have been rhythmic oscillations about a general average. Thus the climate of the earth as a whole has been of about the same order throughout geological time; the oldest sandstone in the British Is., the Torridon sandstone, was formed under desert conditions, and the size of its grains and the position of wind-polished surfaces on the pebbles show that the wind was of the same power and had the same prevalent direction as at present. The evidence of some of the oldest rocks shows that in the earliest geological times glacial con- ditions prevailed in parts of Canada, Spitsbergen, and North Siberia, while a little later (in the Cambrian period), central China was glaciated and ice floes floated in the seas of Australia up to the trop- ics. The climate of the world in the earliest geological periods was therefore no warmer than that of to-day. In Carboniferous times prolific vegetation grew the materials for the world's chief coal-fields, but at the same time glaciers existed in central India, south-eastern Brazil, South Africa and Australia.
Similarly, volcanic action shows no steady decline in power; periods of world-wide volcanic activity due to violent disturbances of the crust have alternated with periods of general volcanic rest.
The shape of the earth has doubtless throughout geological time been approximately an oblate spheroid, but it has been deformed to an irregular geoid by the hard crust sinking in places to follow the shrinkage of the internal mass. The excess of crust was most readily disposed of by subsidences on four surfaces producing the oceanic basins, while raised areas antipodal to the depressions formed the continental masses. The major subsidences have thus produced periodically a tetrahedral deformation, which was corrected, when- ever sufficiently developed to render the crust unstable, by spheroidal recovery accompanied by relatively quick and tumultuous earth movements. These major subsidences and uplifts have been ac- companied by the buckling of belts of the earth's crust into the great fold mountain chains; these movements happened at four main periods of mountain formation, represented by the Grampian folding of north-western Europe in the pre-Cambrian ; the Caledonian move- ments at the end of the Silurian; the Hercynian, which at the end of the Carboniferous produced the older east-to-west mountain chains from Asia across Europe to North America; and the Alpine-Hima- layan and west American mountain systems in the middle of the Cainozoic. Each of these great periods of earth movement was followed by intense climatic disturbance with local glaciations, and rapid biological evolution. Temporary changes in the environment were intensified by rearrangement of ocean and continent due to crustal movements across the older geographical structures.
Earth movements have not only determined the major elevations and depressions of the crust, but also the secondary depressions, such as sunklands, due to the subsidence of areas along peripheral faulting, rift valleys due to sinking of bands of country between parallel faults, and fiords due to formation of valleys along intersecting frac- tures when broad areas of hard rock have been raised in dome-shaped uplifts (Gregory, Origin of Fiords, 1913). The one change on the earth that has been apparently progressive has been the restriction in space of violent crustal movements. The Eozoic rocks, those of the older pre-Cambrian times, are steeply tilted in all parts of the world. Later rocks are often nearly horizontal, and steep tilting in them is confined to belts connected with mountain-forming move- ments; for with the growing strength and thickness of the crust the movements necessary for its accommodation to the reduced size of the earth have been concentrated along narrow bands.
The probable future changes in the nature of earth movements may be inferred from the study of the moon, which of all heavenly bodies is perhaps of most interest to the geologist, as, owing to its proximity, it is the one of which we have fairly precise information as to its topography. The map of the visible side of the moon is indeed more complete than that of the earth. The topography of the moon, like that of the earth, includes wide sunken areas, the "maria" which correspond to oceanic basins, long narrow mountain chains which are composed of parallel ranges, and volcanic craters, some of which are apparently still active. The most characteristic features of lunar topography are numerous ring-shaped mountains named vulcanoids; they surround circular or polygonal depressions and may correspond to volcanic caldrons, but are much larger than those on earth. They were first regarded as volcanic craters, but their intimate structure shows that they are composed of concentric ridges and neither of radial lava flows nor piled rings of volcanic ash. They have also been interpreted as impact rings raised by the fall of colossal meteor- ites; a theory which, however, does not account for their distribution. The moon differs from the earth by the mountain chains being relatively fewer and more fragmentary. The moon probably had a similar early history to that of the earth, but passed through the early stages more quickly because as a smaller body it cooled faster and acquired a thicker and less easily folded crust. It has apparently no shell resting upon a plastic layer, by sliding over which it can accommodate itself to changes of shape rendered necessary by the contraction of the interior. Owing to the absence of this friction- less layer and occurrence of the tremendous monthly changes in tem- perature the vertical movements on the lunar crust are more impor- tant than the horizontal. The chief younger lunar mountains are due to the vulcanoid subsidences, not to the crumpling of narrow belts. The ridges on the moon which appear to be fold mountains are old and broken, for it has reached a stage in which their formation
has either ceased or is dormant. The moon moreover, as it rotates on its axis only monthly instead of daily, is less flattened at the poles, and has not the folds parallel to the equator which are so important in the earth's geography.
