The effect of the heated air which acquires its temperature by radiation from the earth's surface is to produce the winds. This it brings about in a very simple manner, though the details of the process have a certain complication. The best illustration of the mode in which the winds are produced is obtained by watching what takes place about an ordinary fire at the bottom of a chimney. As soon as the fire is lit, we observe that the air about it, so far as it is heated, tends upward, drawing the smoke with it. If the air in the chimney be cold, it may not draw well at first; but in a few minutes the draught is established, or, in other words, the heated lower air breaks its way up the shaft, gradually pushing the cooler matter out at the top. In still air we may observe the column from the flue extending about the chimney-top, sometimes to the height of a hundred feet or more before it is broken to pieces. It is well here to note the fact that the energy of the draught in a chimney is, with a given heat of fire and amount of air which is permitted to enter the shaft, directly proportionate to the height; thus in very tall flues, between two and three hundred feet high, which are sometimes constructed, the uprush is at the speed of a gale.

Whenever the air next the surface is so far heated that it may overcome the inertia of the cooler air above, it forces its way up through it in the general manner indicated in the chimney flue. When such a place of uprush is established, the hot air next the surface flows in all directions toward the shaft, joining the expedition to the heights of the atmosphere. Owing to the conditions of the earth's surface, which we shall now proceed to trace, these ascents of heated air belong in two distinct classes—those which move upward through more or less cylindrical chimneys in the atmosphere, shafts which are impermanent, which vary in diameter from a few feet to fifty or perhaps a hundred miles, and which move over the surface of the earth; and another which consists of a broad, beltlike shaft in the equatorial regions, which in a way girdles the earth, remains in about the same place, continually endures, and has a width of hundreds of miles. Of these two classes of uprushes we shall first consider the greatest, which occurs in the central portions of the tropical realm.

Under the equator, owing to the fact that the sun for a considerable belt of land and sea maintains the earth at a high temperature, there is a general updraught which began many million years ago, probably before the origin of life, in the age when our atmosphere assumed its present conditions. Into this region the cooler air from the north and south necessarily flows, in part pressed in by the weight of the cold air which overlies it, but aided in its motion by the fact that the particles which ascend leave place for others to occupy. Over the surfaces of the land within the tropical region this draught toward what we may term the equatorial chimney is perturbed by the irregularities of the surface and many local accidents. But on the sea, where the conditions are uniform, the air moving toward the point of ascent is marked in the trade winds, which blow with a steadfast sweep down toward the equator. Many slight actions, such as the movement of the hot and cold currents of the sea, the local air movements from the lands or from detached islands, somewhat perturb the trade winds, but they remain among the most permanent features in this changeable world. It is doubtful if anything on this sphere except the atoms and molecules of matter have varied as little as the trade winds in the centre of the wide ocean. So steadfast and uniform are they that it is said that the helm and sails of a ship may be set near the west coast of South America and be left unchanged for a voyage which will carry the navigator in their belt across the width of the Pacific.

Rising up from the earth in the tropical belt, the air attains the height of several thousand feet; it then begins to curve off toward the north and south, and at the height of somewhere about three to five miles above the surface is again moving horizontally toward either pole; attaining a distance on that journey, it gradually settles down to the surface of the earth, and ceases to move toward higher latitudes. If the earth did not revolve upon its axis the course of these winds along the surface toward the equator, and in the upper air back toward the poles, would be made in what we may call a square manner—that is, the particles of air would move toward the point where they begin to rise upward in due north and south lines, according as they came from the southern or northern hemisphere, and the upper currents or counter trades would retrace their paths also parallel with the meridians or longitude lines. But because the earth revolves from west to east, the course of the trade winds is oblique to the equator, those in the northern hemisphere blowing from northeast to southwest, those in the southern from southeast to northwest. The way in which the motion of the earth affects the direction of these currents is not difficult to understand. It is as follows:

Let us conceive a particle of air situated immediately over the earth's polar axis. Such an atom would by the rotation of the sphere accomplish no motion except, indeed, that it might turn round on its own centre. It would acquire no velocity whatever by virtue of the earth's movement. Then let us imagine the particle moving toward the equator with the speed of an ordinary wind. At every step of its journey toward lower latitudes it would come into regions having a greater movement than those which it had just left. Owing to its inertia, it would thus tend continually to lag behind the particles of matter about it. It would thus fall off to the westward, and, in place of moving due south, would in the northern hemisphere drift to the southwest, and in the southern hemisphere toward the northwest. A good illustration of this action may be obtained from an ordinary turn-table such as is used about railway stations to reverse the position of a locomotive. If the observer will stand in the centre of such a table while it is being turned round he will perceive that his body is not swayed to the right or left. If he will then try to walk toward the periphery of the rotating disk, he will readily note that it is very difficult, if not impossible, to walk along the radius of the circle; he naturally falls behind in the movement, so that his path is a curved line exactly such as is followed by the winds which move toward the equator in the trades. If now he rests a moment on the periphery of the table, so that his body acquires the velocity of the disk at that point, and then endeavours to walk toward the centre, he will find that again he can not go directly; his path deviates in the opposite direction—in other words, the body continually going to a place having a less rate of movement by virtue of the rotation of the earth, on account of its momentum is ever moving faster than the surface over which it passes. This experiment can readily be tried on any small rotating disk, such as a potter's wheel, or by rolling a marble or a shot from the centre to the circumference and from the circumference to the centre. A little reflection will show the inquirer how these illustrations clearly account for the oblique though opposite sets of the trade winds in the upper and lower parts of the air.

The dominating effect of the tropical heat in controlling the movements of the air currents extends, on the ocean surface, in general about as far north and south as the parallels of forty degrees, considerably exceeding the limits of the tropics, those lines where the sun, because of the inclination of the earth's axis, at some time of the year comes just overhead. Between these belts of trade winds there is a strip or belt under the region where the atmosphere is rising from the earth, in which the winds are irregular and have little energy. This region of the "doldrums" or frequent calms is one of much trouble to sailing ships on their voyages from one hemisphere to another. In passing through it their sails are filled only by the airs of local storms, or winds which make their way into that part of the sea from the neighbouring continents. Beyond the trade-wind belt, toward the poles, the movements of the atmosphere are dependent in part on the counter trades which descend to the surface of the earth in latitudes higher than that in which the surface or trade winds flow. Thus along our Atlantic coast, and even in the body of the continent, at times when the air is not controlled by some local storm, the counter trade blows with considerable regularity.

The effect of the trade and counter-trade movements of the air on the distribution of temperature over the earth's surface is momentous. In part their influence is due to the direct heat-carrying power of the atmosphere; in larger measure it is brought about by the movement of the ocean waters which they induce. Atmospheric air, when deprived of the water which it ordinarily contains, has very little heat-containing capacity. Practically nearly all the power of conveying heat which it possesses is due to the vapour of water which it contains. By virtue of this moisture the winds do a good deal to transfer heat from the tropical or superheated portion of the earth's surface to the circumpolar or underheated realms. At first, the relatively cool air which journeys toward the equator along the surface of the sea constantly gains in heat, and in that process takes up more and more water, for precisely the same reason that causes anything to dry more rapidly in air which has been warmed next a fire. The result is that before it begins to ascend in the tropical updraught, being much moisture-laden, the atmosphere stores a good deal of heat. As it rises, rarefies, and cools, the moisture descends in the torrential rains which ordinarily fall when the sun is nearly vertical in the tropical belt.

Here comes in a very interesting principle which is of importance in understanding the nature of great storms, either the continuous storm of the tropics or the local and irregular whirlings which occur in various parts of the earth. When the moisture-laden air starts on its upward journey from the earth it has, by virtue of the watery vapour which it contains, a store of energy which becomes applied to promoting the updraught. As it rises, the moisture in the air gathers together or condenses, and in so doing parts with the heat which caused it to evaporate from the ocean surface. For a given weight of water, the amount of heat required to effect the evaporation is very great; this we may roughly judge by observing what a continuous fire is required to send a pint of water into the state of steam. This energy, when it is released by the condensation of water into rain or snow, becomes again heat, and tends somewhat, as does the fire in the chimney, to accelerate the upward passage of the air. The result is that the water which ascends in the equatorial updraught becomes what we may term fuel to promote this important element in the earth's aerial circulation. Trades and counter trades would doubtless exist but for the efficiency of this updraught, which is caused by the condensation of watery vapour, but the movement would be much less than it is.

