The Origin of Continents and Oceans/Chapter 11

by Alfred Wegener

CHAPTER XI

FOLDS AND RIFTS

As far back as 1878, A. Heim had developed in his classical paper, “Untersuchungen über den Mechanismus der Gebirgsbildung,” the idea that the great chains of folded mountains originated in a considerable compression of the earth’s crust. This doctrine was amplified by the discovery of the imbricated sheet folds in the Alps which indicated a still more powerful compression.

According to this new conception, to which he assented, A. Heim now calculated the compression of the Alps, which he had first estimated at a half, to be from a quarter to an eighth. Later, Ampferer assumed flow-movements of the deeper layers which, directed towards each other from both sides, pass downwards under the mountain chain and carry the upper beds passively along with them (under-currents). Koszmat has quite recently referred to the curvature of the mountain chains, and their fan-shaped association at many places. This appears to him only explainable by great horizontal displacements; “many features in the relief and structure of the earth seem to make it certain that an explanation of mountain-building must take into account enormous tangential movements of the crust.^[1] This conception is in itself almost identical with that of the displacement theory, for it only means a small step forward if the mountain chains of the Himalayas are explained by a gigantic thrusting-forward of a long portion of the crust, the southern end of which, India of the present day, once lay near to Madagascar.

This development, step by step, of the theory of horizontal thrusting, has been accompanied by a series of other attempts to explain mountain-building, according to which the heaving-up of the mountains takes place through internal forces, such as volcanic agencies, the pressure of growing crystals, chemical transformations, or the space requirements of volcanic intrusions.^[2] Although the occasional co-operation of such causes should not be denied, it appears to me hopeless to seek herein the main causes of mountain-building. Hence we cannot go any further into these ideas.

Gravity measurements are important for the understanding of folding processes. Koszmat has investigated them for Central Europe in an interesting work,^[3] from which the map of Fig. 31 is taken. The gravity values actually observed are, as usual, reduced in such a way that we may suppose that the whole relief of the earth was planed down to sea-level and the measurement was carried out at this zero altitude; that is, besides the reduction to sea-level, the influence of the mass above it is also taken away from the result. The observed value so reduced is then compared

Fig. 31.—Gravity abnormality beneath mountain ranges, after Koszmat.

with the received normal value for the geographical latitude in question and the difference, the gravity anomaly, is given on the map. It shows us immediately the mass-defect under mountain chains through which the latter are isostatically compensated. “One can only arrive at the conception already expressed by many geophysicists, and by Heim, that it was not a dilatation that caused the defect, but that the upper relatively light parts of the earth’s crust are greatly thickened by the folding, and that this accumulation, during its formation, sank into the plastic lower layer. A folded mountain chain grew not only upwards, but, owing to its weight, downwards also: the folded elevation, is, as Heim expressed it, opposed by a still greater folded depression.” We can thus see directly in the map the approximate topography of the underside of the sial crust; beneath the Alps, where the gravity anomalies reach the greatest negative value, the lower side of the sial crust also sinks deepest into the sima.

If, on the other hand, we had not taken away from the observed values the influence of the masses above the sea-level, we should not have obtained such gravity anomalies in the region of the mountains, but—although with minor deviations—have found normal values there. Subterranean mass-defect and mass-excess above the sea-level therefore mutually counterbalance, isostasy thus prevailing in mountain masses. Since this is proved to be the same in ancient as well as recent mountains, we arrive at the principle: the folding of mountains is a compression subject to the preservation of isostasy.

It is advisable to make clear, as in Fig. 32, what this means. In the compression of a block swimming in the sima, the ratio of the portions above and below the mean level of the sima must always remain the same. Since we assume that the continental block rising 5 km. out of the sima is 100 km. thick, we can take this ratio to be about 1:20. Hence the downwardly directed part of the compressed sial must be 20 times greater than that directed above. What we see in the mountain chains is thus only a very small portion of the whole squeezed up mass, and this consists only of those strata which already lay above the ocean-floor even before the compression. All that lay below this level remains there after the compression, if disturbances are disregarded. If thus the upper structure of the block consisted of a layer of sediments 5 km. thick, then the whole mountain mass would originally consist of sedimentary deposits. Only when this becomes worn away by erosion does a central chain of
Fig. 32.—Compression without disturbance of isostasy. primitive rocks rise up to isostatic compensation, until finally a broad massive of practically the same average height is visible after the more complete peeling of the sedimentary cover.

