43. The principle on which the Aneroid Barometer depends is the varying pressure of the atmosphere upon an elastic metallic chamber partially exhausted of its air, and so constructed that by a system of levers a motion is given to an index-hand which moves upon a dial. As it is very portable, and is not liable to be broken in carriage like mercurial barometers, it is peculiarly suited for nautical purposes, and for taking the heights of mountains. It requires, however, to be occasionally compared with a standard mercurial barometer, being liable to variation from the elasticity of the brass of the chamber changing, or from changes in the system of levers which work the index-hand. Though aneroids may be constructed showing great accuracy in their indications, yet none can lay claim to the exactness of mercurial barometers. The internal machinery of this instrument is liable to get fouled, in which state I have seen them change 0.300 inch in a few weeks, and indicate pressure so irregularly that no confidence could be placed in them even for a few days. Aneroids therefore, like our watches, require to be occasionally cleaned.

44. The Sympiesometer, fig. 8, was invented by Adie. It consists of a glass tube, 18 inches in length and 3 inch in diameter, with a small chamber at the top, and an open cistern below. The upper part of the tube is now filled with common air, and the lower part and cistern with glycerin. Hence when the pressure of the atmosphere is increased, the air is compressed, and the fluid rises; but when it is diminished, the air expands and the fluid falls. To obviate error from the increased pressure of the air when its temperature is raised, a thermometer and sliding scale are added to the instrument, so that it may be adjusted to the temperature at each observation. It is a very sensitive instrument, and well adapted for being used at sea and by travellers,

Fig. 8.

but it is not suited for exact observation. The glycerin used in filling the tube requires to be of a particular consistence, otherwise the air in the tube will be partially absorbed, and the instrument be thereby liable to change, as happened when the tube was filled, as it originally was, with hydrogen gas and oil of almonds.

45. The Hermetic Barometer, sometimes called Poor Man's Weather-glass, fig. 9, consists of a glass tube with a large, nearly flat, bulb of thin glass at the base, filled with common air and spirits of wine, and then hermetically closed. The thin glass bulb, being elastic, is subject to compression and dilatation as the atmospheric pressure is increased or diminished; and as a small quantity of air is left in the top of the tube, the column of spirits of wine rises or falls with the pressure of the air. To insure uniformity in its working, it requires to be kept as near as possible at the same temperature. It is of no use as a scientific instrument; but it may be referred to as the cheapest instrument by which variations in the atmospheric pressure may be roughly indicated. From the rapid absorption of the air by the spirits of wine, the indications change so quickly that in the short space of a year most of them will be found reading four or five tenths of an inch too high. If the tubes were filled as those of sympiesometers are, this objection would be obviated.

46. How to use the Vernier.-A vernier is an instrument for reading off the gradu- so ated scale of the barometer true to the Tooth orth part of an inch. It consists (figs. 10 and 11) of a piece similar to the scale of the barometer, and along which it slides. It will be observed from fig. 10, that ten divisions of the vernier are exactly equal to eleven divisions of the scale; that Fig. 10.

Fig. 9.

0

30

2

2

6

4

8

6

29

29

10

8

10

Fig. 11.

is, to eleven tenths of an inch. Hence each division of the vernier is equal to a tenth of an inch, together with the tenth of a tenth, or a hundredth, or to ten hundredths and one hundredth-that is, to eleven hundredths of an inch. Similarly two divisions of the vernier are equal to twenty-two hundredths of an inch, which, expressed as a decimal fraction, is 0.22 inch; three divisions of the vernier to 0.33 inch, &c. Suppose the vernier set as described and figured at page 19— that is, having the zero line of the vernier a tangent to the convex curve of the mercury in the column. If the vernier and scale occupy the relative positions as in fig. 10, then the height of the barometer is 30.00 inches. But if they stand as in fig. 11, we set about reading it in this way :—(1.) The zero of the vernier being between 29 and 30, the reading is more than 29 inches, but less than 30 inches, we obtain the first figure

29 inches.

(2.) Counting the tenths of an inch from 29 upward, we find that the vernier indicates more than 7 tenths, and less than 8 tenths, giving the second figure

7 tenths, or 0.7 inch.

