entire height, and that its upper limit is clearly determined. Let AB (pl. I. fig. 1) represent the surface of the earth, and CD the limit of the atmosphere, parallel to AB. If the whole surface AB were equally heated, the air would expand and remove the limit CD further from the earth. But, if the portion EF is heated, while AE and FB preserve the same temperature, then the column of air EFIL will expand, and its upper limit will be more elevated in GH than in IL for example. But, like as a drop of water, when it falls on a liquid surface, extends equally in all directions, so also the drop of air situated between GH and IL passes away in all directions and produces winds, which, as the upper arrow indicates, blow from the hot towards the colder countries.

Whilst these phenomena are going on in the higher regions of the atmosphere, the equilibrium is also destroyed at the level of the ground; the weight of the columns ACIE and FBLD being increased by the whole weight of air, which has been diffused over their upper surface, this increase of weight is communicated in all directions with a facility bearing proportion to the comparative diminution of the column, which has EF for the base. Like as air compressed in a globe rushes out as soon as an opening is made, so does the air flow away from the colder towards the hotter regions, in the direction of the lower arrows.

If the region EF is considerably cooled, the atmosphere situated above it would contract; and the columns of air AE and FB would pass towards EF, whilst two inverse currents would take place at the surface of the globe. The combination of these facts leads to the following conclusion:

If two neighbouring regions are unequally heated, there is produced, in the upper strata, a wind blowing from the hotter to the colder region; and, at the surface of the soil, a contrary

current.

This is the cause of the winds that we observe. The little experiment which follows, and which is due to Franklin, very well represents what takes place in the atmosphere. Open in winter a door communicating between a hot and a cold room, there will be two currents: the one above, and directed from the hot to the cold apartment; the other below, and in a contrary direction.

To be convinced of this, it is only necessary to place two tapers by the door, the one high up and the other low down; the flame of the former will be directed from within outwards, that of the latter in the contrary direction. Some

times these two currents exist above and below a pane of glass when imperfectly puttied in; and we observe, in winter, that a thicker layer of ice is accumulated at its lower part. In a chimney and in a lamp-glass an ascending current is kept up, which feeds the flame; and this current is stronger as the sides of the chimney-funnel or the lampglass are more heated.*

Some philosophers have endeavoured to find in the terrestrial globe itself the cause of all the winds. In mountains, say they, the winds are more violent, because they escape more easily from the bosom of the earth. But, although winds may be stronger in mountainous countries, this is merely because the valleys and the summits determine local currents, the velocity of which is added to that of the principal wind. If violent south winds prevail in the middle of Europe, they acquire a violence, of which we can form no idea, in the valleys of the Swiss Alps, where they are known by the name of foen. Water conducts itself in the same manner, when the bed of a river becomes narrow, or studded with rocks; rapid currents rush in all directions among the rocks, even where the water, a few metres higher up, preserves its tranquillity. De Saussure has likewise observed on the Alps alternations of calm and very violent gales.

Some philosophers have also laid stress upon a fact well known to miners. During and before violent tempests they have observed very strong ascending currents; but it must not be forgotten that storms are almost always preceded, or accompanied, by a great fall in the barometric column. The atmospheric pressure becoming less, the air in the bowels of the earth expands, and ascends to the surface. This phenomenon reminds one of the experiment which has often been made with the air-pump. Place under the receiver a

The winds produced in this way have been called winds of aspiration. The following is a remarkable example. On the 10th of November, 1822, the corvette, Coquille, commanded by Captain DUPERREY, was suddenly assailed by a pampero, a wind of frequent occurrence at the mouth of the Rio de la Plata, although it was more than 1000 kilometres E.N.E. of this latitude. The circumstance, which serves to characterise this wind coming from the land, as a wind of aspiration, occasioned by a rarefaction of the atmosphere of the sea, is, that at the moment when it was felt there was a rapid fall of the barometer. (Comptes rendus de l'Académie des Sciences, t. vii. p. 312.)-M.

