Plate boundaries: divergent, convergent, transform; seafloor spreading and Wilson cycles

Intuition Beginner

Where plates pull apart, the ground splits and magma rises to build new crust. These are divergent boundaries. The Mid-Atlantic Ridge is a vast underwater mountain chain where new ocean floor forms as Eurasia and the Americas drift apart. On land, the East African Rift shows an early stage: the continent is stretching and beginning to tear. Lakes fill the valleys formed as crust drops between separating blocks. Given enough time, a new ocean will open.

Where plates collide, one bends down and plunges into the mantle beneath the other. These are convergent boundaries. When ocean crust meets continental crust, the denser ocean plate subducts, forming a deep trench and a volcanic arc on land. The Andes grew this way as the Nazca Plate dives beneath South America. When two continents collide instead, the crust crumples and thickens into vast mountains. The Himalaya rose as India rammed into Asia and keeps rising today.

Where plates slide horizontally past each other, no crust is created or destroyed. These are transform boundaries. The San Andreas Fault in California is the most famous. The Pacific Plate grinds northwest past the North American Plate at roughly 5 centimeters per year. The motion is not smooth: the plates lock, accumulate strain, then snap in an earthquake. The 1906 San Francisco earthquake released centuries of built-up strain along this boundary in seconds.

At mid-ocean ridges, magma wells up and solidifies into fresh basalt. The Earth's magnetic field reverses direction every few hundred thousand years. Each reversal is recorded in the newly formed rock as it cools. The result is a pattern of magnetic stripes mirrored on both sides of the ridge, like a barcode stamped onto the seafloor. This symmetry was the decisive proof that seafloor spreading is real.

Oceans open and close in a repeating pattern called the Wilson cycle. A continent rifts apart and a new ocean basin grows. After hundreds of millions of years, subduction begins at the ocean margins and the basin shrinks. Eventually the bordering continents collide, thrusting up mountain ranges. The Atlantic opened roughly 200 million years ago when Pangaea broke apart. Before that, an earlier ocean called Iapetus had opened and closed in the same region, building the Appalachians.

Visual Beginner

Boundary type	Relative motion	Creates	Destroys	Example
Divergent	Plates move apart	New oceanic crust	Nothing	Mid-Atlantic Ridge
Convergent (ocean–ocean)	Plates collide	Island arc, trench	Oceanic crust	Aleutian Islands
Convergent (ocean–continent)	Plates collide	Volcanic arc, trench	Oceanic crust	Andes
Convergent (continent–continent)	Plates collide	Mountain range	Neither subducts	Himalaya
Transform	Plates slide past	Neither	Neither	San Andreas Fault

Worked example Beginner

The Red Sea lies between Africa and the Arabian Peninsula. Geologists recognize it as a young ocean basin in the making. How did it form, and what stage of the Wilson cycle does it represent?

Twenty-five million years ago, the land that is now eastern Africa and the Arabian Peninsula was a single continuous block of continental crust. Then tensional forces began pulling it apart. The crust stretched and thinned. Fractures opened, and basaltic magma rose through the cracks. A long, narrow rift valley developed, much like the East African Rift today.

As rifting continued, the valley floor dropped below sea level and seawater flooded in from the Indian Ocean through the Strait of Bab el-Mandeb. New oceanic crust began forming by seafloor spreading along the axis of the rift. The Red Sea now has a narrow band of young oceanic crust at its center, flanked by thinned and subsided continental crust on both sides.

The Red Sea represents an early stage of the Wilson cycle: continental rifting has progressed to the point where a narrow ocean basin has formed, but spreading has only produced a thin strip of new ocean floor. If spreading continues for tens of millions of years, the Red Sea will widen into a mature ocean like the present-day Atlantic.

At current spreading rates of about 1 to 2 centimeters per year, the Red Sea grows roughly 10 to 20 kilometers wider every million years. Over 100 million years, it could reach a width of 1,000 to 2,000 kilometers, comparable to the Atlantic at an early stage of its growth.

