Plate Tectonics
The Earth’s own magnetic field settled an argument the coastlines could never win.

“Utter, damned rot!” said the president of the American Philosophical Society. A British geologist warned that no one who “valued his reputation for scientific sanity” would dare support such a thing. A third objected that believing it meant forgetting everything learned in the last seventy years and beginning again. The proposition drawing this fire is now taught to nine-year-olds with a cardboard jigsaw: the continents move, and Africa and South America were once joined. The man taking the abuse was a German meteorologist named Alfred Wegener, and the fifty years it took to settle the question turn on what a geologist will accept as proof.
Wegener came at the problem sideways, from weather and balloons and the Greenland ice. “Doesn’t the east coast of South America fit exactly against the west coast of Africa, as if they had once been joined?” he wrote to his future wife in December 1910. “This is an idea I’ll have to pursue.” He pursued it in front of the Geological Association at Frankfurt on 6 January 1912, and at book length three years later in Die Entstehung der Kontinente und Ozeane, which reached English and four other languages with its third revision in 1922.
His case ran well past the jigsaw. Identical fossil species lay in continuous bands across coasts now separated by an ocean, among them the freshwater reptile Mesosaurus, the seed fern Glossopteris, and the land reptiles Lystrosaurus and Cynognathus, none of which could have crossed the Atlantic. Coal, which forms from swamp vegetation, sat under the ice of Antarctica. Glacial deposits sat in the Vaal River valley of South Africa, which is now hot and dry. He also dismantled the reigning alternative, that continent-sized land bridges had once joined the landmasses and later sunk. Continents are granite and the deep floor is basalt, the lighter rock rides on the denser in a balance geologists call isostasy, and a granite bridge forced down to the sea bottom would bob back up the moment the force was released. The only remaining possibility was that the continents themselves had moved.
He put the argument into one image that has outlived everything else he wrote, and he was honest, in the same book, about the hole in it.
IIt is just as if we were to refit the torn pieces of a newspaper by matching their edges and then check whether the lines of print ran smoothly across. If they do, there is nothing left but to conclude that the pieces were in fact joined in this way.
IIThe forces which displace continents are the same as those which produce great . Continental drift, faults and compressions, earthquakes, , [ocean] and [apparent] are undoubtedly connected on a grand scale.
IIIIt is probable the complete solution of the problem of the forces will be a long time coming. has not yet appeared.
Alfred Wegener, The Origin of Continents and Oceans, passage I from the argument of 1915 and II–III from the final revision of 1929, in John Biram’s translation (Dover, 1966). The bracketed words are the translator’s.
That last admission is what sank him. Wegener suggested the continents plowed through the ocean floor, and the English geophysicist Harold Jeffreys answered, correctly, that a slab of solid rock that size would break apart long before it plowed anywhere. Wegener offered the centrifugal force of the Earth’s rotation and a tidal pull from the sun and moon, and knew perfectly well that both were orders of magnitude too weak. At a New York symposium called in 1926 to discuss his theory, speaker after speaker was sarcastic to the point of insult; he said almost nothing, smoked his pipe, and let them finish. Geologists filed the whole idea under geopoetry. He went back to Greenland in 1930 to lead a twenty-two-man expedition onto the ice cap, made a forty-day dogsled run to relieve the mid-ice station in temperatures averaging fifty-eight below, kept his fiftieth birthday there on 1 November, set out for the coast the next morning, and died in his tent. The search party found him on 12 May 1931, dressed, laid on a reindeer skin, his eyes open. They built a mausoleum of ice blocks over him and raised a twenty-foot iron cross above it, and both have long since gone down into the glacier.
The evidence that would have vindicated him was already being collected, by people working on something else entirely. In 1948 a geologist named Marie Tharp took a post at Columbia’s Lamont Geological Observatory as assistant to Bruce Heezen, who spent his years crossing the oceans firing sound at the bottom and timing the echo. The soundings came back to her as numbers on great rolls of paper, and out of them she drew the shape of a seafloor that everyone then assumed to be a featureless basin, a bathtub floor collecting sediment. In 1952 she noticed that every profile crossing the mid-ocean ridge carried the same notch at its crest. She called it the rift. Heezen called it girl talk.
