On the origin of continents

Nothing in geology makes sense except in light of plate tectonics: the idea that mobile plates float on top of a thick layer of molten rock surrounding the Earth’s core, and currents in that rock move continents and the ocean floor. Plate tectonics explains earthquakes, the formation of mountains and deep sea trenches, and the fossilized sea creatures found miles from the ocean. By explaining so much both within and outside its field, plate tectonics stands as a sort of Platonic ideal of what a scientific paradigm can achieve.

In other ways, plate tectonics is a very unusual paradigm. It was discovered twice: first intentionally by geologists in Europe and its colonial sphere, and second, accidentally, by North American geophysicists. The two paths offer opposite lessons on the secret of scientific discovery. There is no lone genius who can claim the majority of the credit. The theory went from fringe to total acceptance over the course of ten years, with nary a funeral in sight. And it all came together much later than one would think: around the time of the first moon landing in 1969. This is the story of how two separate groups of scientists near-independently discovered plate tectonics with wildly different approaches, and what this can teach us about paradigm shifts.

Moving mountains

Like many scientific concepts, the idea of mobile continents had been proposed repeatedly. Abraham Ortelius, a sixteenth-century Dutch cartographer, is widely credited as the first person to suggest that the continents had not always been in their current positions. In 1596, he noted the match between the eastern coastline of South America and western coastline of Africa, and proposed that they had been separated by a great flood.

But credit for the idea of continental drift is generally given to meteorologist and polar explorer Alfred Wegener. In 1912, he published an article titled ‘Die Entstehung der Kontinente’ (The Origin of Continents) laying out his case. He expanded on his theory with a book in 1915, while serving in the German military’s weather service. Revised editions followed in 1920, 1922, and 1929; the third edition was translated into English and French in 1924, kicking off international discussion. 

If he wasn’t the first, why is Wegener credited as the origin of the theory of continental drift? The reason is that he was the first to assemble thorough evidence: not just from geology and cartography, but also geodesy (the shape of the Earth, including magnetic and gravitational shape), paleontology, and modern biology. Wegener was a synthesist who brought together information that specialists had ignored.

Wegener’s most famous piece of evidence was that Brazil fit neatly into Africa, just as Ortelius had noted four centuries before. But Wegener went beyond the shape of the coastlines. When fit together, geological formations like the Karoo (in South Africa) and Paraná Basin (in South America), or the Cape Fold Belt (in South Africa) and the Sierra de la Ventana (in Argentina), also lined up, albeit imperfectly.

Second, Wegener pointed to geological evidence that South America, Africa, India, and Australia had all been traversed by glaciers. Yet none of these continents could possibly have hosted glaciers at their current latitudes. Even if the world had been cold enough, there wasn’t enough water in the oceans to support that much ice. These observations would be much easier to explain if the continents had historically been adjacent to each other and closer to a pole.

Third, it was already well known before the time of Wegener’s publications that certain fossils of the same species, both animal and plant, were found on multiple continents, dispersed thousands of miles from each other in unexplainable patterns. This finding was previously ascribed to the mysteries of God, but after natural selection had become widely accepted by scientists, it was still an unsolved puzzle.

Further, Wegener noted that coal and fossilized corals are found in modern North America and Europe, even though these continents are too cold to support coral today, and that desert sandstones are found in modern wet environments in Britain, continental Europe, and the northeastern United States. Finally, he pointed to astronomical measurements of Greenland that suggested the island was moving away from Europe at a rate of ten meters per year (a rate later shown to be overestimated by a hundredfold).

Putting all this together, Wegener proposed that the continents had drifted apart over time, and that the bottoms of continents moved through the ocean floor like icebreaker ships through frozen oceans, powered by tidal forces and the Earth’s spin. His theory was rejected for reasons of varying quality. One was that Wegener was a meteorologist, and geologists didn’t like his intrusion into their field. His proposed mechanism of drift, they argued, was impossible given the relative softness of continents and ocean floor, and much of it could be explained away by the long-term change in continents’ climates.

But climate change couldn’t explain the fossil evidence across continents. The leading explanation at the time was that there had once been long stretches of land uniting the continents that had disappeared without a trace, even though the proposed mechanisms for such disappearances ranged from shaky to nonexistent. Disappearing land bridges are hard enough to explain geologically, but even more so biologically. Over generations, species adapt to their environment. So, if identical fossils appear in different places, it could suggest either that their climates were fairly similar or that the animal’s lineage had adapted to a wide range of climates as they slowly passed through. But if these animals were so adaptable, why wouldn’t they have spread to multiple climates within continents as well? You could instead argue that the climate at the time was near-identical across the land bridge and its anchors on either side, but distinct from nearby locations on each continent, despite the change in latitude and peculiarities of coastal ecosystems.

I sympathize with the Earth scientists who rejected the hypothesis of drift, because the idea that continents move feels crazy. But my inner biologist revolts at the logical contortions needed to argue that land bridges sufficiently explained the fossil evidence, as much as if Earth scientists proposed these specific animals grew wings to cross the Atlantic and immediately lost them upon landing on the other side. 

