Alfred Wegener, in the first three decades of this century, and DuToit in the 1920s and 1930s gathered evidence that the continents had moved. They based their idea of continental drift on several lines of evidence: fit of the continents, paleoclimate indicators, truncated geologic features, and fossils.
fit of the continents
As far back as the voyages of exploration of the New World in the 16th and 17th centuries, as rudimentary maps were produced, scholars had noted the complementary shapes of the coastlines of Europe and Africa with North and South America. Some had even wildly proposed that the continents had been split apart.
Wegener was a meteorologis and geologist. Among other things, he studied paleoclimate indicators in sedimentary strata. He studied the geologic literature and recognized that upper Paleozoic (Carboniferous and Permian) strata in northwestern Europe strata contained extensive coals that could only have formed in a hot wet climate like the present equatorial region. In rocks of the same age in equatorial Africa he knew there were glacial tillites. These indicated to Wegener that the continents must have moved (Europe from near the equator, Africa from the polar region into the equatorial region).
He was also aware of certain problems that had been noted by paleontologists. Fossils of a shallow water reptile, Mesosaurus, were found in both Africa and South America even though they could not swim across the Atlantic Ocean. Fossils of a family of seed ferns, Glossopteris, were found in Africa, South America, India, and Antarctica. These and a number of other fossil groups had identical populations separated by thousands of kilometers of ocean. It seemed impossible to account for such stunning parallel evolution. Paleontologists called on implausible land bridges connecting the continents. But Wegener argued that rising and falling land bridges in the oceans were not likely considering the observation that the ocean crust was made of denser (basaltic) rock than the continents. He argued that this denser oceanic crust could not rise up above sea level. Likewise, if the land bridge was less dense (granitic) continental rock it would be too light to sink into the denser rock below.
truncated geologic features
Wegener also noticed that there were major mountain building event sin the northern Appalachians and in northwestern Europe of the same age. The Acadian Orogeny in the northern Appalachians and the Caledonide orogeny in northern Great Britain and Scandinavia occurred during the Devonian period yielding extensive folding, faulting, igneous intrusion, metamorphism, and the development of major sedimentary features (clastic wedges) from the erosion of the rising mountains. If North America and Europe are reconstructed into a nice fit the Caledonides of northwestern Europe are seen to be a continuation of the Appalachians. The Paraña basalts of South America and the Etendeka basalts of Africa were both extruded about 130 million years ago on the conjugate margins of these continents and both are cut off at the coast as if cut in half.
These and other clues indicated that the continents had once been together as part of a larger supercontinent. Wegener first presented his ideas in 1912 and they were elaborated with successive editions of his book, The Origin of Continents and Oceans, through 1929. In it he described a supercontinent that he called Pangea, containing all of the present major continental masses. DuToit's book, Our Wandering Continents, was published in 1937. It updated the continuing work of Wegener and his own. DuToit proposed, in addition to Pangea, a southern supercontinent which existed in the Paleozoic, made up of Africa, South America, Antarctica, India, and Australia, called Gondwanaland.
The hypothesis of continental drift was read with interest but also with much skepticism because there was no plausible mechanism to account for continental motions. Basic intuition wonders how a continent can be slid across the ocean against tremendous frictional forces. Wegener believed in the isostacy concept (continents afloat in the mantle) and that the problem was more akin to pushing an iceberg across the ocean. Nevertheless there were no known forces of sufficient magnitude to account for continental motions.
Mantle Convection Proposed
The British geologist Arthur Holmes suggested a possible mechanism in the late 1920's. Radioactive heat generated in the Earth's interior might cause the Earth's interior to heat up unless there was some mechanism to remove the heat. Holmes proposed that hot mantle, behaving as a very viscous fluid, would rise by convection toward the surface where it would cool and contract (become denser) and then descend back deep into the Earth. There it would heat up and expand (become less dense) and then rise again. Might these proposed convection currents provide a driving force for continental drift?
When rocks are formed small amounts of magnetic
minerals, like magnetite and hematite, are incorporated. The
magnetism of the magnetic minerals is aligned with the Earth's
magnetic field when the rock is formed. The rock retains, in many
cases, a permanent record of this field direction. The Earth's
magnetic field can tell us the direction to the poles from the
familiar compass direction. But the lines of force also are inclined
to the Earth's surface at varying angles (vertical upward at the
south pole, less and less steep moving toward the equator, horizontal
at the equator, then progressively steeper downward moving from the
equator to the north pole where it is vertical downward. The
inclination of the magnetic field is proportional to the latitude.
Paleomagnetic study of ancient rocks can determine the latitude at
which a rock forms and the direction to the North or South Pole.
