Drifting continents

Coastlines that fit

Look at a map of the Atlantic Ocean and the thought comes to nearly everyone: the east coast of South America and the west coast of Africa look like pieces of a jigsaw. Many people noticed it. The Flemish mapmaker Abraham Ortelius remarked on it in 1596 and guessed that the Americas had been “torn away” from Europe and Africa by earthquakes and floods. Francis Bacon noted in 1620 that the match between the coasts was “not to be attributed to mere accident,” though he did not say what it meant. In 1858 Antonio Snider-Pellegrini drew maps of the Atlantic closed and open. An American geologist, Frank Bursley Taylor, argued in 1908 that continents creep over the surface of the earth.

The man who turned the thought into a theory was a German meteorologist. Alfred Wegener, born in Berlin in 1880, held a doctorate in astronomy and spent his career on weather balloons and Arctic expeditions. In January 1911 he wrote to his fiancée, Else Köppen, that the east coast of South America seemed to fit “precisely” into the west coast of Africa, and that he would have to follow up the thought. He did nothing more until the autumn, when he read a survey of fossils that linked Brazil and Africa. On 6 January 1912 he presented his idea to the Geological Association in Frankfurt, and in the spring of that year it appeared in print. In 1915 came his book The Origin of Continents and Oceans, revised in 1920, 1922, and 1929. He proposed that the continents had once been joined in one supercontinent, which he called the Urkontinent; the name Pangaea became standard in the 1920s.

Wegener did not rely on the coastlines alone. He pointed out that the fit is better at the edge of the continental shelf than at the shoreline, and in 1965 a computer fit at the 500-fathom depth contour confirmed it. He collected three kinds of evidence, besides the coasts themselves:

• Matching rocks and mountains. The Appalachians in North America line up, across the closed Atlantic, with the mountains of Scotland and Scandinavia, which are of the same age and type.
• Ice-age deposits in the wrong places. The great glaciation of about 300 million years ago left rocky debris, called tillite, in southern Africa, South America, India, Australia, and Antarctica, lands that are now far apart and mostly warm. If the continents are assembled around the South Pole, one ice sheet covers them. Many geologists later regarded this as Wegener’s strongest argument.
• The same plants and animals on separate continents. The fern-like plant Glossopteris grew in India, South America, Africa, Australia, and Antarctica. Mesosaurus, a small reptile of shallow water about 275 million years old, left fossils in Brazil and in South Africa, and it could not have swum an ocean. Lystrosaurus, a stocky animal of 250 million years ago, turned up in Africa, India, and later in Antarctica.

Why most geologists said no

Wegener’s theory was widely discussed, and then widely rejected, especially in the United States. Its weakest point was the engine. He proposed that a force from the earth’s spin, the “pole-fleeing force,” pushed the continents toward the equator (tides were also invoked). In 1924 the Cambridge geophysicist Harold Jeffreys showed that such forces were far too weak to move continents through a rigid sea floor; and he was right about Wegener’s forces. In 1926 the American Association of Petroleum Geologists held a symposium on the theory in New York. Rollin Chamberlin of the University of Chicago wrote that, to believe Wegener, we must “forget everything which has been learned in the last 70 years.” Most American geologists, in the words of one historian, treated the idea as bad science.

Some did not. The South African geologist Alexander du Toit and the Englishman Arthur Holmes defended it; du Toit published his comparison of South America and South Africa in 1927. In 1928 Holmes proposed the mantle itself might churn, like a pot on a stove heated by radioactivity, carrying the continents along. It was the right sort of idea, but there was no way to test it. Wegener kept on, and went back to Greenland in 1930 to study the climate. On 1 November, his fiftieth birthday, he set out from the central ice-cap station, Eismitte, with a young Greenlander, Rasmus Villumsen, to return to the west coast. He died on the ice in mid-November, probably of heart failure from exertion. Villumsen buried him, went on alone, and was never seen again. In May 1931 a search party found his grave, marked by skis, about 190 kilometers short of the coast. His widow left him there.

