On Sunday afternoons, often a social time at the joint household, Alfred would retire to his study after the midday meal, leaving Kurt to play with the children. Else made it clear in her memories of this time that Alfred was not in any case very good at playing with the children, and that for him social time and small talk were generally considered time wasted.54 Johannes Georgi, Wegener’s younger colleague at the observatory, noted that when he was a Sunday guest at the Köppen household, Wegener invariably excused himself “to attend to urgent work in his study,” something that Georgi very much admired.55 This “urgent work” was his work on continents and oceans, for which he had to seize every spare hour.
We know fewer of the details of the collaboration between Wegener and Köppen in 1919–1920 than in any previous period; shared residence and daily contact made letters unnecessary. We do know that Köppen became so deeply involved in the book that Wegener would write in the foreword to it that “several chapters grew out of an exchange of ideas with him [Köppen] so close and continuous, that it is no longer possible to partition credit.”56 He went on to say that “just as the first edition was advanced through the generous geological advice and collaboration of Cloos, so is the second characterized by the no less valuable collaboration of a climatologist; its development proceeded in a daily exchange of views with W. Köppen.”57 We know from Else’s recollections of this time that “it was an uncommonly productive collaboration. Many days my father waited impatiently for Alfred’s return from the Observatory so that he might discuss a newly discovered argument with him. At this time he always carried a small globe of the Earth in his pocket, so that he might at any time check his reflections on pole positions and climate zones.”58
The small globe that Köppen carried in his pocket at all times in 1919–1920 is a key to a central difficulty in the debate about continental displacements in this period. Paradoxical as it may seem, geologists (Wegener’s fiercest opponents) lived and worked on a flat or nearly flat Earth. Geologic mapping proceeded on a much finer scale than geographical reconnaissance; this explains to some extent why geographers were easier to convince than geologists: their sense of adequate work was not so aggressively fine-grained.
Thinking about the whole Earth as a planetary body was relatively new to geology, even in 1900. The first major geology text in the United States to consider the origin of Earth as a geological topic was Chamberlin and Salisbury’s Geology, published between 1905 and 1907.59 One of the striking novelties of Edward Sueß’s Das Antlitz der Erde had been his insistence on viewing Earth from space and seeing Earth’s surface on a continental scale, emphasizing the continuity of Earth’s largest rather than its smallest surface features. This sort of presentation stands in marked contrast to the work of someone like Oswald Heer (1809–1883), who, like Sueß, worked in great detail on the geology of Switzerland. Heer’s most popular work, Die Urwelt der Schweiz (1865), contained stunning lithographs of the appearance of Switzerland in different geological eras, complete with tropical plants and animals and sometimes oceans. These scenes are, however, always viewed from the vantage of someone standing at ground level; it is the world seen from the surface of Earth, not seen by a “visitor from space.”60
Geological fieldwork was so tightly constrained by concern for detail that the curvature of Earth rarely came into play; nevertheless, in the margins of geological and topographical maps there are lines called “Mercator grid ticks,” which show the offset of the position on a spherical Earth from the position on a Mercator map of Earth. For the lower and middle latitudes, within 45° of the equator, these differences are rather small, but they increase dramatically at higher latitudes, becoming infinite at the poles. This unreflective commitment by geologists to the Mercator map created real problems for them in visualizing Wegener’s argument. It turns out that a number of objections by geologists to Wegener’s continental reconstructions, as well as a number of their confusions in understanding his scheme of continental motions and reconstructions, came from their automatic resort, both physically and in memory, to the “Mercator world.” To understand this, we have to stop for a moment and discuss the problem of map projections.
To portray the spherical Earth on a plane surface, one struggles to balance the preservation of true angles (conformal mapping) with the preservation of the true surface (equivalence mapping). These two properties cannot be perfectly attained in the same projection, which is why there are different projections for different purposes. The Mercator map, introduced by Gerardus Mercator (1512–1594) in 1569, was a revolutionary aid to navigation and is the ultimate conformal map, with every aspect of projection sacrificed to true angular measurements and thus true direction. It allowed sailors on a spherical Earth to navigate dependably while charting their courses using compass directions on a flat map. To accomplish this, Mercator drew the meridians of longitude so that they would be equally spaced and did not converge toward the poles. Latitude and longitude were then in a rectangular grid. However, this progressively distorted the area of land surfaces at high latitudes, making them many times their true area. This mattered little for most European navigation when it was largely confined to midlatitudes and especially the Atlantic. If you kept to your course on a Mercator map, you would eventually arrive at the destination you chose, but because of the distortion of areas at higher latitudes, you might be some days off in your estimate of when you would arrive.
This Mercator map is even today the most familiar of all projections, and it is also the reason why most children think that Greenland is the same size as South America (which is how it appears on the Mercator map), when actually it is the size of Mexico. The Mercator map’s smaller distortions in midlatitudes, in contrast, give a fairly accurate representation of the land surface of the United States, southern Canada, most of Mexico, and most of Europe and the Mediterranean. It can therefore be employed for a number of purposes, including the portrayal of climate zones in the most familiar inhabited areas. When Köppen published his epochal Versuch einer Klassification der Klimate in 1901, with the climates of Earth defined in terms of groups of plants and animals found at different latitudes, he used the Mercator map to delineate these latitude zones.61 Köppen’s mental world was also a “Mercator world”—thus his need for constant resort to the globe in his pocket in 1919: he had to learn to see Earth differently, and to do this, he had constantly to look at a physical globe instead of the Mercator map firmly lodged in his consciousness and memory.