Therefore, paleomagnetists can determine north-south motions and
rotations of continents. Paleomagnetists in the 1950s and 1960s
(Runcorn, Irving, Creer, Cox, and others) were studying magnetism in
rocks. They found that the magnetism recorded in old rocks usually
did not coincide with the present direction of the Earth's magnetic
field. Rocks of a given age from one continent all contain
magnetizations that point in a common direction (normally not toward
the present north pole). The apparent position of the pole is
progressively farther from the present north pole as recorded by
older and older rocks. Had the north pole moved over time or had the
continents moved? When the apparent polar wander paths from separate
continents are compared we can see that they are different. So the
continents must have moved independent of one another. Furthermore
the paleomagnetic results from upper Paleozoic rocks fit very well
with the observations and reconstructions of Wegener and DuToit. The
latitudes implied by the magnetization of late Paleozoic age rocks
from the various continents are consistent with the latitude that
Wegener place them at in his Pangea reconstruction.
From the above it appeared that continental drift must have occurred, but there was no mechanism that could account for pushing the continents across the ocean crust. Therefore it was not taken seriously by most.
During the 1950's paleomagnetists also discovered that the magnetization in some layers of volcanic rock pointed toward the North Pole and other layers were magnetized toward the South Pole. They had discovered that the Earth's magnetic field reverses occasionally.
Harry Hess was a U.S. Naval officer during World War II on a destroyer escorting ship convoys to England. His ship towed a sensitive magnetometer in an attempt to detect the steel hulls of Nazi submarines that preyed on Allied shipping. He noticed that as the ship sailed over the mid-Atlantic Ridge the magnetometer recorded small fluctuations in magnetic field intensity. After the war Hess went to Princeton and studied this phenomenon. He suggested that these fluctuations were due to varying magnetizations of the ocean crust. The magnetometer recorded primarily the direction and intensity of the Earth's magnetic field but also could detect changes in the magnetization of the ocean crust. Apparently the ship sailed over some sections of ocean crust that were magnetized such that they complemented the Earth's magnetic field therefore making the recorded intensity stronger. Other sections must be magnetized in such a way as to subtract from the Earth's magnetic field strength. The sections with complementary magnetization must be sections of the crust magnetized with normal polarity like the present field of the Earth. The sections of the ocean crust whose magnetism subtracts from the main field of the Earth must have been magnetized and formed during periods of reverse polarity.
Vine and Matthews (1963) mapped the varying magnetic intensity on one side of the midocean ridge. They reported linear stripes of alternately higher and lower magnetic field intensity, marine magnetic anomalies, parallel to the midocean ridge. They were the first to completely state the hypothesis of seafloor spreading. They believed that ocean crust was continuously created at the midocean ridges by igneous intrusion and volcanic activity; the newly-formed crust then breaks in two and spreads away from the ridge. The newly forming strips of crust become magnetized alternately in normal or reverse polarity as the Earth's magnetic field reverses.
Pitman and Heirtzler (1966) mapped the magnetic anomalies across a section of the Pacific-Antarctic Ridge and the Reykjanes Ridge south of Iceland. They showed that the magnetic anomalies were symmetric about the ridge; the same pattern of changing intensity was found on both sides of the midocean ridge. This was the conclusive evidence for the seafloor spreading hypothesis.
T.J. Wilson (1965) proposed and Lynn Sykes (1967) confirmed transform faults offsetting midocean ridge segments. Seismic evidence gathered from earthquakes by Sykes showed strike-slip (side-by-side) motion on the transform faults, no earthquakes on the fracture zones, normal fault (stretching) earthquakes on midocean ridges, and thrust fault (compression) earthquakes near deep ocean trenches. Other seismologists showed that there was a descending plane of earthquakes (Benioff Zone) descending from the trenches. Volcanic arcs like the Andes and Cascades mountains and volcanic island arcs like the Mariannas and Aleutians lie over Benioff Zones, set back from deep ocean trenches. The deep ocean trenches were then locations where the ocean crust was being subducted, or returned back into the Earth's interior.
Wilson (1963) showed that islands generally got older the farther they were from the midocean ridges. He set the stage for the understanding of hot spots, chains of volcanic islands in the middle of plates, that get progressively older away from the midocean ridge, suggesting that they form as their plate moves slowly over plumes of hot rising mantle material. The Hawaiian islands are one of the best examples of a hotspot track.
The theory of plate tectonics offers a mechanism, acceptable to the physics community, that can account for the continental motions described by Wegener, DuToit, and the paleomagnetists. It accounts for all of the major features of the ocean basin and the surface of the Earth in general. Beginning in 1966-1967 it has become overwhelmingly accepted by the scientific community. It is the primary predictive tool for understanding such ongoing processes as volcanoes and earthquakes.
Central to plate tectonics is the understanding of the linkage between seafloor spreading and currents of upwelling hot mantle rock that yield the molten rock that solidifies as new crust at the midocean ridges. The midocean ridge system is the main avenue for the release of heat from the interior of the Earth. Upheaval of mountains is the result of continental collisions, such as the formation of the Appalachians when Pangea formed by the collision of Gondwana and Laurasia. Continental collisions are in turn the result of tectonic plate motions which result from seafloor spreading which releases heat from the Earth's interior.