When cutting first began, the Monument’s canyons were narrow, steep, and V-shaped. When the top of the hard, dark Proterozoic rocks was reached, however, downcutting slowed just as it had earlier in Unaweep Canyon. While the streams were thus hung up, other erosional processes caused the cliff walls to recede away from the streams, forming broad, flat-bottomed, U-shaped canyons. Recession of the cliffs away from the middle of the canyons probably was caused partly by undercutting of the soft Chinle Formation by wind and in places by streams. This allowed slabs of the overlying Wingate Sandstone and younger rocks to break off and fall into the canyons—eventually to break up and to be carted off as sand and mud by streams. But other processes are probably the ones chiefly responsible for the present shape and width of the canyons. The summer sun heats the cliff faces until they are hot to the touch, but in the present desert climate of the Monument the rocks cool rapidly after sundown. Oftentimes the hot cliff faces are chilled rapidly by summer thundershowers. Repeated heating, cooling, wetting, and drying causes expansion and contraction of the rocks so that thin layers break off and fall. This process goes on slowly even in winter on sun-facing cliffs, but it does not occur on the cliffs that face away from the winter sun. Even more important, perhaps, is the alternate freezing at night and thawing by day on sun-facing cliff faces during the winter. Water in cracks near the cliff faces alternately freezes and melts, gradually prying off slabs of rock. Canyon walls that are shaded from the sun most of the winter, however, stay cold or frozen much of the winter; hence, they are not subject to repeated heating and cooling or freezing and thawing. Thus, you will notice that because of talus accumulation many such canyon walls are sloping rather than vertical. To illustrate the above conjectures concerning the cutting and shaping of the canyons, let us consider several canyons that trend in different directions. We have seen in figure 12 that the left side of northeastward trending Red Canyon is a nearly vertical cliff that faces the sun most of the winter; whereas, the right side, which is shaded most of the winter, slopes gently enough to be climbed at many places. The sides of Ute Canyon, which trends more nearly northward (fig. 52), slope about equally, as would be expected. However, the west arm of the Canyon, which trends slightly southeastward, has sides whose slopes differ markedly (fig. 35).