The Aneroid Barometer.—The aneroid barometer is so convenient on account of its portability that, although much less trustworthy than a mercurial barometer, it is much more likely to be used by a traveller. Care should be taken in using it to see that the pointer has come to a position of equilibrium, and it should be tapped gently before reading. The eye must be brought directly over the end of the pointer, and the reading made to one-hundredth of an inch, the barometer being held in a horizontal position. Every opportunity of comparing the aneroid with a standard mercurial barometer should be taken, and a note made of the readings of both. The mercurial barometer will require to be corrected for temperature before its indications can be used for correcting the aneroid, as all good aneroids are compensated for changes of temperature. The readings of an aneroid give a very fair idea of the changes of atmospheric pressure, and are very much better than none at all, although they cannot in any case be accepted as of the highest order of accuracy.

The Watkin mountain aneroid, which is so constructed as to be thrown into gear at the moment when it is read, appears to be free from the worst errors of the ordinary aneroid.

For climatological purposes, it is impossible to make barometric observations of value while travelling unless the altitude of each camping-place is accurately known. This is practically never the case except when travelling along the sea-shore or the margin of a great lake the elevation of which has been determined. But, meteorology apart, barometric readings in any little known country are of value, because by comparing them with simultaneous readings taken at a neighbouring fixed station, new data as to the altitude of the country may be obtained. While in camp, it would be an extremely useful thing to make barometer readings, even with an aneroid, every two hours, in order to get some information as to the normal daily range of atmospheric pressure.

The Boiling-point Thermometer.—The temperature at which water boils depends on the pressure of the atmosphere, so that an accurate observation of the boiling-point of water enables the pressure of the atmosphere at the moment of observation to be determined with the utmost accuracy. This method of determining atmospheric pressure having been used hitherto almost solely for the purpose of measuring altitudes, the boiling-point thermometer is usually known as the Hypsometer, but its records are quite as valuable for use at fixed stations as in mountain climbing. Mr. J. Y. Buchanan recommends the use of a boiling-point thermometer with a very open scale graduated to fiftieths of a degree Centigrade and entirely enclosed in a wide glass tube through which steam from water boiling in a copper vessel is passing. On a thermometer of this kind change of pressure can be measured by the change of boiling-point more accurately than with the aid of a mercurial barometer. See Table, Vol. I., p. 293.

2. Observations for Forecasting the Weather.—The familiar name of “weather-glass” is appropriately applied to the barometer, for in most parts of the world it is the surest indicator of any approaching storm.

The scientific prediction of the weather by means of the barometer involves the comparison of the simultaneous readings of barometers over as wide an area as possible, and can only be carried out where there is a complete telegraph system and a public department charged with the work. The storms of wind and rain which break the more usual steady weather are usually associated with the formation of centres of low atmospheric pressure towards which wind blows in from every side. These atmospheric depressions move, as a rule, in fairly regular tracks, the rate of movement of the centre of the depression having no relation to the rate at which the wind blows or to the direction of the wind. The term cyclone is usually applied to such a moving depression, because of the rotating winds round the centre; but the size of a cyclone may vary from a vast atmospheric eddy extending across the whole breadth of the Atlantic to one only a few miles in diameter. The strength of the wind in a cyclone depends on the barometric gradient; in other words, the greater the difference in atmospheric pressure between two neighbouring points the stronger is the wind that blows between them. Or, when a cyclone is passing over an observer, the more rapidly the barometer falls or rises the stronger may the wind be expected to blow.

In direct contrast to the cyclone or depression is the system of high pressure rising to a centre from which the wind blows out on every side. This is called an anticyclone, and is a condition which, once established, may last for many days, or even weeks, without change. It is the typical condition for dry calm weather in all parts of the world.

Fig. 9.—Cyclone Paths and Circulation of Winds in Cyclones in the Northern and Southern Hemispheres.

The direction of movement of the centres of cyclones in the northern hemisphere is usually westward and northward near the equator, the path of the centre bending to the right as it proceeds, and becoming ultimately eastward and southward. In the southern hemisphere the direction of the centre near the equator is westward and southward, turning towards the left as it proceeds. The rotation of the wind about the centre of a cyclone in the northern hemisphere is inwards towards the centre in the direction opposite to the hands of a watch, and in the southern hemisphere it is in the direction in which the hands of a watch move (Fig. 9). In the centre of a cyclone there is a calm, a well-known danger to sailing ships caught in such a storm at sea, because there is no wind to move the vessel, but a tremendous sea driven in from the gale which rages all round from every point of the compass. The law of storms has been very fully studied, and rules have been drawn up to enable sailors to ascertain the direction in which the centre of an approaching cyclone lies and the direction in which it is moving. In a work intended mainly for travellers on land it is not necessary to give these rules; all that is required is to tell how the approach of dangerous storms may be ascertained some time in advance. The fact that the barometer is high or low is in itself of no value for prediction. The important thing to know is the distribution of atmospheric pressure at a given moment over a considerable area. To the isolated observer this is impossible, and he can only judge of the state of the atmosphere by observing the rate at which the barometer is falling or rising. Thus, if for several days the barometer has been steadily and slowly rising, he will probably be right in believing that an anticyclonic condition is establishing itself, and that the weather may be expected to continue fine for many days to come, even after a gradual fall of the barometer begins. A sudden fall of the barometer, on the other hand, is always a sign of wind, and usually of wet weather as well. This is a particularly valuable sign of approaching storm in those parts of the world where, as in the tropics, the normal weather is very uniform and steady. In such places a very sudden fall, say one-tenth of an inch in an hour, is a sure precursor of a violent storm. As the barometer continues to fall, the wind will probably continue to increase in force, and when the barometer reaches its lowest point it will either fall calm (if the centre of the cyclone is passing over the observer) or suddenly change in direction. The rapid rise of the barometer after a great depression is also always accompanied by strong wind, though not so frequently by rain.