Stripes on the Seafloor: How a Rejected Idea Became Plate Tectonics

A theory can be right and still be rejected for fifty years — until the planet itself leaves a receipt.

In 1912 a German meteorologist named Alfred Wegener proposed that the continents were not fixed. He pointed out that the coastlines of South America and Africa fit together like torn pieces of paper, that identical fossils of land animals and plants turned up on continents now separated by oceans, and that distinctive rock formations matched across the Atlantic. His conclusion was that the continents had once been joined in a single landmass and had since drifted apart. Geologists mostly hated it. The evidence was suggestive, but Wegener could not say what force could possibly shove a continent across the planet, and without a mechanism the idea looked like coincidence-hunting. He died in 1930 on an expedition to Greenland with his hypothesis still in disrepute. It took the seafloor to vindicate him, and the story of how is a good illustration of how a theory finally wins.

The trouble with the continents is that they are old and complicated. They have been folded, eroded, and reworked for billions of years, so reading their history is like trying to reconstruct a manuscript that has been rewritten a hundred times. The ocean floor, it turned out, kept a cleaner record, but nobody could read it until the mid-twentieth century, when wartime and postwar technology let ships map the seabed and measure its magnetism in detail for the first time. What they found was unexpected. Running down the middle of the Atlantic, and through the other oceans, was a continuous mountain range, a mid-ocean ridge, with a rift valley along its crest.

In 1962 the geologist Harry Hess offered an explanation that he himself described, half-jokingly, as geopoetry because the evidence was still thin. He suggested that the mid-ocean ridges are where new ocean floor is born. Hot material rises from the mantle along the ridge, cools into fresh crust, and then spreads outward to either side, carried slowly away from the ridge like goods on two conveyor belts running in opposite directions. If new crust is constantly being made, the Earth must be shedding old crust somewhere to stay the same size, and Hess pointed to the deep ocean trenches, where the seafloor plunges back down into the mantle. This was seafloor spreading. It was a beautiful idea, and like Wegener’s it needed proof.

The proof came in 1963, from a property of the rocks that no one had thought to use this way. The Earth’s magnetic field flips every so often, with north and south magnetic poles trading places, and it has done so many times over geological history. When fresh basalt erupts at a mid-ocean ridge and cools, the iron-bearing minerals in it lock in the direction of the magnetic field at that moment, like a compass frozen in stone. Fred Vine and Drummond Matthews in Britain, and independently Lawrence Morley in Canada, put two and two together. If Hess was right and the seafloor spreads outward from the ridge, then the crust should carry a recording of every magnetic reversal that happened while it was forming. As you move away from the ridge you would cross bands of rock magnetized normally, then in reverse, then normally again, one stripe for each interval between flips.

And the stripes should be symmetric, a mirror image on each side of the ridge, because the two conveyor belts are laying down the same record as they move apart. When surveys of the magnetism of the seafloor came in, that is exactly what they showed: parallel stripes of alternating magnetization, running along the ridge and matching side for side across it. There was no good way to explain a pattern like that except spreading. The seafloor had tape-recorded its own creation, and the tape was running in both directions from the ridge.

Everything fell into place quickly after that. If the seafloor spreads, the rock should get older the farther you go from the ridge, and drilling ships sent down to sample the crust confirmed it: youngest at the ridge, progressively older toward the continents, and nowhere older than about 180 million years. That last number is striking. The ocean floor is geologically young because it is constantly recycled, made at the ridges and destroyed at the trenches, while the continents, which ride on top and are too buoyant to sink, preserve rocks billions of years old. The earthquakes that cluster along the trenches trace out planes that dip down into the mantle, marking the path of the descending slabs.

By the end of the 1960s these pieces had been assembled into the theory of plate tectonics. The Earth’s rigid outer shell is broken into a set of plates, each carrying continents and ocean floor together, that move over the softer mantle beneath. They pull apart at the ridges, where new crust forms, grind past one another along faults, and collide at the trenches and mountain belts, where one plate dives under another or two continents crumple together to raise ranges like the Himalayas. Wegener’s drifting continents were real, but he had the picture slightly wrong. The continents do not plow through the ocean floor. They are passengers on plates that include the ocean floor, and the engine is the slow circulation of heat in the mantle, the very mechanism whose absence had sunk his idea fifty years earlier.

It is worth remembering how recent all this is. The theory that now underlies all of geology, that explains earthquakes, volcanoes, mountain ranges, and the distribution of fossils, was not settled until around 1968. The clinching evidence was not a continent at all but a set of magnetic stripes on the floor of the sea, written by a planet that reverses its own compass and keeps the receipts.

Wegener was vindicated not by a continent but by a magnetic barcode on the ocean floor — a reminder that what an unloved theory usually lacks is not more evidence but a mechanism.