warmed by the solar rays as that luminary approaches to the summer solstice. In summer and autumn the opposite winds prevail, and the temperature is more moist and warm; because the balance of temperature being now in favour of the land, the winds blow from the ocean. As the year advances, the winds first noticed again predominate, but they are accompanied by greater humidity than in the spring."* Usually the wind is stronger in February and March than at any other time, but at all seasons is most powerful at noon.

The following table represents the relative prevalence of winds in different countries:-†

The relative temperatures of the different winds, although the circumstances under which the observations are made vary much, are found, on the whole, to present some interesting facts. The following table, relating to the winds at the under-mentioned places, is derived from M. Otto Eisenlohr :

This table conclusively proves that a north-east is the coldest wind there is—a fact which all whose pursuits necessitate their being out much in the open air will corroborate.

The subject of trade winds, monsoons, cyclones, typhoons, local hurricanes, and storms in general,‡ would occupy more space

*Thomson's Introd. to Met. p. 426.
+ Scoffern, Meteorology, p. 484.
Refer to Reid's Law of Storms, &c.

than we can now afford; indeed it belongs rather to the department of Physical Geography.

The northern parts of China are frequently visited by remarkable dust-storms, which may be well referred to here. They come on suddenly, and a thick darkness covers the land for many hours. On March 26, 1862, an extraordinary one occurred, which is thus described by Dr. Lamprey, of the 67th Regiment :

"During the greater portion of the day the wind was blowing from the south in rather strong gusts, almost approaching to a gale. At 3h. 15m. P.M. the wind suddenly shifted to NNW. in a strange manner, when all at once the air was filled with dust, and the sun and light became obscured as if a total eclipse had suddenly occurred. The wind blowing a hurricane, the darkness had something appalling in it; it was so sudden and unusual. It was noticed that the darkness would lighten somewhat at intervals of a quarter of an hour or 20 minutes, as if about to abate, when it would suddenly increase again along with [accompanied by] a slight change of wind to the NNE. The wind, endeavouring to become more northerly, would suddenly turn again to this point, and bring along with it an increase of the dust. This state continued till about midnight, when there was a slight calm, which lasted till about 8 o'clock the following morning.

"On the occurrence of this dust-storm there was a rapid fall of the thermometer, and during the night a hard frost set in. At the place where I made these observations-about 30 miles from Tien-tsin-I noticed a poor man of 70 years of age, in a very excited state, crying out 'What day is this?' and low-le,' the words corresponding to O Providence.' He would give me little or no information on the subject; but from others I learnt that such heavy storms had occurred before; and on my return to Tien-tsin I ascertained that one happened 36 years ago which lasted a fortnight, and that the light during that time was somewhat like the dusk of evening. I also learned that the Chinese designated this storm as a red one-their classification being as follows: white, yellow, red, and black; depending on the amount of light, or more properly speaking, [on] the amount of dust in the atmosphere at the time. The quantity of dust that covered one's person was astonishing, and the clothes, &c., inside the coat were thickly covered with it. On drawing a feather through the fingers it became strongly electrified.

"When I returned to Tien-tsin I learned that the dust-storm occurred there about the same time we experienced it 30 miles NW., and that the electric conductor showed an extraordinary quantity of electricity-a large blue flame poured out of the end of the conductor without intermission. There being little or no 'day-] light, this showed a beautiful appearance, while the sound of it was quite audible at some distance off, and the shock felt on touching the conductor was powerful. The quantity of dust that entered the house was very great. There was a recurrence of the duststorm on the 27th, but not approaching in strength to the one on the 26th. It was also ascertained that it occurred at Pekin and at Taku, where the fury of the storm was greatest. Several country boats were wrecked in the river Pei-ho. Repeated observations have enabled me to come to the conclusion that these dust-storms are owing to the electric condition of the atmosphere."* G. F. C.

*Letter in North China Herald, cited in Fortune's Visits to Japan and China, pp. 335-7. 8vo. London. 1863.

THE BAROMETER.

(BY G. F. CHAMBERS, F.R.A.S.)

As its name implies,* this instrument is used for measuring the pressure of the atmosphere. The principle upon which it is constructed is extremely simple. It is an instrument which, by virtue of its peculiar construction, weighs or balances a column of air 40 or 50 miles in height, and of a diameter equal to its own tube. Suppose that this was 4th of an inch in diameter, if it were possible directly to weigh a column of air 4th of an inch in diameter and 40 or 50 miles in height, and also to weigh a column of mercury 4th of an inch in diameter and about 30 inches in height, we should find that the weight of these columns would be equal.

This can be put to an indirect test in the following manner: Take a glass tube about 30 inches long, closed at one end, and fill it with mercury; place the open end in a cup containing mercury, keeping the finger pressed tightly against the open end until it is quite immersed. Erect the tube and withdraw the finger, and it will be found that the mercury will not run out of the former, because the pressure of the superincumbent air acting on the surface of the mercury in the cup sustains that which is in the tube. Should the pressure of the atmosphere be diminished, the fact will soon become known by the depression which takes place in the level of the column in the tube. If we employ for this experiment a tube, say 36 inches in length, and, after filling it, plunge it as before into a cup of mercury, it will be found that the mercury in the tube empties itself until the height of the column is reduced to (about) 30 inches, thus clearly proving the accuracy of the reasoning adduced above.