† I was at Grindelwald, in the canton of Berne, during the night of the 17th or 18th of July, 1841. In the evening a hot wind began to blow in the valley; it entered by the hollow which separates the Eiger from Mettenberg. Its violence continued increasing during the night, and the following morning nothing was visible on all sides but trees uprooted or broken, and roofs carried away, and transported to great distances. A vast quantity of fragments of ice, detached from the lower glacier of Grindelwald, were stranded in the bed or on the banks of the black Lutschine.-M.

bladder which is well closed, but quite flat, it will swell out as the vacuum is made.

The violent tempests, which frequently accompany volcanic eruptions, are a last argument invoked by some authors. But this coincidence is explained in a very simple manner; to wit, the heat of the volcano determines an ascending current, and the cool air rushes from all sides towards the mountain. The winds, therefore, have a direction the direct opposite to that which they would have if they escaped from the crater of the volcano.

DIFFERENCES PRESENTED BY THE WINDS IN THE DIFFERENT REGIONS OF THE GLOBE. - On examining the winds in all parts of the world, we find important differences, which serve to characterise the climates. On the sea-shore, especially within the tropics, a very regular period is observed every day. At certain determinate hours, the wind blows from the sea,—it is a seabreeze; at other hours the wind comes from the land. In the Atlantic, and the great ocean along the equatorial line, the winds blow almost all the year from the same point of the horizon; those which come from the east are called trade-winds. In India and the neighbouring seas, an annual period is observed in the direction of the wind. For six months the wind blows constantly from one point of the horizon, and for the other six months from another point. These variable winds are called monsoons. In the higher latitudes, all the winds are variable, and the same wind seldom lasts for several successive days.

LAND-WINDS AND SEA-BREEZES. - On coasts, when the weather is calm, no movement is perceived in the air until eight or nine o'clock in the morning, but at that time a sea-breeze gradually rises. It is at first gentle, and is limited to a small space; it gradually increases in force and in extent until three o'clock in the afternoon, then it decreases to give place to the land-wind, which rises soon after sunset, and attains its maximum of velocity and extent at the moment when this body rises.

The direction of these two breezes is perpendicular to that of the coast, but if another wind blows at the same time it is modified in various ways. If the east wind blows near an island, the sea-breeze will be stronger on the east coast of the island, and the land-wind will be weak; on the west side, on the contrary, the land-wind will be stronger than the sea-breeze. On the south coast the direction of the breeze will not be normal to that of the coast; the landwind will blow from the S.E. at the time of its greatest

violence, and the sea-breeze from the N.E. In the course of the day the wind will take all the intermediate directions. In the bosom of gulfs the sea-breezes are very weak, on promontories the land-breezes are weak. These breezes exist between the tropics, and some traces of them have even been noticed in Greenland.

The alternation of these winds is explained by the unequal heating of the land and of the sea. About nine o'clock in the morning the temperature is nearly the same on the land and the sea, and the air is in a state of equilibrium. In proportion as the sun rises above the horizon, the earth becomes more heated than the water; and hence there results an upper land-wind, which is recognised by the motion of elevated clouds, and a sea-breeze blowing in the contrary direction. At the time of the maximum temperature of the day this breeze acquires its greatest force; but, towards evening, the air on the land becomes cool, and, at sunset, it has the same temperature as the sea air. causes a few hours of perfect calm. During the night the land becomes colder than the sea, and a land-wind prevails, the maximum force of which coincides with the time of the minimum temperature of the twenty-four hours, which is likewise that at which the difference of temperature between the earth and the sea is at its extreme.*

This

* M. FOURNET has shewn that there exist in mountains day and night breezes analogous to those of the land and sea. The following is the summary of this memoir, as given by the author himself:

1st. The asperities of the soil daily determine an atmospheric flux and reflux, which is betrayed by ascending and descending breezes, or winds, known, from time immemorial, in certain localities, under the names of thalwind, pontias, vesine, solore, vauderou, rebas, vent du Mont Blanc, aloup du vent.

