Alfred Wegener

by Mott T. Greene

  Wegener imagined a history of Earth in which mountain building causes a continuous thickening of the continental blocks at the expense of their horizontal extent. He imagined that once a salic (or “sialic”) crust, roughly 30 kilometers (19 miles) thick, had covered the entire Earth, covered in its turn by a “Panthalassa,” a universal ocean about 3 kilometers (1.9 miles) deep. This vision is the exact opposite of that put forward by Edward Sueß in The Face of the Earth. Sueß imagined the “breaking up of the terrestrial globe,” in which the continents would eventually sink entirely beneath just such a universal ocean.59 Rather than the destruction of continental masses and their transformation into oceanic deeps as a consequence of ongoing Earth contraction, Wegener imagines the successive emergence of thicker continental blocks separated by broader oceans. Once again the Copernican motif, in consonance with his view of the direction of the evolution of the crust, emerges in high relief.

  Wegener’s treatment of the question of volcanism, like his treatment of mountain building, considered the topic from a geophysical rather than geological standpoint. “Up to now,” Wegener wrote, “we have completely disregarded an obvious objection. With every displacement of continental platforms, the underlying high temperature Sima will be laid bare under the oceans. Must this not lead to catastrophic events?”60 Wegener’s answer is that it need not. If the Sima is denser than the Sal of the crust, it cannot rise into an open fissure and cause a volcanic eruption, unless it is under pressure; this is because of isostasy. If a continent should rift into two fragments, these fragments would still float at the same depth in the subcrust, and there would be nothing to force the hot material below upward. Isostasy predicts that the crust and the subcrust are in equilibrium, and there is no reason on the physical side for the hot matter exposed by this continental drift to surge upward. Rather, Wegener argued, one should expect volcanoes to be concentrated in regions of compressive stress (such as the Pacific margin of South America), rather than regions of tension (the Atlantic margin of South America). Of more interest to Wegener was the episodic character of volcanic outbreaks, which seemed to him—or at least he allowed himself to imagine—to correspond precisely to those periods of continental displacement: one in the Carboniferous and Permian, and another great phase in the Tertiary. These would correspond directly to Wegener’s notions of the breakup of the continents first in the Southern Hemisphere and later in the Northern Hemisphere.

  This was important material for anyone wishing to talk to geologists, and one might have expected a fuller discussion. Yet one’s overall impression is that Wegener handled it in a hurried and cursory fashion. One can sense the pressure he was under to complete this manuscript in the short window of time allotted him before his departure for Copenhagen and then on to Iceland and Greenland. The haste, incomplete character, and compression of argument are evident in his brief and somewhat apologetic remarks at the end of the first section of his paper, when he turns to “Remarks on the Causes of the Displacements”:

  The question concerning which forces cause the horizontal displacement of the continents we have advocated is so obvious that I cannot completely overlook it, although I am of the opinion that it is premature. It is unquestionably necessary first to establish exactly the reality and the type of the displacements before one can hope to fathom their causes. It can essentially only be a question of forestalling false interpretations, and, to a lesser extent, to present such interpretations that might already be able to make a claim to correctness … an obvious candidate to bring forward as a cause of variation of the pole of rotation.61

  Wegener was not, however, at this point predisposed to accept variation in the position or inclination of the pole of rotation as a cause of continental displacement, but rather to imagine that displacement of the pole would be a consequence of redistribution of continental masses. The displacement of the pole must then itself have a cause, and Wegener’s hurriedly sketched hypothesis was for lunar tides in the body of Earth leading to meridional fractures, in combination with “fortuitous streaming in the body of Earth,” as the cause of displacement following these initial meridional fractures. One hardly knows what he meant here (and it is not clear that he knew either), except that the continents, especially in the Southern Hemisphere, have long tapering forms with their long axes along meridians of longitude rather than parallels of latitude. The vagueness of these suggestions and the impatient, naïve enthusiasm they betray seem to have struck even Wegener, who completed this section with a firm declaration that “the time seems, however, as I said, not yet right for these questions.”62

  Having established the qualitative plausibility of continental displacements from the standpoint of geophysics, Wegener’s task in the subsequent section of the paper—on the specifically geological arguments—was to bring to the forefront of discussion those structural features of Earth’s crust left completely unexplained, or only partly explained within the context of the contraction theory, in the form given it by Sueß. He aimed to show that the theory of continental displacements did all the epistemic “geological work” of the contraction theory and a good deal of additional geological work that the contraction theory was unable to perform. It explained all the same geological phenomena that the contraction theory explained, as well as a variety of features and phenomena it did not explain. His arguments on these crucial matters are laid out in the next chapter.

  11

  The Theorist of Continental Drift (2)

  MARBURG, FEBRUARY–APRIL 1912

  Before we undertake to pursue the processes of rifting and compression of continental blocks in the history of Earth, discussed in the first part of this work, it should be pointed out once more that such an initial, tentative investigation must of necessity prove incomplete in many respects, and in other respects perhaps incorrect. All the same, it must be ventured. For once the major considerations have been established, it will not be difficult, through detailed investigations, to eliminate errors.