WHIRLING STORMS.

In the region near the equator, or near the line of highest temperature, which for various reasons does not exactly follow the equator, there is, as we have noticed, a somewhat continuous uprushing current where the air passes upward through an ascending chimney, which in a way girdles the sea-covered part of the earth. In this region the movements of the air are to a great extent under the control of the great continuous updraught. As we go to the north and south we enter realms where the air at the surface of the earth is, by the heat which it acquires from contact with that surface, more or less impelled upward; but there being no permanent updraught for its escape, it from time to time breaks through the roof of cold air which overlies it and makes a temporary channel of passage. Going polarward from the equator, we first encounter these local and temporary upcastings of the air near the margin of the tropical belt. In these districts, at least over the warmer seas, during the time of the year when it is midsummer, and in the regions where the trade winds are not strong enough to sweep the warm and moisture-laden air down to the equatorial belt, the upward tending strain of the atmosphere next the earth often becomes so strong that the overlying air is displaced, forming a channel through which the air swiftly passes. As the moisture condenses in the way before noted, the energy set free serves to accelerate the updraught, and a hurricane is begun. At first the movement is small and of no great speed, but as the amount of air tending upward is likely to be great, as is also the amount of moisture which it contains, the aerial chimney is rapidly enlarged, and the speed of the rising air increased. The atmosphere next the surface of the sea flows in toward the channel of escape; its passage is marked by winds which are blowing toward the centre. On the periphery of the movement the particles move slowly, but as they win their way toward the centre they travel with accelerating velocity. On the principle which determines the whirling movement of the water escaping through a hole in the bottom of a basin, the particles of the air do not move on straight lines toward the centre, but journey in spiral paths, at first along the surface, and then ascending.

We have noted the fact that in a basin of water the direction of the whirling is what we may term accidental—that is, dependent on conditions so slight that they elude our observation—but in hurricanes a certain fact determines in an arbitrary way the direction in which the spin shall take place. As soon as such a movement of the air attains any considerable diameter, although in its beginning it may have spun in a direction brought about by local accidents, it will be affected by the diverse rates of travel, by virtue of the earth's rotation, of the air on its equatorial and polar sides. On the equatorial side this air is moving more rapidly than it is on the polar side. By observing the water passing from a basin this principle, with a few experiments, can be made plain. The result is to cause these great whirlwinds of the hurricanes of higher latitudes to whirl round from right to left in the northern hemisphere and in the reverse way in the southern. The general system of the air currents still further affects these, as other whirling storms, by driving their centres or chimneys over the surface of the earth. The principle on which this is done may be readily understood by observing how the air shaft above a chimney, through which we may observe the smoke to rise during a time of calm, is drawn off to one side by the slight current which exists even when we feel no wind; it may also be discerned in the little dust whirls which form in the streets on a summer day when the air is not much disturbed. While they spin they move on in the direction of the air drift. In this way a hurricane originating in the Gulf of Mexico may gradually journey under the influence of the counter trades across the Antilles, or over southern Florida, and thence pursue a devious northerly course, generally near the Atlantic coast and in the path of the Gulf Stream, until it has travelled a thousand miles or more toward the North Atlantic. The farther it goes northward the less effectively it is fed with warm and moisture-laden air, the feebler its movement becomes, until at length it is broken up by the variable winds which it encounters.

A very interesting and, from the point of view of the navigator, important peculiarity of these whirls is that at their centre there is a calm, similar in origin and nature to the calm under the equator between the trade-wind belts. Both these areas are in the field where the air is ascending, and therefore at the surface of the earth does not affect the sails of ships, though if men ever come to use flying machines and sail through the tropics at a good height above the sea it will be sensible enough. The difference between the doldrum of the equator and that of the hurricane, besides their relative areas, is that one is a belt and the other a disk. If the seafarer happens to sail on a path which leads him through the hurricane centre, he will first discern, as from the untroubled air and sea he approaches the periphery of the storm, the horizon toward the disturbance beset by troubled clouds, all moving in one direction. Entering beneath this pall, he finds a steadily increasing wind, which in twenty miles of sailing may, and in a hundred miles surely will, compel him to take in all but his storm sails, and is likely to bring his ship into grave peril. The most furious winds the mariner knows are those which he encounters as he approaches the still centre. These trials are made the more appalling by the fact that in the furious part of the whirl the rain, condensing from the ascending air, falls in torrents, and the electricity generated in the condensation gives rise to vivid lightning. If the storm-beset ship can maintain her way, in a score or two of miles of journey toward the centre, generally very quickly, it passes into the calm disk, where the winds, blowing upward, cease to be felt. In this area the ship is not out of danger, for the waves, rolling in from the disturbed areas on either side, make a torment of cross seas, where it is hard to control the movements of a sailing vessel because the impulse of the winds is lost. Passing through this disk of calm, the ship re-encounters in reverse order the furious portion of the whirl, afterward the lessening winds, until it escapes again into the airs which are not involved in the great torment.

In the old days, before Dove's studies of storms had shown the laws of hurricane movement, unhappy shipmasters were likely to be caught and retained in hurricanes, and to battle with them for weeks until their vessels were beaten to pieces. Now the "Sailing Directions," which are the mariner's guide, enable him, from the direction of the winds and the known laws of motion of the storm centre, to sail out of the danger, so that in most cases he may escape calamity. It is otherwise with the people who dwell upon the land over which these atmospheric convulsions sweep. Fortunately, where these great whirlwinds trespass on the continent, they quickly die out, because of the relative lack of moisture which serves to stimulate the uprush which creates them. Thus in their more violent forms hurricanes are only felt near the sea, and generally on islands and peninsulas. There the hurricane winds, by the swiftness of their movement, which often attains a speed of a hundred miles or more, apply a great deal of energy to all obstacles in their path. The pressure thus produced is only less destructive than that which is brought about by the tornadoes, which are next to be described.

There is another effect from hurricanes which is even more destructive to life than that caused by the direct action of the wind. In these whirlings great differences in atmospheric pressure are brought about in contiguous areas of sea. The result is a sudden elevation in the level of one part of the water. These disturbances, where the shore lands are low and thickly peopled, as is the case along the western coast of the Bay of Bengal, may produce inundations which are terribly destructive to life and property. They are known also in southern Florida and along the islands of the Caribbean, but in that region are not so often damaging to mankind.

Fortunately, hurricanes are limited to a very small part of the tropical district. They occur only in those regions, on the eastern faces of tropical lands, where the general westerly set of the winds favours the accumulation of great bodies of very warm, moist air next the surface of the sea. The western portion of the Gulf of Mexico and the Caribbean, the Bay of Bengal, and the southeastern portion of Asia are especially liable to their visitations. They sometimes develop, though with less fury, in other parts of the tropics. On the western coast of South America and Africa, where the oceans are visited by the dry land winds, and where the waters are cooled by currents setting in from high latitudes, they are unknown.

Only less in order of magnitude than the hurricanes are the circular storms known as cyclones. These occur on the continents, especially where they afford broad plains little interrupted by mountain ranges. They are particularly well exhibited in that part of North America north of Mexico and south of Hudson Bay. Like the hurricanes, they appear to be due to the inrush of relatively warm air entering an updraught which had been formed in the overlying, cooler portions of the atmosphere. They are, however, much less energetic, and often of greater size than the hurricane whirl. The lack of energy is probably due to the comparative dryness of the air. The greater width of the ascending column may perhaps be accounted for by the fact that, originating at a considerable height above the sea, they have a less thickness of air to break through, and so the upward setting column is readily made broad.