(3.) Casting the eye down the scale to see the point at which a division of the scale and a division of the vernier lie in one and the same straight line, we observe this to take place at the figure 6 of the vernier; this gives the last figure—

6 hundredths, or 0.06 inch.

And placing all these figures in one line, we find that the height of the barometer is 29.76 inches. This sort of vernier gives readings true to the hundredth of an inch. If the inch be divided into half-tenths or twentieths, and 25 divisions of the vernier equal 24 divisions of the scale, it follows that the difference of these divisions is two thousandths of an inch. This vernier is always used with the best barometers; and though a little troublesome to most people to read at first, yet if the method of reading the simpler one be understood, the difficulty will be easily overcome.

47. Reduction of Barometric Readings to 32°.-Mercury expands of its bulk for every degree of Fahrenheit's thermometer; if, then, a barometer stands at a height of 30 inches when the temperature of its mercury is 32°, it will stand at 30 inches if the temperature be raised to 69°. This increase of the length of the column by the tenth of an inch is not due to any increased pressure, but solely to the expansion of the mercury under a higher temperature. In order, therefore, to compare together barometric observations with exactness, it is necessary to reduce them to the heights at which they would stand at some uniform temperature. The temperature to which they are generally reduced is 32°. The correction for this reduction is found by dividing by 9990 the difference between the observed temperature and 32°, and subtracting or adding the result to the observed height, according as the temperature is above or below 32°.

48. Table I., at the end, gives the temperature corrections, adopted by the Royal Society of London, in decimals of an inch for every degree from 29° to 90°, and for every half-inch of pressure from 27.0 to 30.5 inches. The scale is supposed to be brass, extending from the cistern to the top of the column, the difference between the expansion of brass and mercury being allowed for in the table. Since the standard temperature of the English yard is 62°, and not 32°, the dif ference of expansion of the scale and the mercurial column carries the point of no correction down to 28°.5. The table may be used for temperatures lower than 28°.5, by noting how many degrees the given temperature is below 28°.5, and then, looking at the temperature which is just so far above 28°.5, the correction will be found, but it must be added instead of being subtracted. Thus, suppose we wish to find the temperature correction at 20°.0, the height of the barometer being 30° inches. Looking for the correction at 37°.0, which is as far above 28°.5 as 20°.0 is below it, we find it to be -.023; and hence the correction for 20°.0 is +.023.

49. The column of mercury in the tube is supported above

the level of the mercury in the cistern by the pressure of a column of the atmosphere having a base equal to that of the column. Hence the weight of this atmospheric column is the same as that of the column of mercury. Now, if we suppose the mercurial column to be 30 inches, which is nearly the average height at the level of the sea, and its base a square inch, it will contain 30 cubic inches of mercury; and since one cubic inch of mercury contains 34263 grains, the weight of 30 cubic inches will be nearly 14.7 lb. avoirdupois. Thus the pressure of the atmosphere on every square inch of the earth's surface is 14.7 lb.

50. Mode of estimating Atmospheric Pressure.-The pressure of the atmosphere is not expressed by the weight of the mercury sustained in the tube by that pressure, but by the perpendicular height of the column. Thus when the height of the column is 30 inches, we do not say that the atmospheric pressure is 14.7 lb. on the square inch, but that it is 30 inches, meaning that the pressure will sustain a column of mercury at that height.

51. In England and America the height of the barometer is expressed in English inches, and in Russia by half-lines. As a half-line equals half an English decimal line, or the twentieth part of an English inch, the Russian barometric observations are reducible to the scale of English inches by dividing by 20. In France and most countries of Europe the height is expressed in millimetres-a millimetre being the thousandth part of the French metre, which equals 39.37079 English inches. The Old French scale, in which the unit is the French or Paris line (0.088814 inch), is still in use in a few countries on the Continent. Table II. will be found useful in comparing millimetres with English inches, and Table III. in comparing Paris lines with English inches. In the Old French barometer, Paris lines are frequently written in Paris inches, 12 lines being equal to an inch. Hence 300 lines equal 25 inches, 312 lines 26 inches, 324 lines 27 inches, and 336 lines 28 inches, &c. The English measure of length being a standard at 62° Fahr., the Old French meas