  ALFRED WEGENER, 1912

  Geological Arguments and Measurement

  Wegener had worked hard to establish the plausibility of his hypothesis of continental motion and the corresponding implausibility of the theory of Earth contraction. His was a hypothesis of very great generality: continents in general, mountains in general, volcanoes in general. But geology is a very concrete science. If physics and physicists took pleasure in simplicity and unity, geologists seemed to glory in variety and heterogeneity. A geologist’s brain is a gigantic lumber room of Platonic forms, not a sparse laboratory containing a few essential shapes. Fortunately, Wegener seems to have had an inkling of this and the necessity of making his case apply to specific kinds of structures and specific sets of geological data.

  His reading of Sueß and of Kayser had given him a broad overview of the major problems of geology seen from the standpoint of geologists. His line of geological attack, as suggested at the end of the previous chapter, was to select major problems discussed by Sueß and others but not solved by them. He wished to show that his theory could not only reorganize existing geological knowledge but also contribute to the solution of outstanding problems.

  He therefore constructed his argument around specific large-scale features of Earth’s surface, features left only partly explained in the existing geological literature. Here are the major geological features of Earth’s crust which, in Wegener’s estimation, had been left unexplained by the contraction theory:

  The character of large fault troughs (Grabens).

  The relationship of the Atlantic to the Andes.

  The extent and location of the paleocontinent Gondwanaland.

  The area and extent of the Permian Ice Age.

  The difference between Atlantic and Pacific sides of Earth.

  The question of the displacement of the poles.1

  Wegener first considered the matter of the fault troughs or Grabens. These structures, appearing across a broad range of scales, had a very important function
in Sueß’s theory of the formation of the continents. For Sueß, the continents were “Horsts,” giant fault blocks flanked by oceanic “Grabens,” which he imagined to be downfaulted sections of former continental crust. Wegener had already argued against the plausibility of this model, based on density data for the continental surfaces and the ocean floors. Gravity surveys taken at the continental margins suggested that gravity anomalies in this region could best be explained by assuming the ocean floor to be more dense than the continental material adjacent to it, and therefore that it was a distinct Earth shell and not a former continental surface. Now Wegener wanted to turn from such general geophysical considerations to specific geological examples of these phenomena.

  The first of Wegener’s geological examples was the Rhine Graben, a North German fault trough well known to his audience. It had been drilled to considerable depth and the sediment cores from the borholes minutely analyzed. These core samples showed that the material at the bottom of the trough was indeed the same sedimentary material, in the same sequence, as that of the remaining “horsts” flanking it. Gravity data, however, indicated no negative gravity anomalies within the graben. Even though very light sedimentary material was supposed to have “sunk” (which should reduce the value of gravity in this location), the specific gravity of the material deep in the trough was higher than in the flanking hills. Wegener took the opportunity provided by these data to infer that the graben was not a coherent “dropped” fault block but a tensional rift contemporary with the splitting of America from Europe. The equivalence of the sediments in the graben with those in the horsts on both sides was the result, rather, of the slippage of sediments at the margins of the widening rift. The gravity data could be explained by the rising of the Sima into the tensional rift, reestablishing isostatic equilibrium.2

  Wegener’s second example bearing on this question was the rift valleys of East Africa. Sueß had offered no dynamical explanation for why these great rifts should be where they are. On purely morphological grounds, he had asserted that they appeared to be tensional structures.3 In these African rift valleys, gravity measurements showed, in contrast to the Rhine Graben, negative anomalies (less than expected values of gravity). Wegener’s interpretation of this case is that the East African rift valleys were tensional rifts opening from the top down which had not yet cut through the full depth of the continental block to let in the Sima. Since the East African rift valleys were generally agreed to be structurally continued in the Red Sea, to the north, and since the Red Sea was isostatically compensated, Wegener was able to argue that the width of the rift at its northern end had allowed infilling with more dense material from depth, something that had not yet happened farther south within the margin of the African continent.4

  The case of the Rhine Graben shows gravity data incompatible with a simple down-dropped block of crust. On the other hand, the East African rift valleys, admitted by Sueß to be tensional structures, could now find an explanation where they had none before, and the two sets of phenomena could be united within the framework of the theory of continental displacements. In both cases the argument for rifting and lateral displacements was conformable with the existing data and resolved anomalies in the previous explanation.