The cyclones of North America appear generally to originate in the region of the Rocky Mountains, though it is probable that in some instances, perhaps in many, the upward set of the air which begins the storm originates in the ocean along the Pacific coast. They gather energy as they descend the great sloping plain leading eastward from the Rocky Mountains to the central portion of the great continental valley. Thence they move on across the country to the Atlantic coast. Not infrequently they continue on over the ocean to the European continent. The eastward passage of the storm centre is due to the prevailing eastward movement of the air in its upper part throughout that portion of the northern hemisphere. Commonly they incline somewhat to the northward of east in their journey. In all cases the winds appear to blow spirally into the common storm centre. There is the same doldrum area or calm field in the centre of the storm that we note between the trade winds and in the middle of a hurricane disk, though this area is less defined than in the other instances, and the forward motion of the storm at a considerable speed is in most cases characteristic of the disturbance. On the front of one of these storms in North America the winds commonly begin in the northeast, thence they veer by the east to the southwest. At this stage in the movement the storm centre has passed by, the rainfall commonly ceases, and cold, dry winds setting to the northwestward set in. This is caused by the fact that the ascending air, having attained a height above the earth, settles down behind the storm, forming an anticyclone or mass of dry air, which presses against the retreating side of the great whirlwind.

In front of the storm the warm and generally moist relatively warm air, pressing in toward the point of uprise and overlaid by the upper cold air, is brought into a condition where it tends to form small subordinate shafts up through which it whirls on the same principle, but with far greater intensity than the main ascending column. The reason for the violence of this movement is that the difference in temperature of the air next the surface and that at the height of a few thousand feet is great. As might be expected, these local spinnings are most apt to occur in the season when the air next the earth is relatively warm, and they are aptest to take place in the half of the advancing front lying between the east and south, for the reason that there the highest temperatures and the greatest humidity are likely to coexist. In that part of the field, during the time when the storm is advancing from the Rocky Mountains to the Atlantic, a dozen or more of these spinning uprushes may be produced, though few of them are likely to be of large size or of great intensity.

The secondary storms of cyclones, such as are above noted, receive the name of tornadoes. They are frequent and terrible visitations of the country from northern Texas, Florida, and Alabama to about the line of the Great Lakes; they are rarely developed in the region west of central Kansas, and only occasionally do they exhibit much energy in the region east of the plain-lands of the Ohio Valley. Although known in other lands, they nowhere, so far as our observations go, exhibit the paroxysmal intensity which they show in the central portion of the North American continent. There the air which they affect acquires a speed of movement and a fury of action unknown in any other atmospheric disturbances, even in those of the hurricanes.

The observer who has a chance to note from an advantageous position the development of a tornado observes that in a tolerably still air, or at least an air unaffected by violent winds—generally in what is termed a "sultry" state of the atmosphere—the storm clouds in the distance begin to form a kind of funnel-shaped dependence, which gradually extends until it appears to touch the earth. As the clouds are low, this downward-growing column probably in no case is observed for the height of more than three or four thousand feet. As the funnel descends, the clouds above and about it may be seen to take on a whirling movement around the centre, and under favourable circumstances an uprush of vapours may be noted in the centre of the swaying shaft. As the whirl comes nearer, the roar of the disturbance, which at a distance is often compared to the sound made by a threshing machine or to that of distant musketry, increases in loudness until it becomes overwhelming. When a storm such as this strikes a building, it is not only likely to be razed by the force of the wind, but it may be exploded, as by the action of gunpowder fired within its walls, through the sudden expansion of the air which it contains. In the centre of the column, although it rarely has a diameter of more than a few hundred feet, the uprush is so swift that it makes a partial vacuum. The air, striving to get into the space which it is eager to occupy, is whirling about at such a rate that the centrifugal motion which it thus acquires restrains its entrance. In this way there may be, as the column rapidly moves by, a difference of pressure amounting probably to what the mercury of a barometer would indicate by four or five inches of fall. Unless the structure is small and its walls strong, its roof and sides are apt to be blown apart by this difference of pressure and the consequent expansion of the contained air. In some cases where wooden buildings have withstood this curious action the outer clapboards have been blown off by the expansion of the small amount of air contained in the interspaces between that covering and the lath and plaster within (see Fig. 9).



The blow of the air due to its rotative whirling has in several cases proved sufficient to throw a heavy locomotive from the track of a well-constructed railway. In all cases where it is intense it will overturn the strongest trees. The ascending wind in the centre of the column may sometimes lift the bodies of men and of animals, as well as the branches and trunks of trees and the timber of houses, to the height of hundreds of feet above the surface. One of the most striking exhibitions of the upsucking action in a tornado is afforded by the effect which it produces when it crosses a small sheet of water. In certain cases where, in the Northwestern States of this country, the path of the storm lay over the pool, the whole of the water from a basin acres in extent has been entirely carried away, leaving the surface, as described by an observer, apparently dry enough to plough.

Fortunately for the interests of man, as well as those of the lower organic life, the paths of these storms, or at least the portion of their track where the violence of the air movement makes them very destructive, often does not exceed five hundred feet in width, and is rarely as great as half a mile in diameter. In most cases the length of the journey of an individual tornado does not exceed thirty miles. It rarely if ever amounts to twice that distance.

In every regard except their small size and their violence these tornadoes closely resemble hurricanes. There is the same broad disk of air next the surface spirally revolving toward the ascending centre, where its motion is rapidly changed from a horizontal to a vertical direction. The energy of the uprush in both cases is increased by the energy set free through the condensation of the water, which tends further to heat and thus to expand the air. The smaller size of the tornado may be accounted for by the fact that we have in their originating conditions a relatively thin layer of warm, moist air next the earth and a relatively very cold layer immediately overlying it. Thus the tension which serves to start the movement is intense, though the masses involved are not very great. The short life of a tornado may be explained by the fact that, though it apparently tends to grow in width and energy, the central spout is small, and is apt to be broken by the movements of the atmosphere, which in the front of a cyclone are in all cases irregular.

On the warmer seas, but often beyond the limits of the tropics, another class of spinning storms, known as waterspouts, may often be observed. In general appearance these air whirls resemble tornadoes, except that they are in all cases smaller than that group of whirlings. As in the tornadoes, the waterspout begins with a funnel, which descends from the sky to the surface of the sea. Up the tube vapours may be seen ascending at great speed, the whole appearing like a gigantic pillar of swiftly revolving smoke. When the whirl reaches the water, it is said that the fluid leaps up into the tube in the form of dense spray, an assertion which, in view of the fact of the action of a tornado on a lake as before described, may well be believed. Like the tornadoes and dust whirls, the life of a waterspout appears to be brief. They rarely endure for more than a few minutes, or journey over the sea for more than two or three miles before the column appears to be broken by some swaying of the atmosphere. As these peculiar storms are likely to damage ships, the old-fashioned sailors were accustomed to fire at them with cannon. It has been claimed that a shot would break the tube and end the little convulsion. This, in view of the fact that they appear to be easily broken up by relatively trifling air currents, may readily be believed. The danger which these disturbances bring to ships is probably not very serious.

The special atmospheric conditions which bring about the formation of waterspouts are not well known; they doubtless include, however, warm, moist air next the surface of the sea and cold air above. Just why these storms never attain greater size or endurance is not yet known. These disturbances have been seen for centuries, but as yet they have not been, in the scientific sense, observed. Their picturesqueness attracts all beholders; it is interesting to note the fact that perhaps the earliest description of their phenomena—one which takes account in the scientific spirit of all the features which they present—was written by the poet Camoens in the Lusiad, in which he strangely mingles fancy and observation in his account of the great voyage of Vasco da Gama. The poet even notes that the water which falls when the spout is broken is not salt, but fresh—a point which clearly proves that not much of the water which the tube contains is derived from the sea. It is, in fact, watery vapour drawn from the air next the surface of the ocean, and condensed in its ascent through the tube. In this and other descriptions of Nature Camoens shows more of the scientific spirit than any other poet of his time. He was in this regard the first of modern writers to combine a spiritual admiration for Nature with some sense of its scientific meaning.

