lar result. It has long been known that the apparent excess of matter in the visible prominence of mountain ranges was com- pensated by a diminished density, and consequent deficiency of matter, in the crust beneath them. The elaborate studies of Messrs. Hayford and Bowie, Becker and others, have shown that the prin- ciple is of wider extent, and that whatever the relative height of the surface may be the amount of matter under, any region is approxi- mately equal, the greater volume indicated by greater surface eleva- tion being compensated by a lesser average density. To this general principle the word isostasy is now generally applied, though the term was originally intended to indicate the process by which the result was brought about. The geological application of this conclusion, de- rived from geodetic observation, has been elaborately investigated by the late Prof. J. Barrel!, who concluded that the strength of the earth's crust as a whoje was about double that of good granite as we know it, that the thickness through which this strength continues is about 30 m., and that below it there is a rapid transition to mate- rial not much more resistant to change of shape than lead. He also calculated that the average departure from exact balance of surface relief and compensation, in the continental region of N. America, is equivalent to a load of average rock reaching a maximum thick- ness of about i, 600 ft., and thinning out to nothing at the ends of a length of about 200 m. ; for a greater length the maximum load would be correspondingly less, and greater for a shorter length. This does not indicate the maximum possible departure from exact isostasy, for even within the American region he finds evidence of about double (in the Pacific islands quadruple) that shown by the figures quoted.
The bearing of the recorded geodetic observations on the origin of the Himalayas and Pamirs, the largest and loftiest mountain region of the world, has been investigated by R. D. Oldham. He found that the excesses and deficits of matter indicated by observation were ranged in alternate bands, following the general course of the major surface relief; that the outer hills of the mountains bordering on the lowland plains showed an excess of matter; that farther in towards the centre of the hills this was followed by a zone of defi- ciency of matter, and that (though observations are wanting here) the most central portion is either in equilibrium or may show some excess of matter. The widths of these zones, the amount of the departure from exact equilibrium, and the distribution are very much the same as those which would result if a crust of the strength found by Prof. Barrell were uplifted by expansion of the material underly- ing, or possibly composing, its lower portion. Such an expansion of the underlying material, being necessarily accompanied by de- crease in density, would not only account for the compensation of the elevated region, but assumes a prominent position as the pri- mary phenomenon, and direct cause, to which the elevation of the mountain mass was due. This idea is not new, but while the only explanation of the changes of density was variation in temperature, the cause was, quantitatively, inadequate. Of late years, however, a new aspect has been given to the problem by Dr. L. L. Fermor, who has shown that a magma of the same chemical composition may solidify in more than one form of mineral complex, and that the density, and solid volumes, of these aggregates may vary by 20 to 30 per cent. To the two different forms of solidification he applied the terms plutonic or infraplutonic, and pointed out that the transition from one to the other form of mineral combination of the same chemical elements would depend on the balance between pressure and temperature; so that a small change in either might result in a passage from one mode, or mineral combination, to an- other of the same norm, or magma, and that this passage would be accompanied by a very considerable change in bulk, which, again, would result in uplift or sinking of the surface level. As the varia- tions in underground density revealed by geodetic measurements are not greater than those provided for by Dr. Fermor's supposition, it is evident that we have here at least one cause which would at the same time account for the major inequalities of surface level, and for what is known to geodesists as compensation.
We have, then, an agreement between two independent lines of re- search, one pointing to the conclusion that the greater inequalities of surface level are the result of underground changes in density, the other indicating with a high degree of probability that such changes do occur. It must, however, be added that a detailed study of published geodetic measurements suggests that this explanation is not quite complete, and that, as suggested by Dutton, the surface transference of material from regions of denudation to areas of de- position is accompanied by a corresponding underground transfer m the reverse direction ; yet the effect of any such transfer is sub- sidiary and trivial in comparison with that of the changes of den- sity to which the predominant proportion of the changes of surface level appear to be due.
This explanation is in accord with the fact that the larger fea- tures of the surface forms of the earth, such as the distinction between continental and oceanic areas, cannot be fully accounted for in terms of surface tectonics alone. It does not touch the importance of these tectonics in determining the secondary irregularities of sur- face; the well-established connexion of these with geological struc- ture of the surface rocks, on the one hand, and the processes of denu- dation and deposition, on the other, is unaffected, and it still re- mains possible that, within limits, hills and ranges may be directly
due to compression of the rocks composing them. These limits can- not be defined with precision, but, so far as figures available go, they may be put at the extreme figures of a breadth of 150 to 200 m. and a height of 2,000 to 3,000 ft. ; for a narrow base the height might be increased, for a greater width it would be less. These figures, though necessarily vague, give an idea of the limit of magnitude which can be allowed for surface inequalities resulting from superficial tec- tonics alone. For the larger features the lofty mountain ranges, the deeps of the ocean, the extensive elevated plateaus and the broad distinctions between ocean and continent we must look to causes and changes operating in the more deeply seated portions of the crust, or in the material underlying it.