Eppur si muove

In the early twentieth century, most Earth scientists dismissed the theory of continental drift. In North America, the case appeared closed after the 1926 symposium of the American Association of Petroleum Geologists, where scientists cemented consensus against it. In Europe, however, a fraction of scientists continued to investigate, and in Europe’s colonies and former colonies, a smaller number, but larger fraction, of Earth scientists also became adherents. ¹

What explains this geographic split? In the first volume of The Continental Drift Controversy, historian Henry Frankel attributes the difference in reception to differences in local geology. When Wegener proposed drift, geological notetaking was not standardized, and this remained the case for several more decades. Two scientists could go to the same field site and return with very different notes. This made geology an incredibly provincial science, with most geologists only accepting work from people they trusted or in areas about which they were well informed. Most geologists worked in a single location near where they lived, because rocks are heavy. If your local rocks didn’t have drift-shaped mysteries, and your friends’ rocks didn’t either, it was easy to dismiss the crazy idea that the continents moved. I find this theory very appealing, although Frankel doesn’t explain what, specifically, were the local geological problems of each area and why some were more or less explained by drift.

Additionally, drift’s biggest supporters tended to be the rare scientists who had traveled to multiple continents – something more common in Europe and its former colonies than in North America, although American geologists who traveled to Antarctica often did come to support continental drift. It seems plausible that the insights gained traveling explain some of the difference in receptivity to drift.

Whatever the reason for the dispute, these few true believers continued to chip away at the problem of continental drift, and by 1962 they had made substantial progress. Alexander du Toit, a geologist and Wegener’s biggest supporter, for example, had traveled from his native South Africa to multiple South American locations, gathering evidence of similarities in the radioisotopes and stratigraphy (rock layering) between South America and the regions of Africa that corresponded by map fit, which he published as Geological Comparison of South America with South Africa in 1927.

The following year, Arthur Holmes, also a geologist and Wegener’s second biggest supporter, proposed that continents were moved by convection currents in the Earth’s liquid mantle. We now know this to be the correct mechanism, although he incorrectly applied it to continents rather than the plates of the ocean floor. 

Then there was the paleomagnetic data, which revealed traces of the Earth’s past magnetic field. Magnetic fields have a direction: one end of a magnet is ‘north’ and the other ‘south’. But if a magnetic metal, such as iron or nickel, melts (for instance when ejected from a volcano), it temporarily loses its magnetic field, and resets in the same direction as its environment when cooled back into solid form. Scientists had known this since 1895; and it means that in the absence of a stronger magnetic field, rocks record the Earth’s magnetic field, but only weakly. In the 1950s, scientists developed instruments to detect that weak field, though early measurements were crude and rockhandling techniques had to advance before they could prevent their magnetic fields from reversing during research.

If continents and the Earth’s magnetic field were static, every rock sample from anywhere on the planet should point to the same pole. But that isn’t what scientists found. Samples from the same location, laid down at different times, pointed in dramatically different directions: sometimes differing by a few degrees, and sometimes by as much as 180! 

If the continents had always been in their present positions, this variation should not exist; if instead you assumed the continents had moved, you could derive their relative locations from the differences in the pole locations implied by their magnetic fields. By sampling in the same location at different depths (which would have acquired their magnetic fields at different times), geologists could draw fairly precise maps of the paths of a continent’s movement over time. 

Pulling this off was especially impressive because these geologists as yet had no idea how much the Earth’s magnetic field changed over time. The poles wander a few tens of kilometers per year, and dramatically flip a couple of times per million years or so. But this was unknown until geologists created the data that showed the paths of the continents.

As plate tectonics was finally taking off in the early 1960s, Harry Hess, a geologist, and Robert Dietz, a geophysicist, independently proposed ‘seafloor spreading’. This is the idea that the Earth’s liquid mantle is covered by solid plates. Currents in the liquid beneath move those plates apart, creating gaps between them. Those gaps are then filled by partially molten rock from the mantle, which quickly cools and solidifies, bonding to the edges of the existing plates. Their stance was speculative – Hess referred to his paper as ‘an essay in geopoetry’ – but as we’ll soon see, it was later shown to be almost exactly right. Their idea was more significant than just understanding the mechanism of spreading: they had also recognized that the ocean floor itself moved, rather than the continents moving about on top of it.

The evidence mounts

Two world wars passed as geologists worked on drift, and the Cold War was about to begin. The success of submarine warfare in the Second World War had the US Navy eager to pursue it further to improve submarine navigation. 