In treating of the atmosphere, meteorologists base their studies largely on changes in the weight of that medium, which they determine by barometric observations. In fact, the science of the air had its beginning in Pascal's admirable observation on the changes in the height of a column of mercury contained in a bent tube as he ascended the volcanic peak known as Puy de Dome, in central France. As before noted, it is to the disturbances in the weight of the air, brought about mainly by variations in temperature, that we owe all its currents, and it is upon these winds that the features we term climate in largest measure depend. Every movement of the winds is not only brought about by changes in the relative weight of the air at certain points, but the winds themselves, owing to the momentum which the air attains by them, serve to bring about alterations in the quantity of air over different parts of the earth, which are marked most distinctly by barometric variations. These changes are exceedingly complicated; a full account of them would demand the space of this volume. A few of the facts, however, should be presented here. In the first place, we note that each day there is normally a range in the pressure which causes the barometer to be at the lowest at about four o'clock in the morning and four o'clock in the afternoon, and highest at about ten o'clock in those divisions of the day. This change is supposed to be due to the fact that the motes of dust in the atmosphere in the night, becoming cooled, condense the water vapour upon their surfaces, thus diminishing the volume of the air. When the sun rises the water evaporated by the heat returns from these little storehouses into the body of the atmosphere. Again in the evening the condensation sets in; at the same time the air tends to drift in from the region to the westward, where the sun is still high, toward the field where the barometer has been thus lowered; the current gradually attains a certain volume, and so brings about the rise of the barometer about ten o'clock at night.

In the winter time, particularly on the well-detached continent of North America, we find a prevailing high barometer in the interior of the country and a corresponding low state of pressure on the Atlantic Ocean. In the summer season these conditions are on the whole reversed.

Under the tropics, in the doldrum belt, there is a zone of low barometer connected to the ascending currents which take place along that line. This is a continuous manifestation of the same action which gives a large area of a disklike form in the centre or eye of the hurricane and in the middle portion of the tornado's whirl. In general, it may be said that the weight of the air is greatest in the regions from which it is blowing toward the points of upward escape, and least in and about those places where the superincumbent air is rising through a temporary or permanent line of escape. In other words, ascending air means generally a relatively low barometer, while descending air is accompanied by greater pressure in the field upon which it falls.

In almost every part of the earth which is affected by a particular physiography we find that the movements of the atmosphere next the surface are qualified by the condition which it encounters. In fact, if a person were possessed of all the knowledge which could be obtained concerning winds, he could probably determine as by a map the place where he might chance to find himself, provided he could extend his observations over a term of years. In other words, the regimen of the winds—at least those of a superficial nature—is almost as characteristic of the field over which they go as is a map of the country. Of these special winds a number of the more important have been noted, only a few of which we can advert to. First among these may well come the land and sea breezes which are remarked about all islands which are not continuously swept by permanent winds. One of the most characteristic instances of these alternate winds is perhaps that afforded on the island of Jamaica.

The island of Jamaica is so situated within the basin of the Caribbean that it does not feel the full influence of the trades. It has a range of high mountains through its middle part. In the daytime the surface of the land, which has the sun overhead twice each year, and is always exposed to nearly vertical radiation, becomes intensely hot, so that an upcurrent is formed. The formation of this current is favoured by the mountains, which apply a part of the heat at the height of about a mile above the surface of the sea. This action is parallel to that we notice when, in order to create a draught in the air of a chimney, we put a torch some distance up above the fireplace, thus diminishing the height of the column of air which has to be set in motion. It is further shown by the fact that when miners sought to make an upcurrent in a shaft, in order to lead pure air into the workings through other openings, they found after much experience that it was better to have the fire near the top of the shaft rather than at the bottom.

The ascending current being induced up the mountain sides of Jamaica, the air is forced in from the sea to the relatively free space. Before noon the current, aided in its speed by a certain amount of the condensation of the watery vapour before described, attains the proportions of a strong wind. As the sun begins to sink, the earth's surface pours forth its heat; the radiation being assisted by the extended surfaces of the plants, cooling rapidly takes place. Meanwhile the sea, because of the great heat-storing power of water, is very little cooled, the ascent of the air ceases, the temporary chimney with its updraught is replaced by a downward current, and the winds blow from the land until the sun comes again to reverse the current. In many cases these movements of the daily winds flowing into and from islands induce a certain precipitation of moisture in the form of rain. Generally, however, their effect is merely to ameliorate the heat by bringing alternately currents from the relatively cool sea and from the upper atmosphere to lessen the otherwise excessive temperature of the fields which they traverse.

Although characteristic sea and land winds are limited to regions where the sun's heat is great, they are traceable even in high latitudes during the periods of long-continued calm attended with clear skies. Thus on the island of Martha's Vineyard, in Massachusetts, the writer has noted, when the atmosphere was in such a state, distinct night and day, or sea and land, breezes coming in their regular alternation. During the night when these alternate winds prevail the central portion of the island, at the distance of three miles from the sea, is remarkably cold, the low temperature being due to the descending air current. To the same physical cause may be attributed the frequent insets of the sea winds toward midday along the continental shores of various countries. Thus along the coast of New England in the summer season a clear, still, hot day is certain to lead to the creation of an ingoing tide of air, which reaches some miles into the interior. This stream from the sea enters as a thin wedge, it often being possible to note next the shore when the movement begins a difference of ten degrees of temperature between the surface of the ground to which the point of the wedge has attained, and a position twenty feet higher in the air. This is a beautiful example to show at once how the relative weight of the atmosphere, even when the differences are slight, may bring about motion, and also how masses of the atmosphere may move by or through the rest of the medium in a way which we do not readily conceive from our observations on the transparent mass. Very few people have any idea how general is the truth that the air, even in continuous winds, tends to move in more or less individualized masses. This, however, is made very evident by watching the gusts of a storm or the wandering patches of wind which disturb the surface of an otherwise smooth sea.



Among the notable local winds are those which from their likeness to the Foehn of the Swiss valleys receive that name. Foehns are produced where a body of air blowing against the slope of a continuous mountain range is lifted to a considerable height, and, on passing over the crest, falls again to a low position. In its ascent the air is cooled, rarefied, and to a great extent deprived of its moisture. In descending it is recondensed, and by the process by which its atoms are brought together its latent heat is made sensible. There being but little watery vapour in the mass, this heat is not much called for by that heat-storing fluid, and so the air is warmed. So far Foehn winds have only been remarked as conspicuous features in Switzerland and on the eastern face of the Rocky Mountains. In the region about the head waters of the Missouri and to the northward their influence in what are called the Chinook winds is distinctly to ameliorate the severe winter climate of the country.

In almost all great desert regions, particularly in the typical Sahara, we find a variety of storm belonging to the whirlwind group, which, owing to the nature of the country, take on special characteristics. These desert storms take up from the verdureless earth great quantities of sand and other fine debris, which often so clouds the air as to bring the darkness of night at midday. Their whirlings appear in size to be greater than those which produce tornadoes or waterspouts, but less than hurricanes or cyclones. Little, however, is known about them. They have not been well observed by meteorologists. In some ways they are important, for the reason that they serve to carry the desert sand into regions previously verdure-clad, and thus to extend the bounds of the desolate fields in which they originate. Where they blow off to the seaward, they convey large quantities of dust into the ocean, and thus serve to wear down the surface of the land in regions where there are no rivers to effect that action in the normal way.

Notwithstanding its swift motion when impelled by differences in weight, the movements of the air have had but little direct and immediate influence on the surface of the earth. The greater part of the work which it does, as we shall see hereafter, is done through the waters which it impels and bears about. Yet where winds blow over verdureless surfaces the effect of the sand which they sweep before them is often considerable. In regions of arid mountains the winds often drive trains of sand through the valleys, where the sharp particles cut the rocks almost as effectively as torrents of water would, distributing the wearing over the width of the valley. The dust thus blown, from a desert region may, when it attains a country covered with vegetation, gradually accumulate on its surface, forming very thick deposits. Thus in northwestern China there is a wide area where dust accumulations blown from the arid districts of central Asia have gradually heaped up in the course of ages to the depth of thousands of feet, and this although much of the debris is continually being borne away by the action of the rain waters as they journey toward the sea. Such dust accumulations occur in other parts of the world, particularly in the districts about the upper Mississippi and in the valleys of the Rocky Mountains, but nowhere are they so conspicuous as in the region first mentioned.