  The next geological example Wegener chose, “Atlantic and Andes,” took up the most famous aspect of his displacement hypothesis: “The large-scale parallelism of the Atlantic coasts is not to be underestimated as an argument in favor of the supposition that these constitute the margins of an enormously widened rift. A single glance at the map suffices to establish that where mountains lie in the East, such are also found in the West, and where such are absent here, they are absent there as well.”5 Wegener’s approach, highly schematic, began with the circumpolar regions: where there are mountains in Greenland, we find them in Scandinavia. Farther south, where there are long sections of North American coast with no mountains, the same condition prevails in Europe. Farther south still, the great concave shapes of the Caribbean, as well as its European counterpart the Mediterranean, form one long, transverse sea-lane. The great South American tableland matches the great African tableland, and where the coast straightens out farther south in South America, it does so in Africa as well.6

  In the North Atlantic, in “the parts of the world best known to us, namely Europe and North America, there prevails almost perfect agreement regarding the details as well.”7 Eduard Sueß’s great work had pursued these continuities in relentless detail right down to the mineralogy of the gneissic rocks on both sides of the Atlantic, matching those of the Lofoten Islands and Hebrides in the East with those of Greenland in the West. Wegener was following Sueß down the Atlantic coast, and one can sense his excitement in his point-for-point substitution of his interpretation for that of the acknowledged master of this material.8

  From Wegener’s point of view the most striking data were maps, published in a number of geological memoirs and books since the later 1880s, showing the continuation of individual and distinct mountain chains across the Atlantic Basin. Different authors had given different names to these ranges—Caledonian, Hercynian, Armorican, Alpine, Altaide—but most observers agreed that between the mountains of Scotland and those of Nova Scotia, between those of northern Europe, Great Britain, and Ireland and the American Alleghenies, and between the mountains of North Africa and those of the northernmost tip of South America, there was demonstrable continuity in structure and mineralogy. European geologists had concluded that these mountain segments were once continuous.

  There was already a very strong argument against the idea that large sections of these mountains had subsided along with the rest of some former Atlantic continent. Albrecht Penck, the great glacial geologist, had pointed out that the missing segments of such mountain ranges would then be thousands of miles longer than their remaining segments, and a former continuity across the abyss would render them the longest mountain ranges on Earth.9 Viewed in this way, Wegener’s notion of an Atlantic rift gave mountain ranges of plausible extent while employing the same geological evidence.

  One may note that the map of the Atlantic in Andreé’s Atlas which had so inspired Wegener in 1910 did not show him what he might have seen in a different presentation: that the depth data for the Atlantic floor of the Challenger showed no transverse structures, while clearly showing the north-south extent of the Mid-Atlantic Ridge. This casts further light on the question of priority between the American Frank B. Taylor and Wegener in the matter of continental displacements. Taylor’s map of the Atlantic floor, using data from John Murray, gave no hint of suboceanic mountain ranges east and west across the Atlantic floor. Had Wegener come upon Taylor’s work while still composing his paper, he would likely have employed this important piece of data in his argument for displacements.10

  More abundant than this structural evidence of former continuity across the Atlantic abyss was the massive compilation of paleontological data suggesting such connections. A former continuity of fossil forms, followed by discontinuity at some later time, might have several explanations: “It is easy to see that these questions are completely independent of whether one accepts the horizontal displacement of continental blocks or believes in a subsidence of land bridges.… Portions of one and the same continental platform can also become faunally and floral distinct through shallow transgressions, and the decision will often be difficult as to whether we are viewing a rifting, or a separation via transgressing seas.”11

  Wegener’s consideration of the paleontological data concentrated, as one might expect, on the area of his original inspiration: the former Africano-Brazilian continent, sometimes known as “Archhelenis.” His treatment of the evidence here suggests his strong dependence on Theodor Arldt and Emanuel Kayser, and through their work he asserted, in a few brief paragraphs, the existence of a general consensus on a former land connection across a broad front (and thus not a narrow “land bridge” or “isthmian link”) between Brazil and Africa throughout the Mesozoic.
There was also general (though not unanimous) agreement that this connection had ruptured in the Tertiary, at the end of the Eocene.

  When Wegener turns his attention to the paleontological evidence of the North Atlantic, one begins to see his larger view of how the separation of the Atlantic had taken place. Wegener concluded that Atlantic rupture opened from south to north, appearing first in southern South America and southern Africa, with the fission of the North Atlantic somewhat later than the South Atlantic: “According to our view, North America, Greenland, and Europe would have constituted a continuous block up into the ice age, and the ice cap would have had a much smaller perimeter than one was heretofore obliged to accept.”12 This geological story was “reconfirmed” by paleontology, with species persisting across the North Atlantic continents while divergence of faunas was already evident between Africa and South America.

  Moving quite briskly, Wegener turned to his consideration of the Andes. These mountains had been left unexplained by Sueß, who had, as we noted above, a complicated theory of the origin of fold mountains in the “overthrust of a foreland subsidence,” a theory that the phenomena of western South America could not support. Wegener’s interpretation is exactly what one might expect:

  As the upfolding of the Andes is essentially contemporaneous with the opening of the Atlantic Ocean, then the concept of a causal connection between them is provided from the start. The American blocks, in their drift to the West, would have encountered opposition from the certainly very old and no longer very plastic floor of the Pacific, whereby the vast shell, which once formed the western margin of the continental block, with its thick sediments was thrust together into fold mountains … the folding of the Andes in no way need be equivalent to the full breadth of the Atlantic (about 4000 km).13