INFLUENCES or AIR.] tight. In the United States, for example, with an annual thermometric range of more than 90° Fahr. this difficulty led to some experiments by Colonel Totten on the amount of expansion and contraction in different kinds of building- stones, caused by variations of temperature. It was found that in fine-grained granite the rate of expansion was 000004825 for every degree Fahr. of increment of heat; in white crystalline marble it was '000005668 ; and in red sandstone 000009532, or about twice as much as in granite. If the daily variations in temperature are large, the effects are still more striking. In tropical climates with intensely hot days and extremely cold nights, the rapid nocturnal contraction produces sometimes a strain so great as to rival frost in its influence upon the surface of exposed rocks, disintegrating them into sand, or causing them to crack or peel off in skins or irregular pieces. Dr Livingstone found in Africa (12° S. lat., 34° E. long.) that surfaces of rock which during the day were heated up to 137° Fahr. cooled so rapidly by radiation at night that, unable to sus- tain the strain of contraction, they split and threw off sharp angular fragments from a few ounces to 100 or 200 lb in weight.‘ 2. I"rosl.—Though properly belonging to the subject of the geological behaviour of water to be afterwards described in more detail, the disintegrating action of frost may be noticed here. In freezing water expands, and thereby exerts an enormous strain upon any enclosed cavities or walls which may confine it. The consequence of this action is that in countries exposed to frost a continual disinteg- ration of the surface of rocks goes on. This superficial decay combines with the chemical and mechanical opera- tions of the atmosphere to produce considerable modifica- tions in the forms of rocks and cliffs. 3. ll'r'ml.—By driving loose sand over rocks, prevalent winds produce on them a scratched and polished surface, as has happened with ancient monuments buried in the sands of the African deserts? It is said that at Cape Cod holes have even been drilled in window glass by the same agency.3 Cavities are now and then hollowed out of rocks by the gyration in them of little fragments of stone or grains of sand kept in motion by the wind. Hurricanes form important geological agents upon land in uprooting trees, and thus sometimes impeding the drainage of a country, and giving rise to the formation of peat mosses. Weat/zerz'ny of Ifoc7cs.—Under the term “weathering” are included all the superficial changes which rocks undergo in consequence of the action of atmospheric processes upon them. The nature and rapidity of the disintegration depend partly on the one hand upon the climate, and partly on the other upon the composition, texture, and exposure of the rocks. In very dry countries, where the range of tempera- ture is not extreme, weathering is reduced to a minimum. But even if the climate be dry, considerable disintegration may be caused, as has been already explained, by rapid changes of temperature between da.y and night. It is where moisture prevails, however, that weathering chiefly takes place. The nature of the changes will be more pro- perly considered in the section which is devoted to the action of rain. II. REPRODUCTIVE I.'FLUE_'CES.——Tl1eSe arise partly from the result of the chemical and mechanical disintegration involved in weathering, and partly from the transporting power of winds and aerial currents. Under the former head is the formation of soil; under the latter may be noticcd the production of sand-hills, the fall of dust-showers and coloured rain, and the transport of seeds. 1 Livingstone’s 'ZambesL, pp. 492, 516. 2 For an account of this action of drifting sand in North America see Blake in I’acz_'/is Railroad Iieport, V. 92, 230. 3 Dana's Jlmmal, p. 631. GEOLOGY 265 Soil.——Of the detritus produced by the action of the air on rocks, and washed away by rains and streams, pa.rt remains on the land and forms soil. All soil may be considered as the result of the decomposition of rocks, mingled with de- cayed vegetable and animal matter. Were it not for the action of rain in washing the loose materials to a greater or less distance from their source, the soil of every locality ought to be merely the decayed upper surface of the rocks underneath. But wherever rain falls, the soil is moved from higher to lower levels. Hence in some cases a good soil is laid down upon rocks which of themselves would only produce a poor one. This action of rain in the formation of soil is further alluded to on p. 270, and the co-operative influence of plant and animal life on p. 289. Sancl-hills or I)zmes.—Vinds blowing continuously upon loose materials, such as sand, drive them onward, and pile them into irregular heaps and ridges, called “ dunes.” This takes place more especially on windward coasts either of the sea or of large inland lakes, where the shores are sandy; but similar effects may be seen even in the heart of a continent, as in the sandy deserts of the Sahara and of Arabia. The dunes travel inland in parallel, irregular, and often confluent ridges, between which rain-water is some- times arrested to form pools (étcmgs of the French coasts), where formations of peat occasionally take place. On the coast of Gascony the sea for 100 miles is so barred by sand- dunes that in all that distance only two outlets exist for the discharge of the drainage of the interior. As fast as one ridge is driven away from a beach another forms in its place, so that a series of huge sandy billows, as it were, is continually on the move from the sea margin towards the interior. A stream or river may temporarily arrest their progress, b11t eventually they push the obstacle aside or in front of them. In this way the river Adour, on the west coast of France, has had its mouth shifted two or three miles. Occasionally, as at the mouthsof estuaries, the sand is blown across so as gradually to exclude the sea, and thus to aid the fluviatile deposits in adding to the breadth of the land.‘ The coast of Norfolk is fringed with sand-hills 50 to 60 feet high. On parts of the coast of Cornwall, the sand consists mainly of fragments of shells and corallines, and through the action of rain becomes sometimes indurated into a compact stone by carbonate of lime or oxide of iron. Long tracts of blown sand are likewise found along many parts of the Scottish and Irish coast—lines. On the western border of the European continent exten- sive sand-dunes exist. They extend for many leagues along the French coast, and thence, by Flanders and Holland, round to the shores of Courland and Pomerania. In Den- mark they are said to cover an area of 260 square miles. On the coast of Holland they are sometimes, though rarely, 260 feet high,—a common average height being 50 to 60 feet. The breadth of this maritime belt of sand varies considerably. On the east coast of Scotland it ranges from a few yards to 3 miles ; on the opposite side of the North Sea it attains on the Dutch coast sometimes to as much as 5 miles. The rate of progress of the dunes towards the interior depends upon the wind, the direction of the coast, and the nature of the ground over which they have to move. On the low and exposed shores of the Bay of Biscay, when not fixed by vegetation, they travel inland at a rate of about 161; feet; in Denmark at from 3 to ‘.74 feet per annum. In the course of their march they envelop 4 For accounts of sand-dunes, their extent, progress, structure, and the means employed to arrest their progress, the reader may consult Andersen's Klit_formationem. 1 vol. Svo, Copenhagen, 1861; Laval in Annales des Ponts et Chaussées, 1847, Qme sem.; and Mal‘-*ll_S Jllan and Nature, 1864, and the works cited by him. See also E110 de Beaumont, Lt-gens dc Géolo_r/ie, vol. i.
..—-34