The movement of plates in Earth's crust formed mountain ranges like the Andes, pictured here surrounding Machu Picchu.
Close Caption
Lukasz Kurbiel/Shutterstock
Contintental Drift and Plate Tectonics
From Grolier's New Book of Popular Science
The San Andreas Fault, located in California.
Close Caption
Stocktrek/Brand X/Getty Images
Earth’s crust is broken into segments called tectonic plates. The major and minor plates outlined on the map above move very slowly in the directions indicated by the arrows, carrying along the continents and ocean floors that lie on top of them.
Close Caption

Jigsaw puzzle—that's what many people think of when they look at a map of the world. Push Europe and Africa over across the Atlantic Ocean, and the coastlines fit—fit very well—against the coastlines of the Americas. Move India, Australia, and Antarctica around, and their coastlines also fit together—like pieces of a giant jigsaw puzzle.

When the Americas were discovered and mapped some five centuries ago, scientists noticed that the opposing coasts along the Atlantic Ocean had shapes that would fit together. They proposed that, early in Earth's history, the continents had been joined, and that later they had been violently torn apart. In the 19th century, this idea was supported by studies of geology and life-forms on both sides of the Atlantic. The studies revealed many similarities between species, suggesting that they had been intermixing and interbreeding across the continents as recently as 150 million years ago. Studies such as these led Alfred Lothar Wegener, a German meteorologist, to propose in 1912 the theory of continental drift.


Wegener believed that the opening of the Atlantic and Indian oceans was not due to an earlier cataclysm, but rather had occurred slowly and gradually. He bolstered his argument with measurements from recent surveys of the distance between Greenland and Europe. The measurements suggested that the two landmasses were moving away from each other at a perceivable rate. Wegener further theorized that, because Earth is a rotating sphere, there exists a force that pushes the continents toward the equator.

The continents, he believed, plow through the rocks of the seafloor like a ship through water. As a continent moves, he added, coastal mountain ranges pile up like low waves along the land's leading edge.

Little Support. Many biologists rallied around Wegener's theories, convinced that his ideas explained the many similarities between species on now-separate continents. But the world's leading geologists and physicists disagreed with Wegener's "radical" ideas. Some argued that similarities between species could be explained in other ways—for example, that animals had crossed from continent to continent on land bridges that have since disappeared.

Scientists also challenged the accuracy of land surveys that recorded Greenland's drift. Many insisted that the forces generated by Earth's shape are small and act in the wrong direction to explain Wegener's ideas. They argued that if continental drift created coastal mountains on a continent's advancing side, then there should be signs of disturbance and "pulling apart" along the continent's trailing edge.

While these arguments did not entirely negate Wegener's continental-drift theory, they weakened its support. When Wegener died in 1930, few geologists had accepted his ideas.

The chief exceptions were a group of geologists in the Southern Hemisphere who were particularly impressed by the near-exact match of their continental coastlines. Some Alpine geologists also embraced Wegener's theory because it explained why certain mountainous rock beds showed tremendous folding and thrusting—as if they had been compressed by colossal forces.

Early Clues. About the time of Wegener's death, several geologists made observations that suggested a mechanism for, and provided an explanation of, continental drift. They realized that the radioactive decay of several naturally occurring elements produced a great amount of heat—enough to melt rocks in Earth's interior.

The slow deformation of deep rock, they theorized, would generate convection currents in Earth's interior. The currents would be like the rising and falling currents of water boiling in a pan, but vastly larger and much, much slower. These geologists further proposed that such currents may be rising under ridges then known only vaguely to occur along the middle of each ocean. They suggested that these same convection currents could be turning downward again under active mountains and deep-ocean trenches like those off the coasts of Chile, Peru, and eastern Asia. They also believed that it was possible for these currents of molten rock to carry the continents along at rates of about 2 to 4 inches (5 to 10 centimeters) per year.