Where prevailing winds from the sea, from great lakes, and even from considerable rivers, blow against sandy shores or cliffs of the same nature, large quantities of sand and dust are often driven inland from the coast line. In most cases these wind-borne materials take on the form of dunes, or heaps of sand, varying from a few feet to several hundred feet in height. It is characteristic of these hills of blown sand that they move across the face of the country. Under favourable conditions they may journey scores of miles from the shore. The marching of a dune is effected through the rolling up of the sand on the windward side of the elevation, when it is impelled by the current of air to the crest where it falls into the lee or shelter which the hill makes to the wind. In this way in the course of a day the centre of the dune, if the wind be blowing furiously, may advance a measurable distance from the place it occupied before. By fits and starts this ongoing may be indefinitely continued. A notable and picturesque instance of the march of a great dune may be had from the case in which one of them overwhelmed in the last century the village of Eccles in southeastern England. The advancing sand gradually crept into the hamlet, and in the course of a decade dispossessed the people by burying their houses. In time the summit of the church spire disappeared from view, and for many years thereafter all trace of the hamlet was lost. Of late years, however, the onward march of the sands has disclosed the church spire, and in the course of another century the place may be revealed on its original site, unchanged except that the marching hill will be on its other side.

In the region about the head of the Bay of Biscay the quantity of these marching sands is so great that at one time they jeopardized the agriculture of a large district. The French Government has now succeeded, by carefully planting the surface of the country with grasses and other herbs which will grow in such places, in checking the movement of the wind-blown materials. By so doing they have merely hastened the process by which Nature arrests the march of dunes. As these heaps creep away from the sea, they generally come into regions where a greater variety of plants flourish; moreover, their sand grains become decayed, so that they afford a better soil. Gradually the mat of vegetation binds them down, and in time covers them over so that only the expert eye can recognise their true nature. Only in desert regions can the march of these heaps be maintained for great distances.

Characteristic dunes occur from point to point all along the Atlantic coast from the State of Maine to the northern coast of Florida. They also occur along the coasts of our Great Lakes, being particularly well developed at the southern end of Lake Michigan, where they form, perhaps, the most notable accumulations within the limits of the United States.

When blown sands invade a forest and the deposit is rapidly accumulated, the trees are often buried in an undecayed condition. In this state, with certain chemical reactions which may take place in the mass, the woody matter is apt to become replaced by silex dissolved from the sand, which penetrates the tissues of the plants. In this way salicified forests are produced, such as are found in the region of the Rocky Mountains, where the trunks of the trees, now very hard stone, so perfectly preserve their original structure that when cut and polished they may be used for decorative purposes. Conspicuous as is this work of the dunes, it is in a geological way much less important than that accomplished by the finer dust which drifts from one region of land to another or into the sea. Because of their weight, the sand grains journey over the surface of the earth, except, indeed, where they are uplifted by whirl storms. They thus can not travel very fast or far. Dust, however, rises into the air, and journeys for indefinite distances. We thus see how slight differences in the weight of substances may profoundly affect the conditions of their deportation.

THE SYSTEM OF WATERS.

The envelope of air wraps the earth completely about, and, though varying in thickness, is everywhere present over its surface. That of the waters is much less equally distributed. Because of its weight, it is mainly gathered in the depths of the earth, where it lies in the interstices of the rocks and in the great realm of the seas. Only a very small portion of the fluid is in the atmosphere or on the land. Perhaps less than a ten thousandth part of the whole is at any one time on this round from the seas through the air to the land and back to the great reservoir.

The great water store of the earth is contained in two distinct realms—in the oceans, where the fluid is concentrated in a quantity which fills something like nine tenths of the hollows formed by the corrugations of the earth's surface; and in the rocks, where it is stored in a finely divided form, partly between the grains of the stony matter and partly in the substance of its crystals, where it exists in a combination, the precise nature of which is not well known, but is called water of crystallization. On the average, it seems likely that the materials of the earth, whether under the sea or on the land, have several per cent of their mass of the fluid.

It is not yet known to what depth the water-bearing section of the earth extends; but, as we shall see more particularly hereafter when we come to consider volcanoes, the lavas which they send up to the surface are full of contained water, which passes from them in the form of steam. The very high temperature of these volcanic ejections makes it necessary for us to suppose that they come from a great depth. It is difficult to believe that they originate at less than a hundred miles below the earth's surface. If, then, the rocks contain an average of even five per cent of water to the depth of one hundred miles, the quantity of the fluid stored within the earth is greater than that which is contained in the reservoir of the ocean. The oceans, on the average, are not more than three miles deep; spread evenly over the surface of the whole earth, their depth would be less than two miles, while the water in the rocks, if it could be added to the seas, would make the total depth seven miles or more. As we shall note hereafter, the processes of formation of strata tend to imprison water in the beds, which in time is returned to the earth's surface by the forces which operate within the crust.

Although the water in the seas is, as we have seen, probably less than one half of the store which the earth possesses, the part it plays in the economy of the planet is in the highest measure important. The underground water operates solely to promote certain changes which take place in the mineral realm. Its effect, except in volcanic processes, are brought about but slowly, and are limited in their action. The movements of this buried water are exceedingly gradual; the forces which impel it about and which bring it to do its work originate in the earth. In the seas the fluid has an exceeding freedom of motion; it can obey the varied impulses which the solar energy imposes upon it. The role of these wonderful actions which we are about to trace includes almost everything which goes on upon the surface of the planet—that which relates to the development of animal and vegetable life, as well as to the vast geological changes which the earth is undergoing.

If the surface of the earth were uniformly covered with water to the depth of ten thousand feet or more, every particle of fluid would, in a measure, obey the attraction of the sun, of the moon, and, theoretically, also of all the other bodies in space, on the principle that every particle of matter in the universe exercises a gravitative effect on every other. As it is, owing to the divided condition of the water on the earth's surface, only that which is in the ocean and larger seas exhibits any measurable influence from these distant attractions. In fact, only the tides produced by the moon and sun are of determinable magnitude, and of these the lunar is of greater importance, the reason being the near position of our satellite to our own sphere. The solar tide is four tenths as great as the lunar. The water doubtless obeys in a slight way the attraction of the other celestial bodies, but the motions thus imparted are too small to be discerned; they are lost in the great variety of influences which affect all the matter on the earth.

Although the tides are due to the attraction of the solar bodies, mainly to that of the moon, the mode in which the result is brought about is somewhat complicated. It may briefly and somewhat incompletely be stated as follows: Owing to the fact that the attracting power of the earth is about eighty times greater than that of the moon, the centre of gravity of the two bodies lies within the earth. About this centre the spheres revolve, each in a way swinging around the other. At this point there is no centrifugal motion arising from the revolution of the pair of spheres, but on the side of the earth opposite the moon, some six thousand miles away, the centrifugal force is considerable, becoming constantly greater as we pass away from the turning point. At the same time the attraction of the moon on the water becomes less. Thus the tide opposite the satellite is formed. On the side toward the moon the same centrifugal action operates, though less effectively than in the other case, for the reason that the turning point is nearer the surface; but this action is re-enforced by the greater attraction of the moon, due to the fact that the water is much nearer that body.

In the existing conditions of the earth, what we may call the normal run of the tides is greatly interrupted. Only in the southern ocean can the waters obey the lunar and solar attraction in anything like a normal way. In that part of the earth two sets of tides are discernible, the one and greater due to the moon, the other, much smaller, to the sun. As these tides travel round at different rates, the movements which they produce are sometimes added to each other and sometimes subtracted—that is, at times they come together, while again the elevation of one falls in the hollow of the other. Once again supposing the earth to be all ocean covered, computation shows that the tides in such a sea would be very broad waves, having, indeed, a diameter of half the earth's circumference. Those produced by the moon would have an altitude of about one foot, and those by the sun of about three inches. The geological effects of these swayings would be very slight; the water would pass over the bottom to and fro twice each day, with a maximum journey of a hundred or two feet each way from a fixed point. This movement would be so slow that it could not stir the fine sediment; its only influence would perhaps be to help feed the animals which were fixed upon the bottom by drawing the nurture-bringing water by their mouths.

