[Illustration: FIG. 41--Kettleholes and other depressions in glacial till, on the Grand Trunk Railway about 5 miles southwest of Schoolcraft, Mich. The distance between the eastern (right) edge of this view and the western (left) of Fig. 37 is about 1 mile. Scale, about 1:15,000.] These photographs were selected from a series taken as an experiment in map-making. In June, 1920, the United States Air Service sent a plane equipped with a K-1 camera from Dayton, Ohio, to Schoolcraft, Mich, where in seven hours’ flying time a fifteen-minute quadrangle, about 220 square miles, was photographed. The prints were matched together and reduced to a scale of 1:48,000. From them such features as roads, streams, forests, land corners, etc., were transferred to plane-table sheets, which the topographic engineers on the ground then used for contouring the relief. Figure 38 is a part of the preliminary proof of this map. It may be added that the experiment is regarded as highly favorable to the use of the airplane camera as an instrument in mapping. CHAPTER X MOUNTAIN FEATURES (FIGS. 42 TO 52) In obtaining photographic illustrations from the ground of mountains, canyons, and associated land forms, the same difficulty, but in exaggerated form, is encountered that obtains in securing an advantageous point of view for small objects. The difficulty is overcome in large measure by the use of aircraft. In an airplane the observer can rise above the obstructions which interfere with the view desired; can look an isolated mountain peak squarely in the face, as in the case of the photograph of Mt. Shasta (Fig. 42); can study the details of its ice cap (Fig. 42) and gaze downward on the lateral and recessional moraines left by the retreat of the mountain’s glaciers (Fig. 43). Few volcanic craters, occurring as they do at the top of cones, have been successfully photographed unless some higher mountain stands near-by on which a favorable viewpoint can be found. From an airplane, however, one can look into the very throat of a crater, as into that of Cinder Cone (Fig. 48), near Lassen Peak, California. Much attention has been given to the interrelations of canyons, gorges, and mountain ridges, but these relations have hitherto been illustrated chiefly by means of maps and charts. Figures 49, 50, and 52 picture three relations more expressively than any map. To the experienced geographer a map may illustrate perfectly the action of a stream working headward into higher land; but the student to whom the conception of headward erosion is new will certainly grasp the idea more readily from the picture presented in Figure 52. No map could give so clear a conception of a maturely dissected highland as does a photograph like that of the Santa Monica Mountains (Fig. 50).