2d. These currents of air obtain the highest degree of developement in the hollows of valleys, without, however, being peculiar to these situations; for they are manifested along the entire slopes, and the current of valleys is nothing more than the result of ascensions, and lateral and partial cascades (valleys of Cogne, Aoste, Quarazza, plain of St. Symphorien, Pilate, Chessy). 3d. The passage from the flux to the reflux and the converse is rapid in narrow gorges, which, after a short passage, tend toward elevated summits (valleys of Anzasca, Sesia, Visbach, Trient, Cogne, Val-Megnier, Martigny, Simplon): it is slower in the general basins, where the flux is usually not fairly established until ten o'clock in the morning, and where the reflux does not commence regularly until about nine in the evening (valleys of Gier, Azergue, Brevanne, Arc, Aoste, Foccia, Upper Rhone). The interval between the ascending and descending tides is occupied by alternate oscillations, or redundances. The hour of this critical moment varies with the seasons, and also with other accidental meteorological circumstances (valleys of Aoste, Maurienne, Nyons, Gier).

4th. The winds of valleys are regular in regular valleys, but present accidents towards their branchings off; these irregularities may be manifested, according to the mode in which the valleys unite, either in the diurnal period (Martigny, Aoste), or in the nocturnal period (Verres, Bannio, St. Jean de Maurienne, Martigny, Firminy).

5th. The configuration of the upper parts of valleys exercises a still

TRADE-WINDS.- Few phenomena excited so much astonishment among the early navigators, who, in the fifteenth century, ventured into the Atlantic Ocean, as the east winds, which regularly blew within the tropics. The companions of Columbus were struck with terror, when they found themselves driven on by continuous east winds, which seemed to forewarn them that they would never return to their country. For several centuries, the explanation was sought after in vain; at last, Halley and Hadley proposed the following theory :

The regions bordering on the equator are the hottest on the earth, because the sun is at no great distance from their zenith; but, setting out from these zones, the temperature goes on diminishing, in proportion as we approach the poles. There is, therefore, formed an upper current from the equator toward the two poles, and a lower one from the poles to the equator. The air from the poles becomes heated in the neighbourhood of the equator; it ascends and

greater influence over these winds according to the hours and seasons: thus they are sometimes more characteristic by day than by night (Maurienne), at other times more by night than by day (pontias, aloup du vent at Chessy); sometimes winter with its snows is more favourable to the nocturnal winds, at other times summer is to the day-winds. It would be curious to examine, under this relation, the elliptical circuits, which the upper and terminal parts of the Jurassic and subalpine valleys form, compared with the gentle and insensible terminations of the primitive mountains. In the valley of Joux, for example, the alternations from heat to cold are so sudden, that variations of 20° are often experienced there in a few hours, and reapers may be seen in the morning cutting ice with their sickles, whilst a few hours afterwards the thermometer in the sun indicates 38°; it is impossible for such differences as these not to produce extraordinary currents.

6th. The effect of these tides is generally more decided in large valleys, and is weakened in lateral ramifications (Maurienne, Aoste). However, when the basin becomes a true plain, capable of supplying large demands, or absorbing a considerable mass, the effects are reduced: thus the pontias rarely reaches the course of the Rhone; and around Geneva the valley breezes of the Arve appear so reduced as not to have excited the attention of the talented philosophers of that city. However, this fact must be verified hereafter.

7th. In comparing the phenomenon of tides about mountains with that of the sea and land-breezes, which are reciprocally produced along the coasts, we perceive that, at the same period when the diurnal sea-winds are driving ships to harbours, the aerial wave is on its part rising about the mountains; and in the night the reverse is the case. It follows, therefore, from this, that the whole of the atmosphere of the Rhone must be daily subject to a motion, which, on the one hand, carries it from the sea to the continent, and from the latter toward the summits of the plain of central France, or of that of the Alps and the Jura; after which it would return, during the night, to its place of departure. But the slowness with which any movement is transmitted in a great mass of an elastic fluid partially nullifies these effects. However, this annihilation is not always complete; and, for the future, I am induced to believe that the light currents which are manifested during the day in the neighbourhood of Lyon, and which may, in some sense, be considered calms, are merely the result of those oscillations, the effects of which I shall develope on another occasion.

8th. The atmospheric tides drive with them bodies that have the power