The Himalayas and their neighbouring mountain chains can be considered as examples of the first stage. The erosion is enormous in these sedimentary folds, so that glaciers are practically buried beneath the débris—as, for example, the Baltoro glacier, the greatest in the Kara-Korum range, which, with a breadth of 11/2 to 4 km. (length 56 km.), carries no fewer than 15 median moraines. The second stage, in which the central chain is already composed of primitive rocks, but is still flanked on both sides with a zone of sediments, is found in the Alps. Since erosion is much slighter in primitive rocks, the Alpine glaciers are poor in moraines, one of the chief reasons for their beauty. Finally, the Norwegian mountains represent the third stage: the sedimentary cover is entirely removed and the elevation of the massive completed. Thus the denudation of the sedimentary capping of a mountain chain is effected without disturbing isostatic adjustment.

We must deal shortly with the asymmetry of folded mountain chains, which can nearly always be recognized. It is a general rule that, on approaching a mountain mass from one side, one has to pass a general gradual rise of ground, often also foothills and similar features, whilst on the other side the “foredeep” reaches close up to the main system of folds. Much has been written about this. A very simple explanation is yielded by our hypothesis. The masses of sial, pushed deep down by the folding, spread themselves out, and to some extent penetrate beneath the unfolded crust, near the fold, bearing it up, naturally at the expense of the altitude of the mountains themselves. This spreading out can only be symmetrical on both sides when the crust has no progressive movement over the sima. But if the crust is displaced—not taking into account its folding movements—as a whole over the sima (which will almost always be the case), then the spreading out of the sial masses must take place unilaterally. The European and Asiatic blocks have doubtless for a long time endeavoured to reach the equator, and thus are thrust southwards relative to the sima. Moreover, they probably take part in the general westerly drift of the continents. Therefore the total movement relative to the sima will be directed to the south-west. The spread of the masses of sial beneath them must follow unilaterally to the north-east. This, as a matter of fact, is the case, and is shown on the map of Fig. 31. The displacement of the gravity anomaly, or, what amounts to the same thing, of the depressed sial masses, is especially distinct in the case of the Apennines. But the mass deficit of the Alps also extends far on the north-east to the northern boundary of Bohemia and towards Central Germany. And conversely a zone of excess of mass extends from the south under the Alpine chains, which indicates that no depression of the sial masses to great depths corresponds to the superficial folds. At this place, however, a considerable deviation from isostasy occurs, which can be easily explained, for the abnormally high position of the base of the crust according to the map should, if isostasy prevails, correspond to a smaller thickness of the block; that is, the upper surface of the block should be of exceptionally low level in the area, lying everywhere under water. If, on the other hand, considerable altitudes above sea-level are here reached, this can only be rendered possible by deviation from isostasy, these belts of crust being held considerably above their position of equilibrium by their rigid connection with the neighbouring rocks. The map of Koszmat does not by any means give us direct information of these deviations from isostasy.

Sir James Hall was the first to note that the thickness of the sediments in folded mountain chains is always greater than in the neighbouring undisturbed areas. In other words, in this region, before the folding, these sedimentary strata were deposited in greater thickness than in the neighbouring area. This rule has been found to be universally true, and has given geologists much to ponder over. Since the deposits under consideration, which are often many kilometres in thickness, must have all been formed in shallow water, it must be assumed that the block sank simultaneously with the deposition, whilst the sedimentation continued, so that the surface for the time being always remained at the same elevation. This was explained by Hall by means of isostatic movements of compensation on account of the weight of the deposits, in a similar manner to the submergence of a continental block beneath the burden of land-ice. But why is it just those portions of the blocks that have especially thick sedimentary deposits which are later involved in the folding? Such regions with heavy sedimentation are known as geosynclinals (basins). Haug formulated the law thus: mountain chains are formed from geosynclinals.^[4] Probably it might be better expressed: from continental shelves; because a marginal shelf—as, for example, that from which the South American Andes are built—can hardly be called a basin. There are a variety of reasons why the shelves are favoured by folding. It has already been mentioned that they perhaps contain especially numerous and large inclusions of sima, and are on that account more plastic. It may also be possible that here the sial crust possesses a lesser total thickness, and thus also a lesser power of resistance. Reade suggested that the primitive rock was forced down by the thick deposits into the region of higher temperature, and was therefore made more plastic. Perhaps all these causes work in co-operation.