Check your understanding Beginner

Formal definition Intermediate+

Divergent boundaries

A divergent boundary is a plate boundary at which two lithospheric plates move away from each other. The separation is accommodated by extensional faulting and the intrusion of magma derived from decompression melting of the underlying asthenosphere. In oceanic settings, divergent boundaries are expressed as mid-ocean ridges. The ridge crest typically features an axial rift valley bounded by normal faults at slow-spreading ridges (e.g., Mid-Atlantic Ridge, full rate less than 4 cm/yr) or an axial high at fast-spreading ridges (e.g., East Pacific Rise, full rate greater than 8 cm/yr).

Continental rifting precedes ocean basin formation. The progression runs through several stages: (1) uplift and volcanism above a mantle thermal anomaly; (2) crustal stretching and normal faulting, forming a rift valley with lacustrine and fluvial sedimentation (e.g., East African Rift); (3) rupture of continental crust and onset of seafloor spreading, producing a narrow ocean with thinned continental margins (e.g., Red Sea); and (4) mature ocean with well-developed passive margins and a mid-ocean ridge (e.g., Atlantic Ocean).

Mid-ocean ridges are segmented by transform faults that offset the ridge axis. The segments between transforms are the loci of magmatic accretion. Between magmatic segments, non-transform offsets accommodate the ridge geometry through a combination of extension and strike-slip motion. The ridge axis is underlain by an axial magma chamber, typically 1 to 2 km wide and a few hundred meters thick, detected by seismic reflection profiling.

Convergent boundaries

A convergent boundary is a plate boundary at which two plates move toward each other. Because the Earth's surface area is approximately constant, convergence is balanced by subduction: one plate descends into the mantle beneath the other. The subducting plate carries oceanic lithosphere, sometimes with attached continental crust or sediment, down a dipping seismic zone known as a Wadati-Benioff zone.

Three principal subtypes are distinguished by the nature of the colliding plates.

Ocean–ocean convergence. One oceanic plate subducts beneath another. The overlying plate develops a volcanic island arc above the mantle wedge. Examples include the Izu-Bonin-Mariana arc in the western Pacific and the Tonga-Kermadec arc. The volcanic rocks are predominantly basalt and basaltic andesite. A forearc basin develops between the trench and the arc, and a back-arc basin may open behind the arc if slab rollback stretches the overriding plate.

Ocean–continent convergence. Oceanic lithosphere subducts beneath a continental plate. The continental margin develops a volcanic arc of andesitic to dacitic composition. The Andes are the type example: the Nazca Plate subducts beneath the South American Plate, producing the Peru-Chile Trench offshore and the Andean volcanic arc on land. Compression across the boundary also thickens the continental crust through folding and thrust faulting.

Continent–continent convergence. When two continental plates collide, the buoyancy of continental crust (average density approximately 2.7 g/cm${}^3$) prevents either plate from subducting deeply. Instead, the crust deforms by folding, thrust faulting, and thickening, producing a major orogenic belt. The Himalaya-Tibetan Plateau system formed as the Indian Plate collided with the Eurasian Plate beginning approximately 50 million years ago. The continental crust there reaches thicknesses of 70 to 80 kilometers, roughly double the normal continental crustal thickness.

The seismicity of subduction zones follows a characteristic pattern. Shallow earthquakes (depth less than 70 km) occur at the trench and in the overriding plate. Intermediate-depth earthquakes (70 to 300 km) and deep-focus earthquakes (300 to 700 km) occur within the subducting slab, defining the Wadati-Benioff zone. The deepest earthquakes occur at approximately 670 km depth, corresponding to the top of the lower mantle.

Metamorphic rocks in subduction zones record the high-pressure, low-temperature conditions of the descending slab and the overlying wedge. Blueschist-facies metamorphism (glaucophane + lawsonite) forms at pressures of 0.4 to 0.8 GPa and temperatures of 200 to 500 degrees Celsius. Eclogite-facies metamorphism (omphacite + pyrope-rich garnet) forms at still higher pressures, representing the transformation of subducted oceanic crust to a density comparable to the surrounding mantle.

Transform boundaries

A transform boundary is a plate boundary at which two plates slide horizontally past each other with neither creation nor destruction of lithosphere. The boundary is a vertical or near-vertical fault surface. J. Tuzo Wilson introduced the concept in 1965 to explain features connecting offset ridge segments on the ocean floor.