Although the case for continental drift was building, most scientists continued to reject the theory until the late 1950s. Then, between 1956 and 1967, several scientific discoveries revived Wegener's ideas and turned the tide in geologic thinking. One discovery was that the past motions of continents could be traced through an analysis of the magnetism of rocks. Another was that there is indeed a continuous mid-ocean ridge throughout the world's oceans.


Since ancient times, it has been known that a few rocks, chiefly iron ores, are natural magnets and can be used as compasses. As scientific instruments improved, it was found that most rocks are magnetized, albeit more feebly than iron ores. It was further discovered that rocks acquire their magnetism from Earth's magnetic field at the time of their formation, and that they tenaciously retain this magnetization thereafter.

Thus, lava ejected near Earth's magnetic poles is magnetized as it cools in the same vertical direction as the field at the poles. Sediments accumulating in the ocean near the poles are similarly magnetized. On the other hand, the magnetic field near Earth's equator is horizontal, and rock beds forming there are magnetized in that direction. Thus, rocks are magnetized in a direction appropriate to the latitude of the place where they form; they retain this direction of magnetization forever.

In the mid-1950s, it was discovered that rocks of recent origin on several continents were magnetized in directions corresponding to their location, but that rock beds known to have been formed at successively older times showed progressively different magnetic orientations. Further, it was found that the changes in magnetic orientation were exactly what would be expected if the continents had been moving in the way Wegener had proposed. This was strong supporting evidence for the theory of continental drift.


A continuous mid-ocean ridge throughout the world pointed to a possible mechanism explaining the movement of the continents. In 1956, U.S. geophysicist Maurice Ewing and others began detailed investigations of the ocean floor. A substantial body of evidence accumulated over several years suggested that a great, broad mountain system lay down the center of the oceans.

In the Atlantic Ocean, the mountainous ridge extends from the Arctic to the southern seas. There the mid-ocean ridge turns toward and through the Indian Ocean, where the ridge splits into two branches: one branch goes into the Gulf of Aden and the Red Sea; the other passes midway between Australia and Antarctica, crossing the Pacific and entering the Gulf of California. There it forms the San Andreas Fault and emerges again off the coast of Canada, extending as far north as Alaska. A rift—a faulted, or cracked, valley—follows the crest of the mid-ocean ridge in most places. Shallow earthquakes occur along the ridge in all oceans.

Seafloor Spreading. By 1960, ships had surveyed the mid-oceanic ridge in its entirety. Based on accumulated evidence, geologist Harry Hess of Princeton University in Princeton, New Jersey, revived the idea that the crest of the ridge is where the ocean floor spreads apart. Hess proposed that as the two sides of the ridge spread apart, intrusions of hot lava from beneath the crust create new ocean floor. He combined this proposal with the idea that the surface crust—which is cool, brittle, and broken in places—could slide, carried by very slow currents of heat in Earth's hot interior. The broken pieces, or plates, jostle and bump one another as they move. When two of these plates converge head-on, one plate may thrust over the other. The lower piece, pushed beneath the first, may dive into the interior of Earth.

This type of plate activity is now believed by scientists to occur under actively forming mountains, ocean trenches, and volcanic-island arcs. Such activity is also thought to be responsible for the many earthquakes that occur in such areas.


By 1965, investigations led to the proposal that Earth's surface was broken into seven large plates and several smaller plates. It was further suggested that these plates are rigid, and that their boundaries are marked by earthquakes and volcanic activity. In recent years, satellite pictures have documented the existence of plate boundaries. An especially visible example is the San Andreas Fault in California.

Plates interact with one another at their boundaries by moving toward, away, or alongside each other. Faults are examples of boundaries where two plates slide horizontally past each other. Mid-ocean ridges mark boundaries where plates are forced apart as new ocean floor is being created between them. Mountains, volcanic-island arcs, and ocean trenches occur at the boundaries where plates are colliding, causing one plate to slide beneath the other. The network of crustal plates and the geologic activity caused by their movement is referred to as plate tectonics.