Listening to the sea floor

The turn came from the deep sea, and from the Second World War. Before the war nobody had mapped the sea floor in any detail. The war put echo sounders on many ships, and gave scientists a way of measuring the earth’s magnetism from a moving vessel, because aircraft and ships needed to find submarines. (Radar is sometimes credited, but radio waves do not travel far through seawater: the tools were sonar and magnetometers.) Harry Hess, a Princeton geologist, commanded a Navy transport in the Pacific, and kept his echo sounder running as the ship crossed the Pacific. He found flat-topped underwater mountains, which he called guyots, and he returned to Princeton with a feel for the shape of the sea floor.

The picture that mattered was drawn by hand. At Columbia University, Marie Tharp turned the lines of depth soundings gathered by her colleague Bruce Heezen into maps of the Atlantic floor. In 1952 she noticed that the ridge in the middle of the ocean had a deep V-shaped notch down its crest, a rift valley. Heezen did not want it to exist, because it looked too much like continental drift; as she recalled it later, he dismissed it as “girl talk.” He came round when the earthquakes of the Atlantic, plotted by a colleague, fell along the rift. Women were not allowed on research ships; Tharp first went to sea in 1968. Together they published the first map of the North Atlantic floor in 1957.

Then, in the 1960s, the pieces fell together. In 1960 Hess circulated a paper (published in 1962) arguing that new sea floor is made at the ridges, where hot rock rises and cools, and moves away on both sides as on a conveyor belt, until it sinks into deep trenches at the edges of the oceans. He called it “an essay in geopoetry.” Robert Dietz gave it the name “sea-floor spreading.” In 1963 two young Cambridge researchers, Fred Vine and Drummond Matthews, published the test in the journal Nature: as the new rock cools it records the earth’s magnetic field, and since the field has reversed many times the sea floor should be striped, symmetrically on both sides of a ridge. A Canadian, Lawrence Morley, had the same idea at the same time, submitted it in February 1963, and was turned down by two journals. In 1965 J. Tuzo Wilson described faults that join one ridge or trench to another, which he called transform faults. In December 1966 a magnetic profile across the Pacific–Antarctic Ridge, taken by the research ship Eltanin, showed the pattern so clearly that it persuaded many on sight. In 1968–69 the drilling ship Glomar Challenger found, as spreading predicts, that the sediment lying directly on the basalt gets older with distance from the Mid-Atlantic Ridge.

In 1967 and 1968 three independent groups, Dan McKenzie and Robert Parker, W. Jason Morgan, and Xavier Le Pichon, showed that the earth’s surface behaves like a mosaic of rigid plates. The new theory was plate tectonics. It explained more than drift ever did. Continents are not plowing through the ocean floor; they ride on plates that carry both. And the plates are not driven by the pole-fleeing force but by the slow churning of the hot mantle below, in which the weight of old, cold plates sinking at trenches is usually thought to play the biggest part (the details are still debated).

A planet of plates

The earth’s outer shell is broken into plates, usually counted as seven large ones (Pacific, North American, Eurasian, African, Antarctic, South American, and Indo-Australian) and many smaller ones. They move at a few centimeters a year: the Atlantic widens at about 2.5 centimeters a year, and the fastest ridge, the East Pacific Rise, spreads at more than 15. Satellite and GPS stations now measure the motion directly. Oceanic plates are young, because they sink back into the mantle: the oldest large patches of sea floor are only about 180 to 200 million years old. The continents are far older, with rocks as old as 4 billion years.

The plates explain the map of the continents’ past. About 152 million years ago, South America, Africa, India, Antarctica, and Australia were still mostly joined in the southern supercontinent Gondwana, and the Atlantic had begun to open to the north. India broke away about 130 million years ago and raced north at 15 to 19 centimeters a year, an extraordinary speed for a continent. About 50 to 55 million years ago (estimates range from about 59 to about 34 million, and the dating is still debated) it met Asia. It slowed to about 4.5 centimeters a year and has been pushing ever since, raising the Himalaya and the Tibetan Plateau. The Himalaya are still rising, by about three millimeters a year, and Mount Everest was measured at 8,848.86 meters by China and Nepal together in 2020.

