In 1912, the German geologist Alfred Wegener proposed that Earth's continents are mobile rafts of lighter crust that have shifted over time by plowing their way through the denser crust of the oceans. The theory, called continental drift, was partly motivated by the apparent fit, like puzzle pieces, of the coastlines of South America and Africa. Wegener first presented his theory of continental drift at a meeting and in a paper, and then as a book, The Origin of Continents and Oceans, published in 1915. He continued to write updated versions of this work until his death in an ill-fated expedition to Greenland in 1930. Wegener maintained that Earth is composed of concentric shells of increasing density from crust to core. The outermost shell is not continuous but made of continental blocks of lighter rock called sial (an acronym for silica- and aluminum-rich rock) floating in the denser sima (silica- and magnesia-rich rock) underlying the oceans. All the continents had been joined in the supercontinent of Pangaea. As the continent broke up, the pieces moving apart left bits behind, explaining the presence of nonvolcanic islands and island chains, according to Wegener. Where the moving pieces collided, mountains formed. They were thrust up either by the plowing of the continents through the sima, as in the case of the Andes, or by the colliding of two blocks of sial, as in the case of the Himalayas. As for the force driving continental drift, Wegener initially invoked Polfluchtkraft- -a force causing flight from the poles as a result of Earth's rotation- -and later the tidal force resulting from the gravitational attraction between Earth and the Moon and Sun.

One of the most influential geologists to join the mobilist camp, as the drifters school became known, was Arthur Holmes. Holmes recognized the importance of radioactive heat- -which had recently been discovered- -and realized that there must be a mechanism to remove it from Earth's interior. That mechanism, he argued, is convection- -the rising of less dense material and the sinking of more dense material. He went on to propose that the mantle (the part of Earth's interior below the outer crust and above the core) convects in large, circulating patterns, and that this motion carries the continents across Earth's surface. He also related crustal movement and mantle convection to the evolution of mountain belts. Wegener adopted Holmes's mechanism in the last rendition of his theory. Holmes, for his part, presented his grand concept of a dynamic Earth in his influential and popular text Principles of Physical Geology, published in 1944.

Although it was eventually supplanted by the theory of plate tectonics, Wegener's theory of continental drift influenced science because it explained disparate observations, because it was placed in the context of existing theories, and because it offered a coherent view of Earth's evolution. For example, Wegener showed not only that the coastlines on opposite sides of the Atlantic fit together, but that geologic features on the different continents fit as well. He asserted that the Appalachians, which can be traced northward through the Canadian Maritime Provinces, match the Caledonian Mountains in Scotland and Norway. He marshaled evidence from the distributions of fossil and living species to argue that land bridges joining continents were less likely than a single continent. The example commonly cited is that of Mesosaurus, a shoreline scavenger reptile that lived in the Permian period and is found as fossils in rocks on both sides of the South Atlantic Ocean. Mesosaurus was thought not to be a great swimmer, certainly not able to cross an ocean.

Wegener also found supporting evidence in ancient climates. Mounting observations indicated that the past climate of many regions was much different from the present climate. In the tropics, geologists had found sand and gravel left by ancient glaciers, and in rainy regions they had located prehistoric deserts. Then there was the discovery, by a British expedition, of plant fossils only 600 kilometers (370 miles) from the South Pole. Particularly puzzling was evidence suggesting that widely different climates in different regions had occurred at the same time, So one could not account for different climates by claiming simply that the whole of Earth was once hotter or once cooler than now. Wegener solved this dilemma by showing that observations of paleoclimate could be explained if the positions of the continents had shifted relative to those of the poles.