Raft Continents. The original continental-drift theory suggested that continents plowed through the ocean floor like ships. Plate tectonics, on the other hand, holds that continents are carried along together with the surrounding seabed in huge plates—much like rafts frozen into the ice on a flowing stream.

There are several major plates. The North American plate comprises North America and the western half of the North Atlantic seafloor. The South American plate includes South America east to the Mid-Atlantic Ridge. The African plate contains Africa and its surrounding seafloor. The Antarctic plate has Antarctica and surrounding seafloor. The Eurasian plate includes Europe, Asia, and nearby seafloor. The Pacific plate underlies the Pacific Ocean. Recently, an international team of geologists and other researchers analyzing seafloor measurements discovered that what was classified as the Indo-Australian plate may actually be two separate plates—one with the Indian subcontinent and the adjacent seabed, and the other with Australia and surrounding waters.


In the late 1950s, magnetic surveys of the ocean floor revealed a regular, striped pattern of alternately stronger and weaker magnetization. It was further determined that these stripes run parallel with the axes of the mid-ocean ridges and follow the ridges around Earth. In the early 1960s, Frederick J. Vine and Drummond H. Matthews of Great Britain and Lawrence W. Morley of Canada combined these observations into a theory that explained the magnetic patterns of the ocean crust.

Previously, researchers had documented that Earth's magnetic field reverses itself at semiregular intervals of a few thousand to a few million years. During this process, Earth's magnetic field weakens—and then returns in a reversed direction. In other words, what had been the north magnetic pole becomes the south magnetic pole and vice versa.

Vine, Matthews, and Morley pointed out that the lava flowing up along the crest of the mid-ocean ridge would be magnetized in the direction of Earth's field at the time the lava cooled, forming new seafloor. Thus, the gradual weakening and complete reversal of the planet's magnetic poles should be recorded as stripes on either side of the mid-ocean ridge. The end result should be a symmetrical, striped pattern of magnetism across the spreading seafloor—exactly what the surveys had found.

Scientists studying the age and magnetic orientation of lava flows on land made enough measurements to create a time scale of magnetic reversals that could be used to date the ocean floors. According to this scheme, it appears that northwestern Africa began to separate from the eastern coast of the present-day United States about 180 million years ago; and Africa began to move away from South America about 160 million years ago. Moreover, no part of the present-day ocean floor has been found to be more than 200 million years old.


The theory of plate tectonics also explains how and when mountain ranges such as the Andes and the Cascades arose. Both these ranges lie at the converging boundaries of two plates. As these two plates collide, portions of Earth's crust are uplifted and folded, causing a great compressing and thrusting of rock. Earthquakes and volcanoes continue to occur in the area of these new and active mountain ranges.

But what about older mountain ranges such as the Urals, which lie in the middle of continents?

Magnetic studies suggest that the continents of Earth have separated and collided many times. Indeed, they have been separating and coming together in different patterns for much of geologic time. This suggests that older mountains such as the Urals may mark the edges of ancient continents far different from those known today. If so, these old mountains may have formed when the plates carrying the ancient continents collided with each other, or with plates of oceanic crust. In other words, older mountains may mark the sites where ancient plates converged far back in time.

True to this theory, geologic evidence suggests that when the Atlantic Ocean started to open, it did so nearly along the closure lines of an earlier ocean. The sides of this earlier ocean joined some 600 million years ago and raised up a mountain range. The opening of the Atlantic millions of years later tore this range into fragments that are now part of the Appalachian Mountains in eastern North America, the mountains farther north in eastern Greenland, and the mountains of western Morocco in Africa.

The theory of plate tectonics also explains other geologic occurrences, such as volcanoes, which typically occur along plate boundaries. Volcanoes also are found at separating ridges—for example, in Iceland, the Azores, and Tristan da Cunha; they also occur along converging plate boundaries, as in the Andes of South America. A few lines of volcanoes have erupted in the middle of tectonic plates—for example, in Hawaii and in Yellowstone Park. The geologic processes that form "midplate" volcanoes provide further evidence of plate movement.