It is often said that the rise of Tibet gave central China its climate: at about 30 degrees north, the belt of the world’s great deserts, the Sahara and Arabia, runs through Cairo (about 25 millimeters of rain a year), while the same latitude in China holds Wuhan (about 1,300 millimeters) and Chongqing (about 1,150). The dry belt itself is a result of the circulation of the atmosphere and does not move because of Tibet. Scientists still debate how much of central China’s green and wet climate at that latitude comes from the plateau, and how much from the East Asian monsoon itself; both are real, and the plateau sits in the path of the high-altitude jet stream, which shapes the rains. A mountain range changes where moisture goes. It does not move the desert belt.

Six pictures of one planet

This course began with one photograph, so it should end with a few more. Look back over the lessons. There are six ideas, each of which changed what “the earth” meant:

Look at the pattern. Each idea first came from reasoning about something other than looking: an eclipse’s shadow, a pendulum’s rate, a bed of rock. Each was settled much later by an instrument: a heliometer, a pendulum, a sonar, a mass spectrometer. And several of the key steps were taken by people working across the borders of fields: a meteorologist, a physician and farmer (Hutton), a clerk who taught himself astronomy (Bessel), a geochemist (Patterson). The earth was the first thing humans lived on, and one of the last things they understood.

Then, in the second half of the twentieth century, the earth was photographed whole. The first crude picture from orbit was taken in 1959, and Yuri Gagarin in 1961 saw the curve of the horizon. The first people to see the whole earth at once were the three astronauts of Apollo 8, in December 1968. On Christmas Eve, orbiting the moon, they saw the earth rise over the lunar horizon, and Bill Anders photographed it. Later that day they gave a television broadcast from lunar orbit, in which they read the opening verses of the Book of Genesis and ended with a wish to “all of you on the good Earth.” NASA estimated that hundreds of millions of people, perhaps a billion, were watching or listening; it was the largest television audience up to that time. The older picture of the world and the newer one were, for a few minutes, in the same room. On 7 December 1972 Apollo 17 took the photograph known as the Blue Marble from about 29,000 kilometers away. And on 14 February 1990 the space probe Voyager 1, at Carl Sagan’s urging and by order of NASA’s administrator, turned around from a distance of about 6 billion kilometers and photographed the earth as a point of light less than one pixel across. Sagan wrote that it was “a mote of dust suspended in a sunbeam.”

What do such pictures mean? This course takes up a thought of the philosopher Martin Heidegger. In a 1938 lecture, later published as “The Age of the World Picture,” Heidegger argued that the modern age is the age in which the world is conceived and grasped as a picture: set before a human subject as something to be represented and mastered. In an interview in 1966 (published ten years later) he said he was shocked when he saw pictures of the earth taken from the moon. In this reading, the whole-earth photograph is the world as picture in its purest form, and our science of the earth, with its calculations and its measured sizes, ages, and speeds, expresses a will to bring everything before us. It adds that in earlier worlds, such as the Greek cosmos or the Christian creation, people lived inside a small world and did not set out to master it; in ours, the world is as big as our ambition.

That is one reading, and not the only one. Many who saw the photographs from space felt something close to the opposite: astronauts have long described an “overview effect,” a sense of the earth’s fragility and of the thinness of the air, and Anders is widely quoted as saying that the most important thing about going to the moon was that we discovered the earth. The cultural geographer Denis Cosgrove argued that whole-earth images carry a particular Western way of seeing; others argue that they helped create the shared sense of care that showed itself in the first Earth Day, in 1970, and in much that came after. Whether the picture is a sign of mastery or of humility depends partly on what we do with it.