WEATHER, rules for judging of.