Wegener was trained as an astronomer, and the image of Earth that always came first to his mind was a rotating sphere revolving around a star. Because of this, the map of Earth which Wegener carried in his mind and used in thinking about Earth was entirely different from the Mercator projection: Wegener (in 1920 and after) always drew continental and paleocontinetal positions using a very different projection: the “azimuthal equal-area projection” constructed in 1772 by the German philosopher, mathematician, and astronomer Johann Heinrich Lambert (1728–1777). Lambert had been something of a hero to Wegener in his university days. Self-taught in science, Lambert studied comets and planetary astronomy, was interested in meteorology, and delighted in mechanical devices—all interests Wegener shared.62 Wegener had argued, in his brief history of cosmology, written while still a student, that Lambert should have had a full share of credit for the origin of the nebular hypothesis, most often attributed to Kant and Laplace.63 Wegener’s reasons for choosing Lambert’s map projections were, however, entirely scientific and unsentimental.
Once again, to portray the spherical Earth on a plane surface, one struggles to balance the preservation of true angles (conformal mapping) with the preservation of the true surface (equivalence mapping). The Lambert azimuthal equal-area projection that Wegener favored is an equivalence map with two outstanding characteristics: all geographic directions from the center of the map are true directions (thus “azimuthal”), and all the area measurements at all latitudes and longitudes on the map correspon
d to the real surface areas on the spherical Earth (thus “equal area”). The one parameter sacrificed in this projection is true distance: distances are compressed at the limits of the plane circle representing a hemisphere of Earth. Lambert also constructed an “azimuthal equidistant projection” for use when distance was more important than area.
Not only was this the first map that came to Wegener’s mind, but it is also the projection he saw in the atlas in 1911 which inspired his idea of continental displacements. The original inspirational pages he had seen in Andree’s atlas had three versions of this map: the main map was the Equatorial Aspect of the Lambert Azimuthal Equal Area projection centered on longitudes 110° west and 70° east (to put the oceans at the margins and minimize distortion of continental surfaces). Along with this map, in the margins of the same page, Andree had used the Hammer Elliptical (Lambert) Azimuthal Equal Area projection centered on the Greenwich Meridian and two aspects of the Lambert Oblique Azimuthal Equal Area centered on the Greenwich Meridian and 50° north latitude (Berlin) and the facing hemisphere on 180° longitude and 50° south latitude.
Wegener, in both 1912 and 1915, had made a fundamental error in assessing his audience. He had thought that he had to teach them geophysics; what he actually had to teach them was how to read a map properly, as well as how to read the proper map. Köppen had been able to demonstrate—in his Hamburg address and in his article in Geographische Zeitschrift—that one could appeal to authority with regard to geophysics, rather than try to teach it to geologists who would or could not learn. On the other hand, it was absolutely essential, in order to represent in two dimensions what happened on a three-dimensional Earth, that the directions of motion of the continents be true directions and the mapped area of the surfaces of former continents be their true areas. Köppen became convinced, and in turn convinced Wegener, that this insistence on the proper visualization of the planet would be the key to persuading the audience. Wegener’s method, even in his own description, was “morphological-empirical,” and it was in this arena that the battle would actually be fought.