The moulding of the moon's surface by denudation must be very different from that on the earth : the moon having practically no atmosphere and no surface waters, its rocks are not subject to or- dinary decay and wear, but, owing to the absence of an atmosphere and its day lasting for a terrestrial fortnight, it undergoes violent changes of temperature which must shatter the rocks and cause the fragments, by their lengthening and contraction at every change from day to night, to creep down even gentle slopes. The level areas on the moon are therefore probably covered by a sea of talus which spreads over them like the flow of the stone rivers of the Falkland Is., but is effective on gentler gradients.
The aspect of the moon indicates the probable future of the earth when, by continued cooling of the interior, the crust has become much thicker and more rigid. Its condition forecasts the state of the earth at some future time when its relief will be due almost entirely to vertical movements as the crumpling of belts of the crust will have ceased, and when horsts and sunklands due to faulting and valleys due to subsidence instead of to erosion will be the dominant fea- tures in the topography. (J. W. G.)
II. DYNAMICAL GEOLOGY
Volcanoes (see 28.178, 11.657). The most important later advance in our knowledge of the processes of volcanic activity is in the study of the gaseous emanations of volcanoes. These have been somewhat elaborately investigated in regard to Kilauea. Water vapour was found even in the gases most directly collected from the central lava column, associated with permanent gases, consisting mainly of carbon dioxide and monoxide, sulphur dioxide, nitrogen and hydrogen, with some free sulphur, chlorine, fluorine and perhaps ammonia. This group of gases, associated together at a temperature of 1000 C., cannot be in equilibrium; the hydrogen could not exist, except temporarily, in presence of the dioxides of carbon and sulphur, nor the free sulphur in the presence of carbon dioxide; chemical reactions would take place, all of which are accompanied by the evolution of heat. Other reactions, between the gases and the protoxide of iron in the lava, equally give rise to the evolution of heat, and in this is found an explanation of Brun's experience that when obsidian is raised to the temperature at which gases are freely evolved nothing can stop its expansion into pumice, the accession of heat from within, owing to chemical reaction, assisting the rapid expansion of the gas by a weakening of the containing walls of the cavities. It may be added that these chemical reactions also afford an explanation of the long distances to which some lava flows have been known to travel, the preservation of sufficient fluidity being due, not merely to the protective effect of a poorly conducting shell of solidified lava, but to the continued accession of heat.
At the time of its appearance much attention was attracted to A. Brun's theory, that the volcanic exhalation was essentially anhy- drous. Subsequent investigations, especially on Kilauea, have not borne out his contention in detail, and the theory has been generally rejected, at any rate in English countries; yet there is a probability that it may not be far from the truth, so far as paroxysmal eruptions are concerned. The generally accepted opinion, that the propulsive agency in these is predominantly steam, seems to be due to the resemblance in form between the cloud formed over a volcano in eruption, and clouds of condensed water vapour, formed in the upper air; to the presence of water vapour in the fumaroles on the sides of volcanoes, and in the emanations from lava flows; and to the occurrence of heavy rainfall in connexion with violent eruptions. The last-named is, however, by no means an invariable accompani- ment, and may be readily explained by the induced uprush of air which, in a humid atmosphere, would give rise to heavy rainfall ; and as regards the first, the resemblance is equally great to the clouds of smoke issuing from a furnace. Nor can dependence be placed on resemblances, for clouds may be caused by other vapours than that of water, and one formed of such vapour, mixed with the impalpable dust of an eruption, may be indistinguishable in appearance from the ordinary aqueous clouds of the atmosphere.
Direct observation is ordinarily impossible in a violent eruption, but, in 1911, G. Ponte was able to collect gas issuing from a lateral outflow of lava, derived directly from the column of lava in the cen- tral neck of the volcano, then in violent eruption. Analysis of the gas showed it to be composed mainly of carbon dioxide, with some nitrogen, sulphurous acid, sulphuretted hydrogen, nitrogen-carbon monoxide and methane but no water vapour. This observation is confirmatory of Brun's contention, and, even as regards Kilauea, there are some general considerations bearing on the question. Water is one of the end products of the reactions between magmatic gases