AUTHORITIES. For Volcanoes see : R. T. Chamberlin, The Gases in Rocks, Publication No. 106, Carnegie Institution of Washington, 1909 (summary m Jour, of Geol., xvii., 1909, p. 534); A. L. Day and E. S. Shepherd, " Water and Volcanic Activity," Butt. Geol. Soc. America, xxiv., 1913, p. 573 ;.G. Ponte, " Richerche sulle escalazioni dell' Etna," At\i d. R. Accad. dei Lincei, xxiii., 1914, p. 341; J. P. Iddings, The Problem of Vulcanism (8, New Haven and London 1914); R. A. Daly, Igneous Rocks and their Origin (London and New York 1914); T. A. Jaggar, Jun., " Volcanologic Investigations at Kilauea," Am. Jour. Sci., 4th ser., xliv., 1917, p. 161.
For Earthquakes see : C. Davison, A Manual of Seismology (Cam- bridge 1921); R. D. Oldham, "Interpretation of the Californian Earthquake of 1906," Quart. Journ. Geol. Soc., Ixv., 1909, p. I ; A. Cavasino, " Frequenza e distribuzione dei terremote italiani," Bull. Soc. Sesmol. Italiana, xx., 1916, p. 9; R. D. Oldham, "A Seasonal Variation in the Frequency of Earthquakes," Quart. Journ. Geol. Soc., Ixxiv., 1919, p. 99; second communication, ibid., Ixxvii., 1921, p. I ; W. H. Goodchild, "The Evolution of Ore Deposits from Igneous Magmas," Mining Mag., xviii and xix., 1918.
For Mountain Ranges see: L. L. Fermor, "The Relationship of Isostasy, Earthquakes and Vulcanicity to the Earth's Infra-Plutonic Shell," Geol. Mag., 1914, p. 65; R. D. Oldham, "The Struc- ture of the Himalayas and of the Gangetic Plain as elucidated : Geodetic Observations in India," Mem. Geol. Survey of India, xlii., pt. 2, 1917; "The Support of the Mountains of Central Asia," Records Geol. Survey of India, xlix., 1918, p. 117; the last two give full reference to earlier works. (R. D. O.)
III. STRUCTURAL GEOLOGY
Isostasy. -During recent years, the consideration of the structure of the outer layers of the earth's crust, and of. the relation between areas of elevation and depression, has been greatly influenced by the doctrine of isostasy. It has been felt that the larger inequalities cannot be supported by the rigidity of the crust. They may, however, when once they have been established, be maintained on a yielding crust by transference of load. The processes of denudation carry material from the up- lands to the lowlands, and ultimately to the ocean basins, thereby lightening the upraised regions and increasing the weight on those already low (see Plate). The down-bent areas thus become fur- ther depressed, while mountains and continents rise as a conse- quence of their own decay. C. E. Dutton, who invented the word in 1889, has defined it as " the tendency to maintain moun- tain profiles in equilibrium " i(uros, equal, + crracrts, condition).
A simple case of isostatic response to change of load was suggested by T. F. Jamieson in 1865 (Quart. Journ. Geol. Soc., vol. 21, p. 178), when he explained the raised beaches of northern Europe as due to the depression of the region by the weight of ice in the Glacial epoch, followed by an upward swing of the crust when the burden was re- moved by melting. Jamieson's view has been extended by N. O. Hoist (1914), who points out that the subsequent rise may foster a second though minor glaciation, followed by renewed submergence; the final melting brought the various beaches to their present eleva- tions above the sea. It will be seen that the theory of isostatic re- sponse allows of irregularity of bulging, and is altogether more satis- factory than one which seeks to account for- raised beaches by a gen- eral change in volume of the sea. The matter is carried further by those who trace a movement of elevation contemporaneous with the northern depression in regions S. of the loaded area. Ph. Negris (1910) thus accounts for the uplift of the floor of the Pliocene sea in the Mediterranean basin, whereby shelly beds were carried to heights of 1,750 metres above the present level of the Aegean. H. Munthe (Internal. Geol. Congress, 1910, Excursions en Suede, No. 25) suggests a similar bulging for the periphery of the glaciated area of northern Europe, and a subsequent falling back as the nor- thern region rose. Isostatic recovery, as W. A. Johnston points out in Canada, may be concealed in certain places by the flooding of valleys with water previously locked up as ice. R. A. Daly believes that, after a general lowering of sea-level, due to a removal of water in a solid form towards the poles, a general rise of the sea took place in warmer times over the whole area of the Pacific Ocean . Fluc- tuations of the water-level, as well as permanent warping, involving imperfections of return to the original crustal relations, must clearly be allowed for in connecting raised beaches with isostasy.
