The Drifting Continents

Alfred Wegener, Arthur Holmes, and the Solid Mantle That Flows Like Tar

For half a century after German meteorologist and polar explorer Alfred Wegener proposed continental drift at a Frankfurt geological meeting in January 1912 and in his 1915 monograph The Origin of Continents and Oceans, geophysicists led by Sir Harold Jeffreys at Cambridge rejected the theory on a valid mechanical ground. The Earth's silicate mantle transmits earthquake shear waves with a stiffness exceeding steel, and rotational or tidal forces are a million times too weak to plow granite continents through a solid ocean floor.

Before Wegener, nineteenth-century geology was dominated by thermal contractionism, led by Eduard Suess in Vienna and James Dwight Dana at Yale. Contractionists held that as the Earth cooled from its primordial molten state, its interior shrank and its outer crust wrinkled like the skin of a drying apple, pushing up mountain ranges while vast blocks of crust collapsed inward to form ocean basins. When paleontologists discovered identical fossils of the Permian freshwater reptile Mesosaurus in Brazil and South Africa and the heavy-seeded fern Glossopteris across South America, Africa, India, Australia, and Antarctica, contractionists explained the distribution by proposing that ancient transoceanic land bridges had once connected the continents before sinking into the oceanic abyss.

Wegener dismantled the sunken-land-bridge orthodoxy using gravimetric isostasy. Dredging and seismic surveys showed that the Earth's crust is bimodal: continents are built of light granitic rock (sial, density near 2.7 grams per cubic centimeter), whereas the ocean floor and upper mantle consist of dense basaltic and peridotitic rock (sima, density 3.0 to 3.3 grams per cubic centimeter). By Archimedes' principle, a buoyant block of continental granite floats in hydrostatic equilibrium atop the denser mantle just as an iceberg floats in seawater. It can never sink to the bottom of the Atlantic unless weighted down by an impossible external load. Pairing that buoyancy argument with the jigsaw match of the Atlantic continental shelves, severed Caledonian and Appalachian fold belts between Scotland and Newfoundland, and Carboniferous glacial scratches in tropical Brazil and India alongside coal beds in polar Spitsbergen, Wegener concluded that all continents had once joined in the supercontinent Pangaea before breaking apart and drifting horizontally.

Where Wegener stumbled was his mechanical model of how the continents moved. He pictured granitic continents acting like icebreakers plowing directly through stationary oceanic basalt, pushed by two tiny planetary forces: the pole-fleeing centrifugal component of Earth's rotation on an oblate spheroid (Polfluchtkraft) and westward gravitational tidal drag from the Moon and Sun. In his 1924 treatise The Earth, Harold Jeffreys calculated the maximum shear stress produced by both forces—roughly $10^2$ to $10^3\text{ Pa}$—and compared it against the $10^8\text{ Pa}$ yield strength of solid oceanic crust and mantle rock derived from seismic transverse waves. Because Wegener's proposed driving forces fell short by five to six orders of magnitude, the 1926 American Association of Petroleum Geologists symposium in New York buried continental drift as a geophysical impossibility.

The resolution required separating instantaneous elastic stiffness from long-term viscoelastic creep. Under the high temperatures ($1,300^\circ\text{C}$ to $3,000^\circ\text{C}$) and gigapascal pressures of the Earth's interior, crystalline olivine and pyroxene in the solid mantle behave as a Maxwell viscoelastic material whose total strain rate sums an instantaneous elastic term and a time-dependent viscous term ($\dot{\varepsilon} = \dot{\sigma}/E + \sigma/\eta$). Over the one-second period of an earthquake shear wave, the elastic modulus $E$ dominates and the solid mantle rings like a steel bell. Over timescales exceeding the Maxwell relaxation time of a few thousand years, thermally activated crystal-lattice vacancies and dislocations migrate through the solid grains, allowing solid rock to flow with an effective viscosity around $10^{21}\text{ Pa}\cdot\text{s}$. As British geologist Arthur Holmes demonstrated between 1928 and 1931, radioactive decay of uranium, thorium, and potassium inside the mantle generates heat faster than conduction can carry it to the surface, driving the mantle's thermal Rayleigh number above $10^6$ and organizing solid-state thermal convection cells. When Harry Hess's 1962 seafloor-spreading model was joined with Fred Vine and Drummond Matthews's 1963 mapping of symmetric magnetic reversals locked inside cooling mid-ocean ridge basalt, the mechanical paradox vanished. Continents do not plow through the ocean floor, but ride passively embedded within rigid lithospheric plates carried atop a convecting viscoelastic mantle.