First we see the rounded boulders and pebbles of all sizes which must have been rolled about for a long time to make them so smooth. Some of them are so very hard that we cannot even scratch them with our knives; others are soft and easily broken. What would be the effect of rolling together stones of such varying hardness? We must think of these stones as the tools with which the brook cuts and grinds, for water without sediment can do little more than slightly to dissolve the rock. Let us go at once to the little waterfall, for we shall be curious to see what lies at the bottom of the whirling eddy that drew our attention yesterday. As we look down into the sunlit pool we see that the eddy is gone, for the volume of water is not great enough to cause it to revolve, but there in the rock on the bottom is a deep basin-like hole. In the bottom of this hole we shall see a number of well-rounded stones, with perhaps some sand and gravel. These stones are the tools which, whirled about by the eddying water, have cut the basin-like holes. Holes of this sort are common in rocky stream beds, especially in the neighborhood of falls or in places where falls have once been; they are called _pot-holes_ and represent another form of stream cutting (Fig. 48). [Illustration: _Fig. 48. A pot-hole cut in the rock of a stream's bed._] Next let us visit the flood plains which we saw forming when the water was high. Now we shall find the brook flowing in its channel with the flood plain deposits left high and dry. If we dig down into the flood plain, we shall see that it is made up of successive layers varying in thickness and in the size of the fragments. Each of these layers represents a period of high water and the size of the fragments in the layer tells us something of the strength of the current, and therefore of the intensity of the flood. Some layers are thicker than others, showing a longer period of flood, or perhaps several floods in which there was little variation. This _stratification_, as it is called, is one of the peculiarities of water deposits and it is due to the assorting power of currents which vary in force. If we were to cut into the delta we should find the same thing to be true,--a regular succession of layers, though sometimes confused by changes in direction of flow. To-day we shall notice something which escaped our attention when it was held by the rushing torrent--the valley bottom is much wider than the bed of the stream; if we keep our eyes open we shall see the explanation of this in the abandoned channels, where, owing to some temporary obstructions, the stream has been turned from side to side of the valley, now cutting on one bank and now on the other. In this turning from side to side the cutting area of the stream is increased, and it goes on widening its valley as well as cutting it downward.