If we consider the regional distribution of the folded mountain systems, we see that two regions are favoured by them: the anterior margin of drifting blocks and the equatorial belt. This strikes us particularly forcibly in the case of the last great period of folding; it is that of the Tertiary, in which, on the one hand, the chief folding is found on the anterior margin of the American blocks as well as of that of Australia-New Guinea, and on the other in the zone of the Tertiary equator, from the Atlas over the Alps and the Caucasus to the Himalayas. Folding on the anterior margin of drifting blocks seems at first sight to offer difficulties in comprehension, since the sima should surely be the more liquid and the sial the more rigid material. But we must remember our comparison between sealing-wax and wax. Thus if the sial block can be considered as solid, like wax, it will nevertheless be folded when the forces of displacement exceed a certain value. The sima certainly gives way and flows as sealing-wax does—but an exceedingly long period of time is needed for that.

The kinds of folding mentioned—at the front margin and at the equator—correspond on the whole to the two movements of the continental blocks, their westerly drift and their drift from the poles. That this rule holds for more ancient periods, in particular the Carboniferous, is shown by the old folds which form the basis of the Andes, as well as the Carboniferous mountain-system which can be followed along the equatorial zone of that time from North America across Europe towards Eastern Asia.

It can frequently be seen that the parallel-folded chains of a mountain system lie in echelon. If such a folded chain be followed, it is found that sooner or later it comes out on the margin of the mountain system and finally disappears, whereupon the next inner chain forms the margin, disappearing likewise some distance farther, and so on. This is the case when the two blocks do not move directly towards each other, but have a shearing movement though with a direct component. The effect of the different movements of the blocks relative to each other may be in a general way illustrated by Fig. 33, in which the left block is supposed to be fixed and the right to move. If the movement be directed at right angles to the border of the blocks, extremely large folds (overfolds or overthrusts) are formed, but none in echelon; if it be oblique to the boundary, folds in echelon are formed which will be narrower and of less altitude the more parallel the direction of movement is to the margin of the blocks. With exact parallelism, there will be a gliding surface with lateral displacement. Finally, if the movement possesses a component which is directed away from the margin of the blocks, there will be either oblique or normal rifting, which then manifests itself in a rift valley. We can illustrate in a very satisfactory manner the relation of the
Fig. 33.—Folding or rifting as the result of differently directed movements of the blocks. normal folds to the folds in echelon with a table-cloth, if we weight down that portion which will represent the immovable block and displace the other part relatively to it.

These general considerations show that folding and rifting are only different effects of one and the same process, namely, the displacement of the parts of a block relatively to one another, and that they pass continuously into each other through echeloned folds and lateral displacements. It is therefore right that we should now consider also the process of rifting.

The East African rift-valleys form the most beautiful example of such rifts. They belong to a great system of faults which can be followed northwards through the Red Sea, the Gulf of Akaba, and the Jordan valley, up to the margin of the folds of the Taurus (Fig. 34). According to more recent work, these faults also continue southwards down to Cape Colony, being, however, best developed in East Africa.^[5] Neumayr-Uhlig^[6] describes them somewhat as follows:

Such a rift-valley 50 to 80 km. wide containing the R. Shire and Lake Nyassa stretches northwards from the estuary of the Zambesi and then turns to the north-west and loses itself. The parallel trough of Lake Tanganyika begins very close to it. Its magnitude is testified by the fact that the depth of the lake amounts to 1700 to 2700 m., and the elevation of the wall-like slope to 2000 to 2400 m., and even up to 3000 m. This rift-valley contains the river Russisi, and the lakes of Kiwu, Albert-Edward and Albert, in its northerly continuation. “The margins of the trough seem to be turned up just as if the fracture of the earth’s crust was connected with a certain upward movement of the broken margins when suddenly set free. On this peculiar turned-up form of the edges of the plateau depends the fact that the sources of the Nile arise immediately to the east of the slope down to the Tanganyika valley, whilst the lake itself discharges itself into the Congo.” A third well-marked trough commences east of the Victoria Nyanza, contains farther north Lake Rudolph, and turns off to the north-east near Abyssinia, where it continues on one side into the Red Sea, and on the other into the Gulf of Aden. In the coastal region and in the interior of German East Africa these faults usually assume the form of step-faults, with down-throws to the east.^[7]

The great triangle in the angle between Abyssinia and the Somaliland peninsula (between Ankober, Berbera and Massowa), shown dotted in Fig. 34 in a similar manner to the floor of the troughs is of special interest. This relatively level and low-lying area consists entirely of late volcanic lavas. Most authors consider it to be an enormous widening of the floor of the rift. This idea is suggested chiefly by the course of the coast-lines on both sides of the Red Sea, the parallelism of which is disturbed by this projection; if it be cut away, the opposite corner of Arabia exactly fits into the notch. It has already been mentioned that we are obviously dealing with masses of sial from the underside of the Abyssinian mountains, which have spread out unilaterally towards the north-east and have thus emerged at the margin of the block. Perhaps the rift was already filled with sima, so that the rising mass of sial carried with it a cap of this material; or else there might have been great inclusions of sima in this welling-up mass of sial, which would be subsequently pressed out in the same way as in Iceland. In any case, the great elevation above the level of the ocean betrays the presence of sial masses beneath the lava-flows.

The date of the origin of these faults, which are arranged in a mesh-like manner in East Africa, is to be placed in a geologically late period. At many places they cut late basaltic lavas, and at once place Pliocene fresh-water formations. Thus in any case they cannot have been formed before the close of the Tertiary period. On the other hand, they appear to have existed in the Pleistocene, as has been deduced from the raised beaches marking higher water-levels in the lakes without outlet, lying in the bottom of the troughs. In Lake Tanganyika, its obviously earlier marine fauna, which later adapted itself to fresh-water conditions (relict fauna), points to a somewhat long existence. But the frequent earthquakes and the strong vulcanicity of the fault zone certainly indicate that in any case the process of separation is similar to that in progress to-day. The only new fact regarding the mechanical significance of such rift-valleys is that they form the first steps of a complete separation of the two parts of the block, whether it is a matter of recent still unfinished rifts, or earlier attempts at one, which have become quiescent again in consequence of the weakening of the tensile forces. According to our ideas, a complete separation would take place somewhat in the manner diagrammatically shown in Fig. 35. First an opening cleft arises in the more brittle upper strata, whilst the more plastic layers below are stretched. Since steep vertical walls, of the height here in question, would make much too great demands on the resistance to pressure of the rocks, inclined gliding planes of slipping are formed simultaneously with the rift, or in place of it, along which the marginal parts of both portions of the block sink into the rift as it opens, to the accompaniment of numerous local earthquakes, so that only a rift-valley of moderate depth is exposed to view, with its floor formed of faulted blocks of the same series of rocks as those which occur in situ high up on the margin of the trough. At this stage the rift-valley is not yet isostatically compensated, and this is the case, according to E. Kohlschütter,^[8] with a large portion of the recent troughs of East Africa. An uncompensated mass-defect exists; therefore a corresponding abnormality of gravity is observed, and, what
Fig. 35.—Rifting (diagrammatic). is more, both margins of the rift rise in isostatic compensation, so that the impression is produced that the trough passes exactly longitudinally through an anticline. The Black Forest and the Vosges on the two sides of the Rhine Valley are well-known examples of this marginal elevation. Finally, if the split travels right through the block, the sima will rise up into it, so that the previous mass-defect is lost, and the trough is henceforth found to be, as a whole, isostatically compensated. In most places the floor of the trough is completely covered with fragments from the margins of the rift, but naturally, by further widening, the time comes when the free surface of the sima is also exposed. In the case of the great rift-valley of the Red Sea, which, according to Triulzi and Hecker, is already isostatically compensated, development may have gone so far that at the deeper parts the sima is uncovered. With a further separation of the blocks, the pieces broken from the margins remain behind as islands. It is to be noticed that these pieces, even if in their highest portions they reach or exceed the level of the continent, need not have the same thickness throughout as the continental blocks. Instead, they need only be essentially broader in the submerged than in the elevated portion. Thus here the only condition which needs to be fulfilled is the one that the proportion between the weights above and below the level of the deep-sea floor is the same as in the great continental platforms. All these views as to the nature of the rift-valleys are not contradictory to those now current, but merely supplement them.

