When the ice is advancing, the glaciers generally have a more convex outline, the icefalls are more broken into towers and spires, and piles of fresh rubbish are found shot over the grass of the lower moraines. Moraines which have been comparatively recently deposited by advancing ice are disturbed, show cracks, and are obviously being pushed forward or aside by the glacier. There is a tendency for the glacier to terminate in a vertical front, or “Chinese wall,” as distinguished from the sloping snout of a stationary or retreating glacier, owing to the more rapid advance of the upper layers. If the advance is rapid the overhanging layers will fall, and an ice talus will collect in front. An inspection of the ice will often disclose horizontal lines of sheering which in a side view are seen to rise towards the terminal wall. Search should be made along these sheer planes for included moraine material, and any proofs of elevation of this material noted. When the ice is in retreat it terminates in a gently sloping snout. The marks of its further recent extension are seen fringing the glacier both at the end and sides in their lower portions, the glacier fails to fill its former bed, and bare stony tracts, often interspersed with pools or lakelets, lie between the end of the glacier and the mounds of recent terminal moraines. Where a glacier has retreated to any considerable extent, careful observations of the form of its bed are of value. What is the nature of the rock surfaces exposed—convex or concave; are they rubbed smooth on their leesides; how far have the contours of the cliffs or slopes, or the sides of any gorge, been modified where they have been subjected to ice-friction? Is there any evidence that the ice has flowed over large boulders, or loose soils, such as gravel, without disturbing them? How has it affected rocks of different hardness, for instance, veins of quartz in a less hard rock? Generally, do the appearances indicate that the glacier has excavated, or only abraded and polished its bed; that it has scooped out new rock-basins, or only cleaned out, scratched, and preserved from filling-up by alluvial deposits or earthslips, existing basins? What is the general character of the valley bottom and the slopes above and below the most conspicuous ancient moraines?