The distinction between transform faults and fracture zones is essential. A transform fault is the active segment between two offset ridge crests where the plates move in opposite directions. The fracture zone extends beyond the ridge-ridge offset on both sides; there the plates move in the same direction at the same rate, so no relative motion occurs. Fracture zones are topographic lineations but are not active plate boundaries.

On the ocean floor, transform faults produce long, linear valleys with steep walls. The depth of the valley reflects the age contrast across the fault: older, deeper seafloor on one side abuts younger, shallower seafloor on the other. Transform faults are also the sites of strong seismic activity, producing earthquakes with predominantly strike-slip focal mechanisms.

Seafloor spreading

Seafloor spreading is the process by which new oceanic lithosphere is created at mid-ocean ridges through magmatic accretion and moves laterally away from the ridge axis. The full spreading rate $v$ is the rate at which the two plates separate, typically 2 to 16 cm/yr. The half-spreading rate is $v/2$, the rate at which each plate moves away from the ridge.

Decompression melting drives magma generation. As the plates diverge, hot asthenosphere rises to fill the gap. The ascending peridotite crosses its solidus because pressure decreases faster than temperature along the adiabatic ascent path. The degree of partial melting, typically 10 to 20 percent, determines the composition and volume of the basaltic melt produced.

The oceanic crust produced by seafloor spreading has a characteristic layered structure. From top to bottom: layer 1 is pelagic sediment that accumulates on the basaltic crust; layer 2 is extrusive basalt (pillow lavas and sheet flows) and a sheeted dike complex; layer 3 is gabbro, the coarsely crystalline equivalent of basalt that cooled slowly in the magma chamber. The total thickness of oceanic crust is approximately 7 km, remarkably uniform worldwide.

The Wilson cycle

The Wilson cycle, named for J. Tuzo Wilson, describes the sequential opening and closing of an ocean basin. The canonical stages are: (1) embryonic rifting of a continent (e.g., East African Rift); (2) juvenile ocean with narrow seafloor spreading (e.g., Red Sea); (3) mature ocean with wide basin and passive margins (e.g., Atlantic); (4) subduction initiation at the margins and declining ocean (e.g., Pacific); (5) terminal ocean with continental collision imminent (e.g., Mediterranean); and (6) continental collision and suturing (e.g., Himalaya). The cycle then resets as thermal insulation beneath the assembled continent builds up and triggers renewed rifting.

The timescale for a full Wilson cycle is approximately 400 to 500 million years. Known supercontinents include Nuna (Columbia), assembled roughly 1.8 Ga; Rodinia, assembled roughly 1.1 Ga and broken apart roughly 750 Ma; Pannotia, assembled roughly 600 Ma; and Pangaea, assembled roughly 335 Ma and beginning to break apart roughly 200 Ma.

Key result: the Vine-Matthews-Morley hypothesis and spreading rate determination Intermediate+

In 1963, Fred Vine and Drummond Matthews at Cambridge, and independently Lawrence Morley of the Geological Survey of Canada, proposed that the linear magnetic anomalies observed parallel to mid-ocean ridges record reversals of the geomagnetic field frozen into the oceanic crust as it forms. This hypothesis provided the first quantitative test of seafloor spreading.

The mechanism is as follows. Basaltic magma erupted at the ridge axis cools through its Curie temperature (approximately 580 degrees Celsius for magnetite) and acquires a thermoremanent magnetization aligned with the ambient geomagnetic field. When the field reverses polarity (the north magnetic pole becomes the south magnetic pole), subsequent crust records the reversed direction. As spreading carries older crust away from the axis in both directions, a symmetrical sequence of normal and reversed magnetization stripes is produced.

The width $w$ of a magnetic stripe of known age $\Delta t$ directly gives the half-spreading rate:

$$v_{\text{half}}=\frac{w}{\Delta t}$$

For example, the Brunhes-Matuyama reversal boundary (age 0.78 Ma) is observed at approximately 15 km from the Mid-Atlantic Ridge axis near 30 degrees North. This gives a half-spreading rate of:

$$v_{\text{half}}=\frac{15\text{ km}}{0.78\text{ Myr}}\approx 1.9\text{ cm/yr}$$

The full spreading rate is twice this, approximately 3.8 cm/yr, consistent with rates determined independently from plate circuit reconstructions.