Although the divided condition of the ocean perturbs the action of the tides, so that except by chance their waves are rarely with their centres where the attracting bodies tend to make them, the influence of these divisions is greatly to increase the geological or change-bringing influences arising from these movements. When from the southern ocean the tides start to the northward up the bays of the Atlantic, the Pacific, or the Indian Ocean, they have, as before noted, a height of perhaps less than two feet. As they pass up the narrowing spaces the waves become compressed—that is, an equal volume of moving water has less horizontal room for its passage, and is forced to rise higher. We see a tolerably good illustration of the same principle when we observe a wind-made wave enter a small recess of the shore, the sides of which converge in the direction of the motion. With the diminished room, the wave gains in height. It thus comes about that the tide throughout the Atlantic basin is much higher than in the southern ocean. On the same principle, when the tide rolls in against the shores every embayment of a distinct kind, whose sides converge toward the head, packs up the tidal wave, often increasing its height in a remarkable way. When these bays are wide-mouthed and of elongate triangular form, with deep bottoms, the tides which on their outer parts have a height of ten or fifteen feet may attain an altitude of forty or fifty feet at the apex of the triangle.

We have already noted the fact that the tide, such as runs in the southern ocean, exercises little or no influence upon the bottom of the sea over which it moves. As the height of the confined waters increases, the range of their journey over the bottom as the wave comes and goes rapidly increases. When they have an elevation of ten feet they can probably stir the finer mud on the ocean floor, and in shallow water move yet heavier particles. In the embayments of the land, where a great body of water journeys like an alternating river into extensive basins, the tidal action becomes intense; the current may be able to sweep along large stones quite as effectively as a mountain torrent. Thus near Eastport, Me., where the tides have a maximum rise and fall of over twenty feet, the waters rush in places so swiftly that at certain stages of the movement they are as much troubled as those at the rapids of the St. Lawrence. In such portions of the shore the tides do important work in carving channels into the lands.

Along the shores of the continents about the North Atlantic, where the tides act in a vigorous manner, we almost everywhere find an underwater shelf extending from the shore with a declivity of only five to ten feet to the mile toward the centre of the sea, until the depth of about five hundred feet is attained; from this point the bottom descends more steeply into the ocean's depth. It is probable that the larger part of the material composing these continental shelves has been brought to its position by tidal action. Each time the tidal wave sweeps in toward the shore it urges the finer particles of sediment along with it. When it moves out it drags them on the return journey toward the depths of the sea. If this shelf were perfectly horizontal, the two journeys of the sand and mud grains would be of the same length; but as the movement takes place up and down a slope, the bits will travel farther under the impulse which leads them downward than under that which impels them up. The result will be that the particles will travel a little farther out from the shore each time it is swung to and fro in the alternating movement of the tide.

The effect of tidal movement in nurturing marine life is very great. It aids the animals fixed on the bottoms of the deep seas to obtain their provision of food and their share of oxygen by drawing the water by their bodies. All regions which are visited by strong tides commonly have in the shallows near the shores a thick growth of seaweed which furnishes an ample provision of food for the fishes and other forms of animal life.

A peculiar effect arising from tidal action is believed by students of the phenomena to be found in the slowing of the earth's rotation on its axis. The tides rotate around the earth from east to west, or rather, we should say, the solid mass of the earth rubs against them as it spins from west to east. As they move over the bottom and as they strike against the shores this push of the great waves tends in a slight measure to use up the original spinning impulse which causes the earth's rotation. Computation shows that the amount of this action should be great enough gradually to lengthen the day, or the time occupied by the earth in making a complete revolution on the polar axis. The effect ought to be great enough to be measurable by astronomers in the course of a thousand years. On the other hand, the records of ancient eclipses appear pretty clearly to show that the length of the day has not changed by as much as a second in the course of three thousand years. This evidence does not require us to abandon the supposition that the tides tend to diminish the earth's rate of rotation. It is more likely that the effect of the reduction in the earth's diameter due to the loss of heat which is continually going on counterbalances the influence of the tidal friction. As the diameter of a rotating body diminishes, the tendency is for the mass to spin more rapidly; if it expands, to turn more slowly, provided in each case the amount of the impulse which leads to the turning remains the same. This can be directly observed by whirling a small weight attached to a string in such a manner that the cord winds around the finger with each revolution; it will be noted that as the line shortens the revolution is more quickly accomplished. We can readily conceive that the earth is made up of weights essentially like that used in the experiment, each being drawn toward the centre by the gravitative stress, which is like that applied to the weight by the cord.

The fact that the days remain of the same length through vast periods of time is probably due to this balance between the effects of tidal action and those arising from the loss of heat—in other words, we have here one of those delicate arrangements in the way of counterpoise which serve to maintain the balanced conditions of the earth's surface amid the great conflicts of diverse energies which are at work in and upon the sphere.

It should be understood that the effects of the attraction which produces tides are much more extensive than they are seen to be in the movements of the sea. So long as the solar and planetary spheres remain fluid, the whole of their masses partake of the movement. It is a consequence of this action, as the computations of Prof. George Darwin has shown, that the moon, once nearer the earth than it is at present, has by a curious action of the tidal force been pushed away from the centre of our sphere, or rather the two bodies have repelled each other. An American student of the problem, Mr. T.J.J. See, has shown that the same action has served to give to the double stars the exceeding eccentricity of their orbits.

Although these recent studies of tidal action in the celestial sphere are of the utmost importance to the theory of the universe, for they may lead to changes in the nebular hypotheses, they are as yet too incomplete and are, moreover, too mathematical to be presented in an elementary treatise such as this.

* * * * *

We now turn to another class of waves which are of even more importance than those of the tides—to the undulations which are produced by the action of the wind on the surface of the water. While the tide waves are limited to the open ocean, and to the seas and bays which afford them free entrance, wind waves are produced everywhere where water is subjected to the friction of air which flows over it. While tidal waves come upon the shores but twice each day, the wind waves of ordinary size which roll in from the ocean deliver their blows at intervals of from three to ten seconds. Although the tidal waves sometimes, by a packing-up process, attain the height of fifty feet, their average altitude where they come in contact with the shore probably does not much exceed four feet; usually they come in gently. It is likely that in a general way the ocean surges which beat against the coast are of greater altitude.

Wind waves are produced and perform their work in a manner which we shall now describe. When the air blows over any resisting surface, it tends, in a way which we can hardly afford here to describe, to produce motions. If the particle is free to move under the impulse which it communicates, it bears it along; if it is linked together in the manner of large masses, which the wind can not transport, it tends to set it in motion in an alternating way. The sounds of our musical instruments which act by wind are due to these alternating vibrations, such as all air currents tend to produce. An AEolian harp illustrates the action which we are considering. Moving over matter which has the qualities that we denote by the term fluid, the swayings which the air produces are of a peculiar sort, though they much resemble those of the fiddle string. The surface of the liquid rises and falls in what we term waves, the size of which is determined by the measure of fluidity, and by the energy of the wind. Thus, because fresh water is considerably lighter than salt, a given wind will produce in a given distance for the run of the waves heavier surges in a lake than it will in the sea. For this reason the surges in a great storm which roll on the ocean shore, because of the wide water over which they have gathered their impetus, are in size very much greater than those of the largest lakes, which do not afford room for the development of great undulations.