Hot Spots

Scientists believe that, in addition to the great, slow convection currents that carry plates about Earth, there are also smaller, rapidly rising mantle plumes, columns of hot material rising from deep within Earth. (Earth is believed to be composed of an inner solid core, a middle mantle, and an outer crust.) These plumes of molten rock, often called hot spots, rise and erupt through the crust of a moving plate.

Most of the isolated midplate volcanoes, such as those of Hawaii and Yellowstone, lie at one end of a line of extinct volcanoes that grow steadily older with distance from the active center. Hawaii's Mauna Loa is at the extreme southeastern end of the rest of the Hawaiian island chain. The volcanoes in this chain become steadily older and less active to the northeast. Likewise, Yellowstone's hot springs and geysers are at the eastern end of a line of extinct volcanoes that extend into Idaho. Such a line of volcanoes suggests that the crust of Earth is passing over a hot spot, or hot spots, in the deeper mantle. As the crustal plate moves, the hot spot "punches" up a line of volcanic and hot-spring activity.

Indeed, the motions of the Pacific plate are compatible with the direction of the Hawaiian chain and the ages of its volcanic islands. Plate motion has slowly moved the volcanic islands away from the hot spot that created them. In other words, the Hawaiian island chain traces the motion of the Pacific plate.

Ridge hot spots.  As molten rock flows up along the mid-ocean ridges to create new seafloor, the lava flows more abundantly in certain spots, producing volcanic islands. Scientists believe that these places of abundant lava flow may be hot spots that occur between two separating plates. Two such plates underlie the large, highly volcanic island of Iceland, which straddles the Mid-Atlantic Ridge. On one side is the North American plate; on the other, the Eurasian plate. Similarly, ridges extend from the active volcanic island of Tristan da Cunha westward to South America and eastward to Africa. Some geologists propose that, although such hot spots do not actually move plates, they may mark weak points in the mantle, which in turn help determine the lines along which plates fracture and separate.

Convergence hot spots.  Hot spots of volcanic activity often occur at the junction where plates collide. Examples are the volcanic Azores, which arise where the North American, Eurasian, and African plates meet. Another—Macquarie Island, south of New Zealand—marks the meeting point for the Pacific, Antarctic, and Indo-Australian plates. These hot spots may be fueled in part by plate collision.


The theory of plate tectonics may also be used to explain circular island arcs and oceanic trenches. Where ocean floor is being carried down freely into the interior, it is likely to do so along circular arcs. Geometry dictates this pattern, a concept illustrated when a person pushes his or her thumb into a dead tennis ball: the resulting depression is circular. This principle may explain the origin of circular volcanic-island arcs, such as the Aleutians.

If, on the other hand, a continental plate pushes past and over an oceanic plate, the ocean floor will be forced down into the interior of Earth along an offshore trench. The deep trenches off the coasts of Peru and Chile are prime examples.


The theory of plate tectonics, which has now been generally accepted by geologists, has several practical applications. It is vitally important for earthquake prediction, because it explains much about the cause and distribution of quake activity. Plate tectonics is also helpful to the petroleum and mining industries, because it helps explain how deposits are formed and where they are most likely to be found. Petroleum deposits, for instance, tend to exist in some of the world's oldest rock beds, which, because of seafloor spreading, should be found farthest from the mid-oceanic ridges. True to theory, important petroleum deposits tend to be found near coasts, both on- and offshore.

As for minerals, exploration of mid-ocean ridges has revealed streams of hot water bearing great concentrations of dissolved metals. These minerals-in-solution pour out of rifts in the Red Sea, near the Galápagos Islands, and in the Gulf of California. The discovery of these mineral-rich hot springs helps explain the source of the valuable precipitated minerals found in manganese nodules along the seafloor. More underwater hot springs will undoubtedly be found, and some may be exploited as sources of mineral deposits or as sources of heat and energy.

J. Tuzo Wilson