not inconsiderable removal of fine soil from the surface. As the upper layers of soil are removed, roots and percolating water are enabled to reach down farther into the solid rock which is broken up into subsoil, and thus the general surface of the land is insensibly lowered.

Besides accumulating in situ as subsoil and soil, the debris of decomposed rock forms talus-slopes and screes at the foot of crags, and a layer of rain-wash or brick-earth over gentler slopes. Where the action of wind comes markedly into play, tracts of sand-dunes may be piled up along the borders of the sea and of lakes, or in the arid interior of continents; and wide regions are in course of time buried under the fine dust which is sometimes so thick in the air as to obscure the noonday sun. But in none of these forms can the accumulation of decomposed material be regarded as permanent. So long as it is exposed to the influences of the atmosphere, this material is liable to be swept away from the surface of the land and borne outwards into the sea.

CHAPTER III.

THE INFLUENCE OF RUNNING WATER IN GEOLOGICAL CHANGES, AND HOW IT IS RECORDED.

It appears, then, that from various causes all over the globe, there is a continual decay of the surface of the land; that the decomposed material partly accumulates as soil, subsoil, and sheets or heaps of loose earth or sand, but that much of it is washed off the land by rain or blown into the rivers or into the sea by wind. We have now to consider the part taken by running water in this transport. From the single rain-drop up to the mighty river, every portion of the water that flows over the land is busy with its own share of the work. When we reflect on the amount of rain that falls annually over the land, and on the number of streams, large and small, that are ceaselessly at work, we realise how difficult it must be to form any fit notion of the entire amount of change which, even in a single year, these agents work upon the surface of the earth.

The influence of rain in the decay of the surface of the land was briefly alluded to in the last chapter. As soon as a drop of rain reaches the ground, it begins its appointed geological task, dissolving what it can carry off in solution, and pushing forward and downward whatever it has power to move. As the rain-drops gather into runnels, the same duty, but on a larger scale, is performed by them; and as the runnels unite into large streams, and these into yet mightier rivers, the operations, though becoming colossal in magnitude, remain essentially the same in kind. In the operations of the nearest brook, we see before us in miniature a sample of the changes produced by the thousands of rivers which, in all quarters of the globe, are flowing from the mountains to the sea. Watching these operations from day to day, we discover that they may all be classed under two heads. In the first place, the brook hollows out the channel in which it flows and thus aids in the general waste of the surface of the land; and in the second place, it carries away fine silt and other material resulting from that waste. Rivers are thus at once agents that themselves directly degrade the land, and that sweep the loosened detritus towards the ocean. An acquaintance with each of these kinds of work is needful to enable us to understand the nature of the records which river-action leaves behind it.

Chemical Action of Running Water. We have seen that rain in its descent from the clouds absorbs air, and that with the oxygen and carbonic acid which it thus obtains it proceeds to corrode the surfaces of rock on which it falls. When it reaches the ground and absorbs the acids termed "humous," which are supplied by the decomposing vegetation of the soil, it acquires increased power of eating into the stones over which it flows. When it rolls along as a runnel, brook, or river, it no doubt still attacks the rocks of its channel, though its action in this respect is not so easily detected. In some circumstances, however, the solvent influence of river-water upon solid rocks is strikingly displayed. Where the water contains a large proportion of the acids of the soil, and flows over a kind of rock specially liable to be eaten away by these acids, the most favourable conditions are presented for observing the change. Thus, a stream which issues from a peat-bog is usually dark brown in colour, from the vegetable solutions which it extracts from the moss. Among these solutions are some of the organic acids referred to, ready to eat into the surface of the rocks or loose stones which the stream may encounter in its descent. No kind of rock is more liable than limestone

FIG. 7. Erosion of limestone by the solvent action of a peaty stream. Durness, Sutherlandshire.

to corrosion under such circumstances. Peaty water flowing over it eats it away with comparative rapidity, while those portions of the rock that rise above the stream escape solution, except in so far as they are attacked by rain. Hence arise some curious features in the scenery of the watercourse. The walls of limestone above the water, being attacked only by the atmosphere, are not eaten away so fast as their base, over which the stream flows. They are con

D

sequently undermined, and are sometimes cut into dark tunnels and passages (Fig. 7). Even where the solvent action of the water of rivers is otherwise inappreciable, it can be detected by means of chemical analysis. Thus rivers, partly by the action of their water upon the loose stones and solid rocks of their channels, and partly by the contributions they receive from springs (which will be afterwards described), convey a vast amount of dissolved material into the sea. The substance thus invisibly transported consists of various mineral salts. One of the most abundant of these-carbonate of lime-is the substance that forms limestone, and furnishes the mineral matter required for the hard parts of a large proportion of the lower animals. It is a matter of some interest to know that this substance, so indispensable for the formation of the shells of so large a number of seacreatures, is constantly supplied to the sea by the streams that flow into it. The rivers of Western Europe, for instance, have been ascertained to convey about 1 part of dissolved mineral matter in every 5000 parts of water, and of this mineral matter about a half consists of carbonate of lime. It has been estimated that the Rhine bears enough carbonate of lime into the sea every year to make three hundred and thirty-two thousand millions of oysters of the usual size. Another abundant ingredient of river-water is gypsum or 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 To of an inch from each square mile in a century, or one foot in 13,200 years.