Just as a single rift can be resolved occasionally into an extensive meshed net of smaller rifts (the system of the East African troughs, which pass over into a solitary rift in the Red Sea, forms an example of this), so instead also of a single rift-valley the breaking down of an extensive area can be accomplished. The Ægean Sea is the best example of this. Here a large area was broken up in quite late geological time into single blocks, which were submerged to dissimilar depths. We must assume that the lower layers of the lithosphere have stretched so that the fault-rifts gradually disappear downwards. The amount of horizontal extension can be measured on the inclined fault planes, shown diagrammatically in Fig. 36, so far as they are free. At many other places a land connection has obviously been submerged in a similar way; for example, in the Bass Straits between Australia and Tasmania. But it is easily seen that there is a limit to this depression and that a complete tearing and separation of the two blocks must take place long before the sinking portions have reached the level of the deep-sea floor. The foundering of the English Channel, the North Sea, and other former land areas, now transformed into portions of the continental shelf around England, took place, according to our idea, immediately before the breaking away of Newfoundland from Ireland. But they became nevertheless only shallow shelves, for the complete separation of the blocks occurred farther to the west.

If the geographical orientation of the chief rifts
Fig. 36.—Extensive collapse due to the stretching of the lower crust (diagrammatic). of the sial crust be considered, the predominance of the meridional direction can be recognized, although many abnormalities occur. Not only does this apply to the rift system of East Africa, to which reference has already been made, and to the Rhine trough formed in the Oligocene, but the Atlantic rift also follows a course which, for the position of the Tertiary poles, is essentially meridional. The same applies to the rift of which one side forms the eastern margin of Africa. The southerly tapering of the continents in South America, South Africa, and India may be traced back to such meridional rifts extending to the pole.

1. F. Koszmat, “Erörterungen zu A. Wegener’s Theorie der Kontinentalverschiebungen,” Zeitschr. d. Ges. f. Erdk. zu Berlin, p. 103, 1921.
2. Compare with this K. Andrée, Über die Bedingungen der Gebirgsbildung, Berlin, 1914. Walther Penck is an extreme advocate of the intrusion theory (“Die Entstehung der Gebirge der Erde,” Deutsche Revue, Sept.–Oct., 1921).
3. F. Koszmat, “Die mediterranen Kettengebirge in ihrer Beziehung zum Gleichgewichtszustande der Erdrinde,” Abh. d. Math. Naturw. Kl. d. Sächs. Akad. d. Wiss., 38, No. 2, Leipzig, 1921.—“Die Beziehung zwischen Schwereanomalien und Bau der Erdrinde,” Geol. Rundsch., 12, pp. 165–189, 1921.
4. Haug, Traité de Geologie, 1, “Les phénomènes géologiques,” p. 160. Paris, 1907.
5. Oskar Erich Meyer, “Die Brüche von Deutsch-Ostafrika,” Neues Jahrb. f. Min., Geol. und Paläont., Beil.-Bd. 38, pp. 805–881, 1915.
6. Neumayr-Uhlig, Erdgeschichte, 1, Allgem. Geol., 2. Aufl., pp. 1–367. Leipzig and Vienna, 1897.
7. Compare “Die Karten des Abflusslosen Rumpfschollenlandes im nordöstlichen Deutsch-Ostafrika,” von E. Obst.
8. E. Kohlschütter, “Über den Bau der Erdkruste in Deutsch-Ostafrika,” Nachr. d. Kgl. Ges. der Wiss. Göttingen, Math.-Phys. Kl., 1911.