Nothing made this clearer than the errors of Carl Diener in his wartime criticism of Wegener’s book. Diener had accused Wegener of using a continental reconstruction in the Northern Hemisphere which had created a 35° separation between Russia and Alaska across the Bering Strait, and in another part of his argument Diener had misplaced the North Pole by 90° relative to Wegener’s reconstruction of the gathering of continents around the South Pole, before their dispersal in the Carboniferous and Permian. Wegener demonstrated to Köppen, first in a letter and then in person in early 1918, that Diener’s visualization of the situation could only have been created by someone looking at a Mercator map, someone who saw Earth distributed on a rectangle, rather than a globe, with consequent distortions of area. It was at this point that Köppen actually understood the significance of map projections when considering continental displacements, so much so that he added a section to his manuscript to be published in Geographische Zeitschrift showing that Diener’s accusations—that Wegener had made absurd claims about continental motions—were actually themselves embarrassing errors by someone looking at a Mercator map who could not visualize Earth as a globe.64
Edgar Dacqué had anticipated these considerations already in 1915, though too late for Wegener to incorporate them into his own book. Chapter 9 of Dacqué’s book on paleogeography was titled “The Design of Paleogeographic Maps and Their Details.”65 In the final section, “Principal Objections against [different] Paleogeographic Maps and Their Data,” he had remarked, “I direct your attention to the theories of continental displacement treated extensively in Chapter IV. Even if such theories are not demonstrably true, but merely arguable, they show already what intricacies and especially what precautions are involved in the design of paleographic maps.”66
He went on to spell out his concern that however suitable the Mercator projection for the demonstration of faunal and floral latitude zones, when discussions moved into polar regions, it was preferable to use equal-area maps, and this was of crucial importance when considering theories of displacement of the poles and displacement of the continents. Even here, not all equal-area maps would be equally suitable, and in consequence, “the best procedure is and will remain, especially for educational purposes in lectures and seminars, the simultaneous use and comparison with the globe, always with the realization, naturally, that neither the equator, nor Earth’s poles are in the same constant positions throughout Earth’s history, and that neither a globe, nor maps that represent the world in the present, can accurately and absolutely represent all paleogeographic details.”67
This idea was foremost in Wegener’s mind in preparing the second edition. When he came to write the introduction to the book in April 1920, after six months of continuous work, he gave this final bit of advice to his readers: “The reader is emphatically advised that a large number of questions—if one does not want to give up one’s own judgment [i.e. does not wish to resort to the arguments of others to make up one’s mind]—certainly require the use of a terrestrial globe. An atlas is not sufficient because of the distortions caused by the projections used. The criticism of the first edition suffers throughout from a failure to refer to the globe.”68
The tone of the criticisms of the first edition was not as troublesome to Wegener as that the majority of objections raised by his (intemperate) critics were based on misunderstandings. Now, in 1919, the major problem for Wegener was that these misunderstandings, misreadings, and inaccurate descriptions of his work had been carried over wholesale into a variety of subsequent reviews. This was quite serious, and he found himself accused of gross errors of fact by men who were being treated (by a second round of reviewers) as authoritative critics, when the errors involved were those of the critics themselves, almost all attributable (as Wegener saw it) to their failure to evaluate the arguments using proper paleogeographic maps, or by reference to a globe of Earth.
While Wegener could not make his critics agree with him, he could make the book easier to understand, make the argument easier to follow, and reorganize its elements in a way that might fit the conceptual world of his readers better than the structure he had chosen in 1915. Assuming that a new edition of the book would produce both new supporters and new critics, at the very least he could try to ensure that these critics would have to disagree with what he had actually said about continental motions and pole displacements on a spherical Earth, not on a Mercator map. Thus, at every junction in his argument, he would strive to remind his readers that they should be thinking of lateral displacements on a globe, not on a flat map.
Given the inherent complexity of his argument (much in the spirit of Darwin’s Origin of Species, the book was “one long argument”), Wegener decided to produce a very detailed analytical table of contents. The previous edition had a table of contents merely twelve lines long, and some of the chapter titles were but a single word: “Splitting,” or “Gondwanaland.” In the second edition, the argument was laid out in full detail in the table of contents. Nearly every page in the book would have its own subhead title, referenced to the page or pages on which that material appeared. The second edition has 135 pages, and there are 114 titled subheadings in its table of contents. This detail would provide an overview of the work for new readers, but it would also alert his critics (and supporters alike) to the location of the revisions and corrections he had made in light of previous criticisms and suggestions.
To further the aim of making the book’s argument transparent and easy to follow, Wegener also decided to include a comprehensive index of authors, concepts, and geological place-names. Of his previous works, only Thermodynamik der Atmosphäre had an index; the circumstances of composing Wind- und Wasserhosen in Europa and the first edition of Die Entstehung der Kontinente und Ozeane—both in wartime—had persuaded him to proceed to print without an index in each case. This had definitely been a mistake, especially in the latter book, in whic
h he sometimes covered the same material in four or five different chapters, leaving the reader to page through the book to try to remember where an argument or concept had appeared before.
In preparing this revision, Wegener also made a radical change in his mode of work. He had corresponded with Köppen about everything he wrote from 1908 onward, but he had worked on the first papers on continental displacements in 1912 almost entirely alone, and he had worked through the first edition of his book in 1915 on the topic with a single collaborator, Hans Cloos, who had given him a firmer footing in the literature of early twentieth-century geology. However, in all the other areas from which he had drawn evidence for his hypothesis, he had depended on his own reading, with mixed results. To remedy this obvious fault, he recruited several scientists to read through his manuscript as he prepared it and to offer criticism in advance of publication.
He first recruited Edgar Irmscher, close by in Hamburg and already a supporter of his hypothesis; Irmscher could review anything having to do with the distribution of plants. Wegener also engaged the cooperation of Ernst Tams (1882–1963), a young seismologist, also at Hamburg, then on his way to becoming one of Germany’s leading geophysicists. Tams was pioneering the use of travel times of seismic waves to show the difference between continental and oceanic crust. Additionally, Wegener got the enthusiastic support of Wilhelm Michaelsen (1860–1937), an assistant at the Zoological Museum in Hamburg, about to be promoted to its curator. Michaelsen was an expert on the distribution of earthworms in the present and the past, and he became intrigued by the possibility that Wegener’s hypothesis of continental drift could help explain how so many earthworm species or closely related species could exist on widely distributed continents, while not being able (obviously) to traverse abyssal oceans in order to get there.
Alfred Wegener Page 75