To the eye, a wave in the water appears to indicate that the fluid is borne on before the wind. Examination, however, shows that the amount of motion in the direction in which the wind is blowing is very slight. We may say, indeed, that the essential feature of a wave is found in the transmission of impulse rather than in the movement of the fluid matter. A strip of carpet when shaken sends through its length undulations which are almost exactly like water waves. If we imagine ourselves placed in a particle of water, moving in the swayings of a wave in the open and deep sea, we may conceive ourselves carried around in an ellipse, in each revolution returning through nearly the same orbit. Now and then, when the particle came to the surface, it would experience the slight drift which the continual friction of the wind imposes on the water. If the wave in which the journey was made lay in the trade winds, where the long-continued, steadfast blowing had set the water in motion to great depths, the orbit traversed would be moving forward with some rapidity; where also the wind was strong enough to blow the tops of the waves over, forming white-caps, the advance of the particle very near the surface would be speedy. Notwithstanding these corrections, waves are to be regarded each as a store of energy, urging the water to sway much in the manner of a carpet strip, and by the swaying conveying the energy in the direction of the wave movement.

The rate of movement of wind waves increases with their height. Slight undulations go forward at the rate of less than half a mile an hour. The greater surges of the deeps when swept by the strongest winds move with the speed which, though not accurately determined, has been estimated by the present writer as exceeding forty miles an hour. As these surges often have a length transverse to the wind of a mile or more, a width of about an eighth of a mile, and a height of from thirty-five to forty-five feet, the amount of energy which they transmit is very great. If it could be effectively applied to the shores in the manner in which the energy of exploding gunpowder is applied by cannon shot, it is doubtful whether the lands could have maintained their position against the assaults of the sea. But there are reasons stated below why the ocean waves can use only a very small part of their energy in rending the rocks against which they strike on the coast line.

In the first place, we should note that wind waves have very little influence on the bottom of the deep sea. If an observer could stand on the sea floor at the depth of a mile below a point over which the greatest waves were rolling, he could not with his unaided senses discern that the water was troubled. He would, indeed, require instruments of some delicacy to find out that it moved at all. Making the same observations at the depth of a thousand feet, it is possible that he would note a slight swaying motion in the water, enough sensibly to affect his body. At five hundred feet in depth the movement would probably be sufficient to disturb fine mud. At two hundred feet, the rasping of the surge on the bottom would doubtless be sufficient to push particles of coarse sand to and fro. At one hundred feet in depth, the passage of the surge would be strong enough to urge considerable pebbles before it. Thence up the slope the driving action would become more and more intense until we attained the point where the wave broke. It should furthermore be noted that, while the movement of the water on the floor of the deep sea as the wave passes overhead would be to and fro, with every advance in the shallowing and consequent increased friction on the bottom, the forward element in the movement would rapidly increase. Near the coast line the effect of the waves is continually to shove the detritus up the slopes of the continental shelf. Here we should note the fact that on this shelf the waves play a part exactly the opposite of that effected by the tides. The tides, as we have noted, tend to drag the particles down the slope, while the waves operate to roll them up the declivity.

As the wave in advancing toward the shore ordinarily comes into continually shallowing water, the friction on the bottom is ever-increasing, and serves to diminish the energy the surge contains, and therefore to reduce its proportions. If this action operated alone, the subtraction which the friction makes would cause the surf waves which roll in over a continental shelf to be very low, probably in height less than half that which they now attain. In fact, however, there is an influence at work to increase the height of the waves at the expense of its width. Noting that the friction rapidly increases with the shallowing, it is easy to see that this resistance is greatest on the advancing front of the wave, and least on its seaward side. The result is that the front moves more slowly than the rear, so that the wave is forced to gain in height; but for the fact that the total friction which the wave encounters takes away most of its impetus, we might have combers a hundred feet high rolling upon the shelving shores which almost everywhere face the seas.

As the wave shortens its width and gains in relative height, though not in actual elevation, another action is introduced which has momentous consequences. The water in the bottom of the wave is greatly retarded in its ongoing by its friction over the sea floor, while the upper part of the surge is much less affected in this way. The result is that at a certain point in the advance, the place of which is determined by the depth, the size, and the speed of the undulation, the front swiftly steepens until it is vertical, and the top shoots forward to a point where it is no longer supported by underlying water, when it plunges down in what is called the surf or breaker. In this part of the wave's work the application of the energy which it transmits differs strikingly from the work previously done. Before the wave breaks, the only geological task which it accomplishes is effected by forcing materials up the slope, in which movement they are slightly ground over each other until they come within the battering zone of the shore, where they may be further divided by the action of the mill which is there in operation.

When the wave breaks on the shore it operates in the following manner: First, the overturning of its crest sends a great mass of water, it may be from the height of ten or more feet, down upon the shore. Thus falling water has not only the force due to its drop from the summit of the wave, but it has a share of the impulse due to the velocity with which the surge moved against the shore. It acts, in a word, like a hammer swung down by a strong arm, where the blow represents not only the force with which the weight would fall of itself, but the impelling power of the man's muscles. Any one who will expose his body to this blow of the surf will recognise how violent it is; he may, if the beach be pebbly, note how it drives the stones about; fragments the size of a man's head may be hurled by the stroke to the distance of twenty feet or more; those as large as the fist may be thrown clear beyond the limits of the wave. So vigorous is this stroke that the sound of it may sometimes be heard ten miles inland from the coast where it is delivered.

Moving forward up the slope of a gently inclined beach, the fragments of the wave are likely to be of sufficient volume to permit them to regather into a secondary surge, which, like the first, though much smaller, again rises into a wall, forming another breaker. Under favourable conditions as many as four or five of these successive diminishing surf lines may be seen. The present writer has seen in certain cases as many as a dozen in the great procession, the lowest and innermost only a few inches high, the outer of all with a height of perhaps twenty feet.

Along with the direct bearing action of the surf goes a to-and-fro movement, due to the rushing up and down of the water on the beach. This swashing affects not only the broken part of the waves, but all the water between the outer breaker and the shore. These swayings in the surf belt often swing the debris on the inner margin over a range of a hundred feet or more, the movement taking place with great swiftness, affecting the pebbles to the depth of several inches, and grinding the bits together in a violent way. Listening to the turmoil of a storm, we can on a pebbly beach distinctly hear the sound of the downward stroke, a crashing tone, and the roar of the rolling stones.

As waves are among the interesting things in the world, partly on account of their living quality and partly because of their immediate and often exceeding interest to man, we may here note one or two peculiar features in their action. In the first place, as the reader who has gained a sense of the changes in form of action may readily perceive, the beating of waves on the shore converts the energy which they possess into heat. This probably warms the water during great storms, so that by the hand we may note the difference in temperature next the coast line and in the open waters. This relative warmth of the surf water is perhaps a matter of some importance in limiting the development of ice along the shore line; it may also favour the protection of the coast life against the severe cold of the winter season.

The waves which successively come against the shore in any given time, particularly if a violent wind is blowing on to the coast, are usually of about the same size. When, however, in times of calm an old sea, as it is called, is rolling in, the surges may occasionally undergo very great variations in magnitude. Not infrequently these occasional waves are great enough to overwhelm persons who are upon the rocks next the shore. These greater surges are probably to be accounted for by the fact that in the open sea waves produced by winds blowing in different directions may run on with their diverse courses and varied intervals until they come near the shore. Running in together, it very well happens that two of the surges belonging to different sets may combine their forces, thus doubling the swell. The danger which these conjoined waves bring is obviously greatest on cliff shores, where, on account of the depth of water, the waves do not break until they strike the steep.

* * * * *

Having considered in a general way the action of waves as they roll in to the shore, bearing with them the solar energy which was contributed to them by the winds, we shall now take up in some detail the work which goes on along the coast line—work which is mainly accomplished by wave action.