In the actual construction of barometers, there are several precautions to be attended to, chiefly, however, of a mechanical character. It is of special importance that the mercury should be free from impurities,† the presence of which would vitiate the results of the observations, and that both the tube and the mercury should be perfectly dry. A perfect expulsion of the atmospheric air is also indispensable for a first-class standard instrument. This is effected by boiling the mercury after it has been placed in the tube-an operation of great delicacy and difficulty, often resulting in the tube being broken.

* Bapos, weight; and μérpov, a measure.

In a crude state, mercury contains varying quantities of lead, tin, and sometimes zinc: nitric or acetic acid is employed to remove these.

The following is a description of a first-class standard barometer. It is on Fortin's principle, reading from an ivory point in the cistern to ensure a constant level :-The tube is enclosed for protection in a tube of brass extending throughout its whole length; in the upper portion of this are two longitudinal openings opposite each other-on one side of the front opening is the scale, reading by a vernier to th of an inch; on the opposite side is sometimes placed a scale, divided to show French milimetres, reading also by a vernier toth of a milimetre. The reservoir or cistern is of glass, closed at the bottom by means of a leather bag, acted upon by a thumbscrew passing through the bottom of an arrangement of brass work, by which it is protected. In another form of good though not standard barometer, which is coming much into use, especially for exhibition in places of public resort, the graduation of the scale is contracted so as fully to compensate for the varying height of the mercury in the cistern; and as this correction is made by mathematical calculation, the result is so perfect as frequently to obtain a Kew verification of 'ool or '002 of an inch.

Fortin's Barometer differs from the ordinary ones in the plan adopted for bringing the lower level of the mercury exactly to the zero point before reading off an observation. It may be described as a Torricellian barometer, in which the bottom of the cistern can be raised or lowered by a screw. An ivory needle points downwards, and the extremity of this is on a level with the zero of the scale. The bottom of the cistern is raised or lowered by the screw, until the point of the needle and its image in the mercury exactly coincide. The reading is then taken. For full account of all the different mercurial barometers ever devised, the reader is referred to the English Cyclopædia.

Aneroid Barometer is the name applied to an ingenious mechanical contrivance for indicating changes in the pressure of the atmosphere. The theory of its action depends on the effect produced by this pressure on a flat circular metallic chamber exhausted of air and hermetically sealed, and kept in a state of tension by a strong metallic spring; thus the chamber and spring is a substitute for the Torricellian tube, and the vacuum for the column of mercury. The external appearance of the instrument is something like that of a ship's chronometer. On the dial-plate is a curved scale to represent inches from 28 to 31. The pressure on the interior vacuum chamber is communicated by a system of levers to an index or clock-hand placed above the face of the dial, and thus minute changes in the atmospheric pressure are (by being multiplied) promptly rendered perceptible to the observer.† * α, without; and vnpòs, fluid.

+ Described fully in Thomson's Introd. to Met.

The aneroid barometer, though not an independent instrument, is, by reason of its compactness and general accuracy (after being once set by a standard mercurial one), much approved of as a "weather-glass; " and Sir H. James considers that "altitudes not exceeding 2,000 feet can be determined with it very approximately," though it cannot, of course, be depended upon in the same way that a mercurial barometer can be.* This instrument was invented by M. Vidi, of Paris, in 1847.

Bourdon's Metallic Barometer externally somewhat resembles the aneroid. It consists of a flattened metal tube, bent into the form of a circle, and nearly exhausted of air. In this condition, when the atmospheric pressure is diminished, the ends recede; when it is increased, they approach each other. A lever attached to one end of the tube by suitable mechanism imparts motion to an index playing above a dial-plate, as in the aneroid. The graduations and method of graduating are the same for both. This instrument was invented in 1850.

Various plans for self-registering barometers have been devised, but the only one which has come into use and that only in large observatories-is the one in which photography and clockwork play the principal parts. Sensitised paper is arranged to pass regularly in front of the vacuum of the barometer, on which is directed the light of a lamp; by this means a sinuous streak is marked on the paper, and by suitable means the varying elevation of the mercurial column may be ascertained. An apparatus of this kind is in use at the Royal Observatory, Greenwich. In the great Exhibition of 1862 there was exhibited an ingenious selfregistering contrivance, invented by Vice-Admiral Sir Alexander Milne, K.C.B.

In taking barometric observations, it is necessary to apply certain corrections for temperature, capillarity, index-error, elevation above the sea-level, &c.

Though mercury is almost invariably used, it may be observed that any liquid will answer the purpose, though not equally well. Mercury possesses two great advantages over all other liquids, which has led to its now universal employment in the construction of barometers. (1.) That it does not give off vapour at ordinary temperatures. If it did, it is evident that the space above the column would be filled with an elastic vapour, which would press downwards against the column, so that its weight would no longer be a measure of the atmospheric pressure, but of the difference of this pressure and the elastic force of the vapour given off. (2.) That the specific gravity of mercury is much greater than that of any other liquid. The height of the column,

* Meteorological Instructions, &c., p. 17.