atmosphere and the ordinary boiling point are then subject to like variations, but these being small, no errors of any importance result from them. 97. Inverted Vessels.-The suspension of water in an inverted vessel is owing to the pressure of the atmosphere. If a tube closed at one end, and of small diameter, be filled with water and inverted, the fluid will remain suspended; and if means be taken to insure the stability of the surface, that is, to prevent the particles from being shaken out of their places, a vessel of any diameter may be inverted. Thus, for instance, if a tumbler be filled with water, and a piece of paper be laid on the surface of the water, the vessel may be inverted, and the atmospheric pressure on the under surface of the paper will prevent the water from running out. The paper is merely a means by which steadiness may be given to the particles at the surface of the fluid. A tumbler so inverted may be set down on a table, and the piece of paper drawn away; the tumbler cannot be raised up, without spilling all the water. The siphon is an instrument in very general use, and whose action may be at once explained by the preceding principles. The water which is sustained in the two legs has a tendency to separate at the upper part, one column running out by each orifice. But this cannot take place without a partial vacuum being formed at the top, and this will be prevented by the atmospheric pressure, unless the columns be more than 32 feet, which is never the case in the siphon: one of these pressures acts directly at the orifice of the longer leg, and the other is transmitted through the surface of the fluid. Now one column is longer than the other, both legs being full, and the longer column being the heavier will draw the other in its direction. This motion will go on continuously, and the siphon will be kept full by the pressure of the air on the surface of the water in the vessel from which it is drawn off. Were the tube not to be kept constantly full there would be a separation at some point which the atmospheric pressure is sufficient to prevent. Thus the motion of a fluid through the siphon is precisely similar to the motion of a smooth chain hanging over a point. If the two parts of the chain are equal the chain remains at rest, but if one portion be longer than the other it moves in the direction of the longer portion. Fresh links, so to speak, are added continuously to this fluid chain by the atmospheric pressure on the surface of the fluid, so that the chain being continuous, the motion is continuous also, and does not cease till one portion of the chain becomes equal to the other; thus the water continues to run through the siphon until the level of the water in the basin coincides with the level of the orifice through which the fluid is discharged. 98. Atmospheric Strata.-The mathematical conditions of equilibrium of a fluid mass, such as the atmosphere is known to be, suggest some important considerations respecting the strata of the atmosphere. The density at any point is proportional to the pressure by the known laws of elastic fluids; hence the density of the atmosphere diminishes gradually from the surface of the earth upwards. The atmosphere, as we shall see presently, must have a limit, and consequently, a bounding surface, and the form of this surface will be nearly spherical; that is, it will be parallel to the surface of the ocean; both the atmosphere and the ocean are seas of fluid matter, and subject to the same forces of gravity and rotation. The bounding surface will therefore be a level surface, and every surface similar to this, or every other level, must, if the atmosphere is in equilibrium, be subject to the following condition: "That the pressure, the density, and the temperature, is the same throughout;'* that is, whatever two or more points of the same level we choose, the pressures on any equal portions must be the same, and consequently, the *Theory of Fluids, Arts. 17-20. densities the same; also they must be of the same temperature. This being the case, the atmosphere will be at rest, and will be composed of concentric level strata, whose densities differ from each other by insensible gradations. These considerations of homogeneous level strata will be found of very great importance in the subject of the refraction of light. The 99. Theory of Winds.—Simple as the preceding condition is, it evidently cannot be fulfilled in the ocean of the atmosphere; an universal calm is inconceivable in a fluid possessing such mobility and elasticity, since a disturbance at a single point will set the whole in motion. But the alternate presence and absence of the sun is the cause to which the disturbances of the atmosphere are to be attributed; this may probably ultimately be considered as the sole cause; it is certainly a very principal one. condition of equilibrium expresses that the temperature must be uniform in the same level, or, in other words, it must be every where the same at the same height above the earth's surface. There are many causes which conspire to prevent this uniformity of temperature. quantity of heat derived from the sun is very various during the twenty-four hours. But besides this, the variations in temperature, arising from local causes, are amply sufficient to disturb the uniformity which is requisite for absolute rest. The The disturbances, however, which must principally be considered, are those due to the temperature for different latitudes. The climates of different parts of the earth's surface are, unquestionably, owing in a great measure to their position with respect to the sun. At the equator, the sun is during all seasons nearly vertical, and any given portion of the surface receives at all times a much greater quantity of heat than an equal portion near the poles; for the passage of the rays of light and heat in a vertical direction is much less interrupted than in an oblique direction. It is owing to the heat lost in consequence of the oblique incidence of the sun's rays that the inhabitants of our latitudes receive so much less heat in winter than in summer, although the earth is much nearer the sun at the former than at the latter period. There is a continuous diminution in the mean temperature as we advance from the equator towards either pole; this may be interrupted at particular places, owing to local causes, but the general law is such as we have stated. The mean temperature of the equator is about 84° F. The temperature of our latitude is exceedingly variable; the mean in January being 36° F., in July and August 61° F., and in December 39° F. ; the mean temperature for the whole year is about 50° F. in London. From the observations of Scoresby, in high northern latitudes, it appears that the mean temperature for latitude 76° 45', is about 18° F., and for latitude 78° about 16° F.: from these and other data the mean temperature of the North pole is supposed to be about 4° F. From these statements it will be at once seen that there must exist about the equatorial regions a belt of air of much higher temperature than exists in other latitudes. The effect of this increased temperature is a dilatation or increase in bulk; the mass so heated and dilated becomes specifically lighter than the surrounding air, rises up, and its place is supplied by colder air, which, becoming heated, is, in its turn, replaced by a fresh quantity. The ascent of the heated air causes a constant upward current, and the motion of the air, which replaces this, causes a constant current flowing in on each side from the poles towards the equator. The heated air having ascended, flows towards each pole. Thus we have two systems of currents, an upper and a lower current; whereof the upper current consists of the hot air, which having ascended from the equatorial regions, is travelling towards the poles, and the lower current is the cold air of higher latitudes, which is travelling towards the equator. We may now endeavour to give some account of the constant winds, such as the trade winds, the monsoons, and other local or periodic winds, which follow some regular law. The upper current would, were there no other causes in operation, travel due north and south; that is, it would be a south wind in the northern, or our hemisphere, and a north wind in the southern hemisphere; but the lower current, on the contrary, would be a north wind in our hemisphere, and a south wind in the southern hemisphere. The directions of these currents are, however, modified by the motion of the rotation of the earth. The atmosphere revolves round with the earth, but since the particles of the air in the lower current are, as they move towards the equator, travelling successively over points of the earth which have a greater and greater linear velocity; the air does not acquire all at once the linear velocity which is due to the latitude to which it has descended, and the particles of the air appear consequently to oppose the motion of the earth, which is from west to east; thus, a north-east wind is created in our hemisphere, and a southeast in the southern hemisphere. These north-east winds are the regular trade winds, which are of such vast importance to mariners, since they furnish a constant wind within 30° on each side of the equator. At some points near the equator the winds will lose their regular easterly character, and will be as irregular as in our own latitude. The air having acquired the velocity of the earth in its approach to the equator will appear to be at rest, or to be simply a north and south wind. The regular trade winds, then, result from the fact that air, being rapidly transferred from higher to lower latitudes, does not acquire all at once the velocity which is due to the distance at which it is from the earth's centre. The motion of the upper current will furnish us with an explanation of some other constant winds, which are known to exist. In the extra-tropical regions there frequently exists a south-westerly wind in the northern hemisphere, and a north-westerly wind in the southern hemisphere. These winds may be seen at once to |