On most coast lines the observer readily notes that the shore is divided into two different kinds of faces—those where the inner margin of the wave-swept belt comes against rocky steeps, and those bordered by a strand altogether composed of materials which the surges have thrown up. These may be termed for convenience cliff shores and wall-beach shores. We shall begin our inquiry with cliff shores, for in those sections of the coast line the sea is doing its most characteristic and important work of assaulting the land. If the student has an opportunity to approach a set of cliffs of hard rock in time of heavy storm, when the waves have somewhere their maximum height, he should seek some headland which may offer him safe foothold whence he can behold the movements which are taking place. If he is so fortunate as to have in view, as well may be the case, cliffs which extend down into deep water, and others which are bordered by rude and generally steeply sloping beaches covered with large stones, he may perceive that the waves come in against the cliffs which plunge into deep water without taking on the breaker form. In this case the undulation strikes but a moderate blow; the wave is not greatly broken. The part next the rock may shoot up as a thin sheet to a considerable height; it is evident that while the ongoing wave applies a good deal of pressure to the steep, it does not deliver its energy in the effective form of a blow as when the wave overturns, or in the consequent rush of the water up a beach slope. It is easy to perceive that firm-set rock cliffs, with no beaches at their bases, can almost indefinitely withstand the assaults. On the steep and stony beach, because of its relatively great declivity, the breaker or surf forms far in, and even in its first plunge often attains the base of the precipice. The blow of the overfalling as well as that of the inrush moves about stones of great size; those three feet or more in diameter are often hurled by the action against the base of the steep, striking blows, the sharp note of which can often be heard above the general roar which the commotion produces. The needlelike crags forming isles standing at a distance from the shore, such as are often found along hard rock coasts, are singularly protected from the action of effective waves. The surges which strike against them are unarmed with stones, and the water at their bases is so deep that it does not sway with the motion with sufficient energy to move them on the bottom. Where a cliff is in this condition, it may endure until an elevation of the coast line brings its base near the level of the sea, or until the process of decay has detached a sufficient quantity of stone to form a talus or inclined plane reaching near to the water level.

As before noted, it is the presence of a sloping beach reaching to about the base of the cliff which makes it possible for the waves to strike at with a hammer instead of with a soft hand. Battering at the base of the cliff, the surges cut a crease along the strip on which they strike, which gradually enters so far that the overhanging rock falls of its own weight. The fragments thus delivered to the sea are in turn broken up and used as battering instruments until they are worn to pieces. We may note that in a few months of heavy weather the stones of such a fall have all been reduced to rudely spherical forms. Observations made on the eastern face of Cape Ann, Mass., where the seas are only moderately heavy, show that the storms of a single winter reduce the fragments thrown into the sea from the granite quarries to spheroidal shapes, more than half of their weight commonly being removed in the form of sand and small pebbles which have been worn from their surfaces.

We can best perceive the effect of battering action which the sea applies to the cliffs by noting the points where, owing to some chance features in the structure in the rock, it has proved most effective. Where a joint or a dike, or perhaps a softer layer, if the rocks be bedded, causes the wear to go on more rapidly, the waves soon excavate a recess in which the pebbles are retained, except in stormy weather, in an unmoved condition. When the surges are heavy, these stones are kept in continuous motion, receding as the wave goes back, and rushing forward with its impulse until they strike against the firm-set rock at the end of the chasm. In this way they may drive in a cut having the length of a hundred feet or more from the face of the precipice. In most cases the roofs over these sea caves fall in, so that the structure is known as a chasm. Occasionally these roofs remain, in which case, for the reason that the floor of the cutting inclines upward, an opening is made to the surface at their upper end, forming what is called in New England a "spouting horn"; from the inland end of the tunnel the spray may be thrown far into the air. As long as the cave is closed at this inner end, and is not so high but that it may be buried beneath a heavy wave, the inrushing water compresses the air in the rear parts of the opening. When the wave begins to retreat this air blows out, sending a gust of spray before it, the action resembling the discharge of a great gun from the face of a fortification. It often happens that two chasms converging separate a rock from the cliff. Then a lowering of the coast may bring the mass to the state of a columnar island, such as abound in the Hebrides and along various other shores.

If a cliff shore retreats rapidly, it may be driven back into the shore, and its face assumes the curve of a small bay. With every step in this change the bottom is sure to become shallower, so that the waves lose more and more of their energy in friction over the bottom. Moreover, in entering a bay the friction which the waves encounter in running along the sides is greater than that which they meet in coming in upon a headland or a straight shore. The result is, with the inward retreat of the steep it enters on conditions which diminish the effectiveness of the wave stroke. The embayment also is apt to hold detritus, and so forms in time a beach at the foot of the cliff, over which the waves rarely are able to mount with such energy as will enable them to strike the wall in an effective manner. With this sketch of the conditions of a cliff shore, we will now consider the fate of the broken-tip rock which the waves have produced on that section of the coast land.

By observation of sea-beaten cliffs the student readily perceives that a great amount of rocky matter has been removed from most cliff-faced shores. Not uncommonly it can be shown that such sea faces have retreated for several miles. The question now arises, What becomes of the matter which has been broken up by the wave action? In some part the rock, when pulverized by the pounding to which it is subjected, has dissolved in the water. Probably ninety per cent of it, however, retains the visible state, and has a fate determined by the size of the fragments of which it is composed. If these be as fine as mud, so that they may float in the water, they are readily borne away by the currents which are always created along a storm-swept shore, particularly by the undertow or bottom outcurrent—the "sea-puss," as it is sometimes called—that sweeps along the bottom from every shore, against which the waves form a surf. If as coarse as sand grains, or even very small pebbles, they are likely to be drawn out, rolling over the bottom to an indefinite distance from the sea margin. The coarser stones, however, either remain at the foot of the cliff until they are beaten to pieces, or are driven along the shore until they find some embayment into which they enter. The journey of such fragments may, when the wind strikes obliquely to the shore, continue for many miles; the waves, running with the wind, drive the fragments in oscillating journeys up and down the beach, sometimes at the rate of a mile or more a day. The effect of this action can often be seen where a vessel loaded with brick or coal is wrecked on the coast. In a month fragments of the materials may be stretched along for the distance of many miles on either side of the point where the cargo came ashore. Entering an embayment deep enough to restrain their further journey, the fragments of rock form a boulder beach, where the bits roll to and fro whenever they are struck by heavy surges. The greater portion of them remain in this mill until they are ground to the state of sand and mud. Now and then one of the fragments is tossed up beyond the reach of the waves, and is contributed to the wall of the beach. In very heavy storms these pebbles which are thrown inland may amount in weight to many tons for each mile of shore.

The study of a pebbly beach, drawn from crest to the deep water outside, will give an idea as to the history of its work. On either horn of the crescent by which the pebbles are imported into the pocket we find the largest fragments. If the shore of the bay be long, the innermost part of the recess may show even only very small pebbles, or perhaps only fine sand, the coarser material having been worn out in the journey. On the bottom of the bay, near low tide, we begin to find some sand produced by the grinding action. Yet farther out, below high-tide mark, there is commonly a layer of mud which represents the finer products of the mill.

Boulder beaches are so quick in answering to every slight change in the conditions which affect them that they seem almost alive. If by any chance the supply of detritus is increased, they fill in between the horns, diminish the incurve of the bay, and so cause its beach to be more exposed to heavy waves. If, on the other hand, the supply of grist to the mill is diminished, the beach becomes more deeply incurved, and the wave action is proportionately reduced. We may say, in general, that the curve of these beaches represents a balance between the consumption and supply of the pebbles which they grind up. The supply of pebbles brought along the shore by the waves is in many cases greatly added to by a curious action of seaweeds. If the bottom of the water off the coast is covered by these fragments, as is the case along many coast lines within the old glaciated districts, the spores of algae are prone to take root upon them. Fastening themselves in those positions, and growing upward, the seaweeds may attain considerable size. Being provided with floats, the plant exercises a certain lifting power on the stone, and finally the tugging action of the waves on the fronds may detach the fragments from the bottom, making them free to journey toward the shore. Observing from near at hand the straight wall of the wave in times of heavy storm, the present writer has seen in one view as many as a dozen of these plant-borne stones, sometimes six inches in diameter, hanging in the walls of water as it was about to topple over. As soon as they strike the wave-beaten part of the shore these stones are apt to become separated from the plants, though we can often notice the remains or prints of the attachments adhering to the surface of the rock. Where the pebbles off the shore are plenty, a rocky beach may be produced by this process of importation through the agency of seaweeds without any supply being brought by the waves along the coast line.