At the time, such navigation consisted of compasses (which provided limited data), dead reckoning (calculating current location based on approximations of past location, speed, and direction, which accumulates errors), and celestial navigation (which required surfacing and good visibility). The Navy hoped to supplement these methods with marine maps filled with landmarks in the magnetic field that navigators could use for orientation. Ideally, those landmarks could be detected silently, without emitting sonar and giving away navigators’ positions. To pursue this goal, the Navy invested heavily in marine geology; it provided nearly 90 percent of all academic oceanographic funding in the 1950s. And landmarks they found. Along the western coast of North America, two geophysicists detected stripes of alternating polarity, pointing to the north and south pole in succession, switching so often that, when drawn out, the stripes resembled a zebra’s coat.

Further investigation found similar striping at the Reykjanes Ridge (southwest of Iceland), the East Pacific Rise (west of South America), and the Mid-Atlantic Ridge (running down the Atlantic ocean). There was clearly a pattern, but caused by what?

This didn’t remain a mystery for long. Such patterns are exactly what you’d expect to see if Hess’s seafloor spreading hypothesis was correct. As magma rose between tectonic plates, it would cool, solidify, and take on the same magnetic orientation as the Earth’s current magnetic field. Every few hundred thousand years, when the planet’s magnetic poles flipped, new magma would take on the opposite orientation, producing regular stripes. Geophysicist Lawrence Morley realized this in 1963, but his letters to this effect were rejected by publishers. Instead, that same year, two other geophysicists, Frederick John Vine and Drummond Hoyle Matthews, published a similar theory of seafloor spreading to explain the zebra stripes. Vine and Matthews did not cite Hess, and might have come up with the idea independently. The idea of continental drift through seafloor spreading ultimately became known, rather longwindedly, as the Vine-Matthews-Morley hypothesis. 

Further reinforcement came over the next few years, as geophysicists published empirical and theoretical follow-ups detailing the movement of rigid plates over a sphere and using earthquake data to map the plate boundaries. By 1970, plate tectonics had reached near total acceptance, though there was only indirect evidence for how fast the continents were moving. More information would come a decade later, when astronomers developed ‘Very Long Baseline Interferometry’. This technology uses telescopes placed in different locations but aimed at the same quasar, a pattern of very bright light billions of light years away, to measure miniscule differences in the time it takes for light to travel from the quasar to each telescope. By comparing these differences over time, scientists have been able to estimate that the continents move at 1.5 to 5 centimeters per year – roughly the same speed at which human fingernails grow. A decade later, NASA provided further confirmation with thousands of GPS trackers that showed continents moving relative to each other. 

Geology’s theory of everything

Since measuring the speed of drift in the mid-1980s, geologists have built up a more detailed picture of the processes involved: more granular maps of plate boundaries, more complete models of the convection currents that move them, and an appreciation of how small these plates can be (hundreds of ‘microplates’ have now been identified, with most tectonic plate maps providing only a summary). But the basic principles remain unchanged.

The continents emerge from the continental floor, which together with the ocean floor forms Earth’s outermost solid layer. This layer is made up of giant plates that float atop a deeper layer of hot, slowly moving rock called the mantle. Heat from the Earth’s scorching hot core drives motion within the mantle, causing hot mantle rock to rise, spread sideways, and sink again.

Over millions of years, this slow drift has carried the continents vast distances. Around 200 million years ago, all of the continents formed a single landmass known as Pangea. Convection currents in the mantle gradually split this landmass apart, carrying the continents into their present locations. The discovery that Earth’s surface is divided into floating, mobile plates explained a number of baffling observations no one realized had the same cause. The matching shape of the South American and African coastlines, along with shared minerals and fossils, reflect how they were once united. Evidence of glaciers in now-tropical Africa reflect that it wasn’t always tropical, while coral grew in regions that are now temperate because they once lay at tropical latitudes. The Wallace Line – which separates neighboring islands between Asia and Australia that host completely different species – arose because those islands were once far apart.

Plate tectonics solved more than mysteries of location. The impact of plates crashing into each other also explains earthquakes, the formation of mountains, and the creation of deep ocean trenches. Where plates pull apart, molten rock rises to the surface as lava, forming volcanoes. As magma fills the gaps between separating plates and cools, magnetic stripes form, recording the Earth’s magnetic field. This paradigm has had enormous staying power. A North American student of geology from 1925 who time traveled to 1975 would find the field almost unrecognizable. A student who traveled from 1975 to today would find many new details to study, but the framework largely unchanged. 

Plate tectonics is a scientific paradigm that has two origin stories, and one could draw opposite lessons from each. The first story of plate tectonics is of the power of a visionary genius inspiring fifty years of research by scientists who sought evidence to shine light on a specific hypothesis and gradually refine it until they reached the truth. The second is of the power of giant piles of military money and former physicists to find surprises and turn those into theories.

But the more striking feature of plate tectonics is what was lost in the decades before it was accepted. While continental drift remained fringe, geologists were accumulating observations without the framework to connect them: fossils filed separately from magnetic stripes, mountain ranges explained without reference to ocean floors, earthquakes catalogued without a theory of what caused them. The matching coastlines of South America and Africa were known for centuries before anyone understood what they meant.

Wegener’s synthesis, and the geophysicists’ confirmation, revealed that each piece had been part of the same puzzle all along.