sulphate of lime, of which the Thames is computed to carry annually past London not less than 180,000 tons. The total quantity of carbonate of lime, removed from the limestones of its basin by this river in a year, amounts, on an average, to 140 tons from every square mile, which is estimated to be equal to the lowering of the general surface to the extent of 10 of an inch from each square mile in a century, or one foot in 13,200 years.

9

Mechanical Action (1) Transport.-The dissolved material forms but a small proportion of the total amount of mineral substances conveyed by rivers from land to sea. A single shower of rain washes off fine dust and soil from the surface of the ground into the nearest brook which then rolls along with a discoloured current. An increase in the volume of the water enables a stream to sweep along sand, gravel, and blocks of stone lying in its channel, and to keep these materials moving until, as the declivity lessens and the rain ceases, the current becomes too feeble to do more than lazily carry onward the fine silt that discolours it. Every stream, large or small, is ceaselessly busy transporting mud, sand, or gravel. And as the ultimate destination of all this sediment is the bottom of the sea, it is evident that if there be no compensating influences at work to repair the constant loss, the land must in the end be all worn away.

Some of the most instructive lessons regarding the work of running water on land are afforded by the beds of mountaintorrents. Huge blocks, detached from the crags and cliffs on either side, may there be seen cumbering the pathway of the water, which seems quite powerless to move such masses and can only sweep round them or find a passage beneath them. But visit such a torrent when it is swollen with heavy rains or rapidly melted snow, and you will hear the stones knocking against each other or on the rocky bottom, as they are driven downwards by the flood. Or when the stream is at its lowest, in dry summer weather, follow its course a little way down hill, and you will see that by degrees the blocks, losing their sharp edges, have become rounded boulders, and that these are gradually replaced by coarse shingle and then by finer gravel. In the quieter reaches of the water, sheets of sand begin to make their appearance, and at last when the stream reaches the plains, no sediment of coarser grain than mere silt may be seen in its channel. It is thus obvious that in the constant transport maintained by watercourses, the carried materials, by being rolled along rocky channels and continually ground against each other, diminish in size as they descend. A

river flowing from a range of mountains to the distant ocean may be likened to a mill, into which large angular masses of rock are cast at the upper end, and out of which only fine sand and silt are discharged at the lower.

Partly, therefore, owing to the fine dust and soil swept into them by wind and rain from the slowly decomposing surface of the land, and partly to the friction of the detritus which they sweep along their channels, rivers always contain more or less mineral matter suspended in their water or travelling with the current on the bottom. The amount of material thus transported varies greatly in different rivers, and at successive seasons even in the same river. In some cases, the rainfall is spread so equably through the year that the rivers flow onward with a quiet monotony, never rising much above nor sinking much below their average level. In such circumstances, the amount of sediment they carry downward is proportionately small. On the other hand, where either from heavy periodical rains or from rapid melting of snow, rivers are liable to floods, they acquire an enormously increased power of transport, and their burden of sediment is proportionately augmented. In a few days or weeks of high water, they may convey to the sea a hundredfold the amount of mineral matter which they could carry in a whole year of their quieter mood.

Measurements have been made of the proportions of sediment in the waters of different rivers at various seasons of the year. The results, as might be expected, show great variations. Thus the Garonne, rising among the higher peaks of the Pyrenees, drains a large area of the south of France, and is subject to floods by which an enormous quantity of sediment is swept down from the mountains to the plains. Its proportion of mud has been estimated to be as much as I part in 100 parts of water. The Durance, which takes its source high on the western flank of the Cottian Alps, is one of the rapidest and muddiest rivers in Europe. Its angle of slope varies from 1 in 208 to 1 in 467, the average declivity of the great rivers of the globe being probably not more than 1 in 2600, while that of a navigable stream ought not to exceed 10 inches per mile or I in 6336. The Durance is, therefore, rather a torrent than a river. With this rapidity of descent is conjoined an excessive capacity for transporting sediment. In floods of exceptional severity, the proportion of mud in the stream has been estimated at one-tenth by weight of the water, while the average proportion for nine years from 1867 to 1875 was about 이 Probably the best general average is to be obtained from a

river which drains a wide region exhibiting considerable diversities of climate, topography, rocks, and soils. The Mississippi presents a good illustration of these diversities, and has accordingly been taken as a kind of typical river, furnishing, so to speak, a standard by which the operations of other rivers may be compared, and which may perhaps be assumed as a fair average for all the rivers of the globe. Numerous measurements have been made of the proportion of sediment carried into the Gulf of Mexico by this vast river, with the result of showing that the average amount of sediment is by weight 1 part in every 1500 parts of water, or little more than one-third of the proportion in the water of the Durance.

