_Effects of heat and cold on rocks._—Most rocks are aggregates of several simple minerals; a few only (limestone, quartzite and a few others) expand or contract alike in all their parts. Of the minerals composing the compound rocks, scarcely any expand to exactly the same extent under the influence of the sun’s heat, especially when their colors differ; nor, in the great majority of cases, does one and the same mineral expand alike in all three directions. It follows that at each change of temperature there is a tendency to the formation of minute fissures between adjacent crystals or masses of different simple minerals; and especially in the case of large crystals of certain kinds, this action alone will gradually result in the disruption of the rock surface, so that individual crystals may be detached with little difficulty. In any case, the cracks so formed are gradually widened by a frequent repetition of the changes of temperature, coupled with access of air, water, dust, and the rootlets of plants; all of which brings about a gradually increasing rate of surface crumbling. This is especially conspicuous at the higher elevations of mountains, where the temperature changes are very great and abrupt; and also in the clear atmosphere of deserts, where owing to the extent and suddenness of temperature-changes between day and night, caused by the free radiation of heat into the clear sky, even homogeneous pebbles are known to be almost explosively disrupted in the mornings and evenings of clear days. Such effects may often be strikingly observed on small surfaces of compound crystalline rocks, such as granite, exposed on glaciers, where the daily changes of temperature are often extreme, viz., from below the freezing point to as much as 130 degrees Fahr. (54.4 degrees C). In such cases one may sometimes scoop off the disintegrated rock by the handful, while yet the mineral surfaces are almost perfectly fresh. On a larger scale, the disruption and scaling off of huge slabs of granite, and rocks of similar structure, may be observed in southern California on the southwestern side of rock exposures, where slabs from a few inches to ten and more feet in length and eight or ten inches thick, have slid off, perhaps still leaning against the parent rock, which has been rounded off by a succession of such events into the domelike form so characteristic of granite mountains. Merrill[4] reports similar exfoliations to occur especially on the peninsula of California, on Stone Mountain in Georgia, and elsewhere. [4] See Rocks, Rock-weathering, and Soils, page 246; also paper on Domes and Dome Structure, by G. K. Gilbert, in Bulletins of the Geol. Society Am., Vol. 15, pp. 29-36.