These erratic blocks, now found all over Europe, tell us a good deal about the ice on which they were transported. The blocks that fall on the edges of the glacier remain on the side where they descend. Hence, if there is any notable difference in the composition of the rocks on either side of the valley, the existence of this difference will be preserved in the _moraines_. If, therefore, in a country where the glaciers have disappeared we can trace the scattered blocks up to their sources among the mountains, we can say what was the track followed by the prehistoric glaciers. In Europe there are several examples of the uses of this detective evidence. Thus the peculiar blocks of the Valais mountains can be traced right on to Lyons; and this shows us that the glacier from which the River Rhone sprang extended once right across the east of France to Lyons, and probably farther. It was therefore once at least 170 miles longer. Similarly, blocks which are exactly like the characteristic rocks peculiar to Southern Scandinavia are found in Northern Germany, Belgium, and East Anglia; and we therefore believe that a great sheet of ice once filled up the Baltic and the German Ocean, carrying with it immense numbers of northern "erratics." In our own country, in fact, glacier boulders are found in nearly every county, and show that once the greater part of the country was buried under ice. But, as we have said, it is not only on its shoulders, but in its interior and beneath its base that a glacier rolls and pushes its rubbish along. It is not all stones. Clay and earth mingle with it, often enclosing the stones; and the debris left by extinct glaciers of ages ago is sometimes called the boulder-clay. This is the deposit, earthy and stony, that the glacier leaves on the floor of the valley as it shrinks--unless the river which usually springs from the end of glaciers sweeps it away. Most of the stones thus left are smoothed or polished and covered with scratches or ruts, such as would be made by rubbing against other hard pointed fragments of stone. This is to be explained by the fact that these stones as they were carried on by the glacier were rubbed on the floor of rock over which the glacier was slipping. If their journey was long enough, they stood a chance of being rubbed away altogether and of finishing their existence as sand or mud. What the valley did to the glacier's stones, the stones did to the valley. They scratched it and scored it. Every promontory of rock which stood in the path of the ice had its angles and corners ground away. The polish and the directions of the scratches are especially remarkable, because, whether the marks are mere lines or deep-worn ruts, they are all on smooth surfaces, and they all run one way. That way is the direction in which the glacier moved. How high a degree of polish or how deep the markings may be depends a good deal on the kind of rock over which the glacier moved.