If now we assume that, all over the world, the general average proportion of sediment floating in the water of rivers is I part in every 1500 of water, we can readily understand how seriously in the course of time must the land be lowered by the constant removal of so much decomposed rock from its surface. Knowing the area of the basin drained by a river, and also the proportion of sediment in its water, we can easily calculate the general loss from the surface of the basin. The ratio of the weight or "specific gravity" of the silt to that of solid rock may be taken to be as 19 is to 25. Accordingly the Mississippi conveys annually from its drainage basin an amount of sediment equivalent to the removal of part of a foot of rock from the general surface of the basin. At this rate, one foot of rock will be worn away every

6000

6000 years. If we take the general height of the land of the whole globe to be 2120 feet, and suppose it to be continuously wasted at the same rate at which the Mississippi basin is now suffering, then the whole dry land would be carried into the sea in 12,720,000 years. Or if we assume the mean height of Europe to be 973 feet and that this continent is degraded at the Mississippi rate of waste until the last vestige of it disappears, the process of destruction would be completed in rather less than 6,000,000 years. Such estimates are not intended to be close approximations to the truth. As the land is lowered, the rate of decay will gradually diminish, so that the later stages of decay will be enormously protracted. But by taking the rate of operation now ascertained to be in progress in such a river basin as the Mississippi, we obtain a valuable standard of comparison, and learn that the degradation of the land is much greater and more rapid than might have been supposed.

(2) Erosion. But rivers are not merely carriers of the mud, sand, and gravel swept into their channels by other agencies.

By

keeping these materials in motion, the currents reduce them in size, and at the same time employ them to hollow out the channels wherein they move. The mutual friction that grinds down large blocks of rock into sand and mud, tells also upon the rocky beds along which the material is driven. The most solid rocks are worn down; deep long gorges are dug out, and the watercourses, when they have once chosen their sites, remain on them and sink gradually deeper and deeper beneath the general level of the country. The surfaces of stone exposed to this attrition assume the familiar smoothed and rounded appearance which is known as water-worn. The loose stones lying in the channel of a stream, and the solid rocks as high up as floods can scour them, present this characteristic aspect. Here and there, where a few stones have been caught in an eddy of the current, and are kept in constant gyration, they reduce each other in dimensions, and at the same time grind out a hollow in the underlying rock. The sand and mud produced by the friction are swept off by the current, and the stones when sufficiently reduced in size are also carried away. But their places are eventually taken by other blocks brought down by floods, so that the supply of grinding material is kept up until the original hollow is enlarged into a wide deep caldron, at the bottom of which the stones can only be stirred by the heaviest floods. Cavities of this kind, known as pot-holes, are of frequent occurrence in rocky watercourses as well as on rocky shores, in short, wherever eddies of water can keep shingle rotating upon solid rock. As they often coalesce by the wearing away of the intervening wall of rock, they greatly aid in the deepening of a watercourse. In most rocky gorges, a succession of old pot-holes may be traced far above the present level of the stream (Fig. 8).

That it is by means of the gravel and other detritus pushed along the bottom by the current, rather than by the mere friction of the water on its bed, that a river excavates its channel, is most strikingly shown immediately below a lake. In traversing a lake, the tributary streams deposit their sediment on its bottom, because the still water checks their current and, by depriving the water of its more rapid movement, compels it to drop its burden of gravel, sand, and silt (see p. 42). Filtered in this way, the various streams united in the lake escape at its lower end as a clear transparent river. The Rhone, for instance, flows into the Lake of Geneva as a turbid stream; it issues from that great reservoir at Geneva as a rushing current of the bluest, most translucent water

which, though it sweeps over ledges of rock, has not yet been able to grind them down into a deep gorge. The Niagara, also, filtered by Lake Erie, has not acquired sediment enough to enable it to cut deeply into the rocks over which it foams in its rapids before throwing itself over the great Falls.

One of the most characteristic features of streams is the singularly sinuous courses they follow. As a rule, too, the flatter

the ground over which they flow, the more do they wind. Not uncommonly they form loops, the nearest bends of which in the end unite, and as the current passes along the now straightened channel, the old one is left to become by degrees a lake or pond of stagnant water, then a marsh, and lastly, dry ground. We might suppose that in flowing off the land, water would take the shortest and most direct road to the sea. But this is far from being the case. The slightest inequalities of level have originally determined sinuosities of the channels, while trifling differences in the hardness of the banks, in the accumulation of sediment, and