Turning from the conditions of the atmosphere which affect the disruption of rock to the conditions of the rock which influence the same process, several points are to be noted. In the first place, the disrupting effects of changes of temperature are slight or nil where the solid rock is protected by soil, clay, sand, gravel, snow, or other incoherent material. If the constituent parts of the loose material are coarse, like bowlders, their surfaces are affected like those of larger bodies of rock. The color of rock, its texture and its composition, also influence its range of daily temperature by influencing absorption and conduction. Dark-colored rocks absorb more heat than light-colored ones, and compact rocks are better conductors than porous ones. Great absorption and slow conduction favor disruption. A given range of temperature is unequally effective on rocks of different mineral composition. In general crystalline rocks (igneous and metamorphic) are more subject to disruption by this means than sedimentary rocks, partly because they are more compact, but especially because they are made up of aggregates of crystals of different minerals which, under changes of temperature, expand and contract at different rates, while the common sedimentary rocks are made up largely of numerous particles of one mineral. [Illustration: +Fig.+ 29.—Top of Notch Peak, Bighorn Mountains, Wyo. Shows the thoroughly broken character of the rock on the summit, the absence of soil, vegetation, etc. (Kümmel.)] [Illustration: +Fig.+ 30.—A detail from Fig. 29 showing the size of the rock blocks. (Kümmel.)] [Illustration: +Fig.+ 31.—Peak north of Kearsarge Pass, the Sierras. Shows the way in which serrate peaks break up into angular blocks.] The freezing of water in the pores of rock is effective in disrupting them only when the pores are essentially full at the time of freezing. Otherwise there is room for the expansion attending the freezing. If the pores of the rock are large, the expansion on freezing may force out sufficient water to balance the increase of volume, even though the rock was completely saturated. If the pores be very small the water passes out less readily, and if the rock is saturated, freezing is more likely to be attended with disruption.[21] In view of these considerations the breaking of rock by changes of temperature should be greatest on the bare slopes of isolated elevations of crystalline rock, where the temperature conditions of temperate latitudes prevail, and where the atmosphere is relatively free from moisture. All these conditions are not often found in one place, but the disrupting effects of changing temperatures are best seen where several of them are associated (Figs. 29, 30, and 31).