He came round roughly a year later, when the epicentres of ocean earthquakes were plotted and fell along the line she had drawn. Tharp then ran the rift on through the places where soundings were thin, down the South Atlantic, into the Indian Ocean, and around to the East African Rift Valley: one continuous crack wrapped most of the way around the planet. Being right about this was professionally dangerous. Supporters of continental drift were called drifters, and by Tharp’s own account being one could get you fired. Heezen waited four years to publish the rift, and when the paper appeared it carried his name and that of Lamont’s director, Maurice Ewing. Hers was not on it.
The explanation arrived in 1962, from a Princeton geologist who had spent the war commanding a Navy transport across the North Pacific and running its echo sounder the whole way, on his own initiative, to map the bottom he was crossing. Harry Hess had found the ridges standing as much as a kilometre and a half above the abyssal plain, and the trenches near the continental margins going down past eleven. In History of Ocean Basins he proposed that oceans grow from their middles: molten basalt rises along the ridge, makes new floor, and the floor travels away from the axis in both directions, cooling and sinking as it goes, until it reaches a trench and is drawn back down into the mantle. Robert Dietz named the process seafloor spreading. Hess, who had no proof and said so, borrowed a word from an earlier writer and offered the paper as “an essay in geopoetry,” wearing on purpose the sneer that had been used to bury Wegener.
The test came the next year, and it worked because of an accident in the chemistry of basalt. Lava cooling past roughly 580 degrees locks in the direction of the magnetic field surrounding it at that moment, held there by the magnetite crystallising out of the melt. The Earth’s field reverses at irregular intervals, north and south trading places. Put those two facts beside seafloor spreading and a prediction falls out with no room to wriggle: the floor leaving a ridge is a magnetic tape drawn under a recording head at two centimetres a year, so it must carry stripes of alternating polarity, of unequal widths matching the unequal lengths of the reversals, arranged in mirror image on either side of the axis. Fred Vine and Drummond Matthews at Cambridge published exactly that in Nature in September 1963, in three pages, and the survey ships towing magnetometers found the stripes lying where the argument required them.
Lawrence Morley of the Geological Survey of Canada had reached the same conclusion first and sent it to Nature, and then to the Journal of Geophysical Research, early in 1963. Both refused it as too speculative. A referee wrote that such speculation might make interesting talk at cocktail parties but was not the sort of thing that ought to be published under serious scientific aegis. Morley read the paper to the Royal Society of Canada that June, three months before Vine and Matthews appeared in print. The idea is now called the Vine–Matthews–Morley hypothesis, and Morley’s manuscript is sometimes described as the most important paper in the earth sciences ever to be turned down.
Once the stripes can be read, the floor can be dated, and the map of ocean-crust ages is the whole argument made visible at a glance.
The rates are unremarkable, which is exactly why nobody could see them. The Mid-Atlantic Ridge opens at about 2.5 centimetres a year, the Arctic Ridge manages less than that, and the East Pacific Rise near Easter Island runs past fifteen. The Pacific plate has been sliding along the San Andreas past North America at roughly five centimetres a year for ten million years, which is what put those bent streams in the photograph at the top of the page. Since the late 1970s satellite geodesy has measured the motion directly, and the USGS notes that plate speeds and directions averaged over a few years of that data compare well with the speeds and directions averaged over millions of years of magnetic stripes. Two instruments six orders of magnitude apart in timescale return the same numbers.
Wegener’s evidence had been sound the whole time. What he lacked was an engine, and the engine was never in the continents at all; it was underneath them, and its record was lying on the seafloor waiting for someone to tow a magnetometer over it. So when continental drift comes up, the fact worth having is that the coastlines convinced nobody who mattered. The argument was won by the Earth’s own magnetic field, written into the floor of the Atlantic at about the speed a fingernail grows.