The magnetic polarity timescale, calibrated by radiometric dating of lava flows on land and by astronomically tuned sedimentary records, extends back more than 150 million years. By matching the anomaly pattern on the ocean floor to this timescale, the age of the ocean floor can be determined at any location. The ages increase symmetrically away from the ridge axis, providing a self-consistent test of the spreading model.

The rate of crustal production at a ridge segment of length $L$ is:

$$Q=v_{\text{half}}\cdot L\cdot d$$

where $d$ is the thickness of the newly formed crust (approximately 7 km). This quantity represents the volume rate of oceanic crust generation. Globally, the total crustal production rate is approximately 21 km${}^3$/yr, balanced by an equal rate of subduction.

Exercises Intermediate+

Plate boundary dynamics and the Wilson cycle Master

Force balance at plate boundaries

The motion of tectonic plates is governed by a balance of forces acting on the lithosphere. The principal driving forces are slab pull, ridge push, and basal drag. Slab pull arises from the negative buoyancy of the cold, dense subducting slab as it sinks through the mantle. Estimates place slab pull at 2 to 3 times the magnitude of ridge push, making it the dominant driving force for plates attached to subducting slabs. Ridge push results from the gravitational sliding of the elevated ridge flank down the sloping lithosphere-asthenosphere boundary. Basal drag is the viscous resistance (or assistance) exerted by the flowing asthenosphere on the base of the lithosphere.

Resistive forces include slab resistance (viscous and frictional resistance to subduction), mantle drag on the base of the plate, collision resistance at convergent boundaries, and transform resistance along strike-slip faults. The net torque on each plate must vanish in the steady state, a condition used to test global plate motion models for self-consistency.

Thermal structure of subducting slabs

A subducting slab enters the mantle with a thermal structure inherited from its age at the trench. Old oceanic lithosphere (age greater than 100 Ma) has a thick thermal boundary layer (lithospheric thickness approximately 100 km) and is colder and denser than the surrounding mantle at any given depth. This negative thermal anomaly drives slab pull. The temperature anomaly $\Delta T$ of the slab relative to the ambient mantle at depth $z$ governs its density contrast $\Delta \rho =\alpha {\rho }_0\Delta T$, where $\alpha$ is the coefficient of thermal expansion and ${\rho }_0$ is the reference density.

As the slab descends, it is heated by conduction from the surrounding mantle. The characteristic timescale for thermal equilibration is $\tau =d^2/\kappa$, where $d$ is the slab thickness and $\kappa$ is the thermal diffusivity (approximately $10^{-6}$ m${}^2$/s for mantle silicates). For a 100-km-thick slab, $\tau \approx 3\times 10^{17}$ s, or about 10 billion years. Slabs therefore remain colder and denser than the ambient mantle throughout the upper mantle, and many penetrate into the lower mantle before approaching thermal equilibration.

Slab rollback and back-arc spreading

When the rate at which the subducting slab sinks through the mantle exceeds the rate of plate convergence, the trench migrates oceanward. This process is called slab rollback. It is facilitated by a steepening slab dip and by the subduction of old, dense lithosphere. Slab rollback stretches the overriding plate, creating extension behind the volcanic arc. If extension is sufficient, the crust thins and seafloor spreading begins, forming a back-arc basin.

The western Pacific hosts the most active back-arc spreading systems. The Mariana Trough behind the Izu-Bonin-Mariana arc and the Lau Basin behind the Tonga arc are examples. Back-arc basin basalts have geochemical compositions intermediate between mid-ocean ridge basalt (MORB) and island arc basalt, reflecting the mixing of depleted asthenospheric melt with fluid-enriched mantle wedge material.

Obduction and ophiolite emplacement

Ordinarily, oceanic lithosphere subducts beneath continental or other oceanic lithosphere because it is denser. Obduction is the anomalous process by which slices of oceanic lithosphere are thrust onto continental crust during convergence. The resulting allochthonous body is an ophiolite, a section of oceanic crust and upper mantle preserved on land.

Well-studied ophiolites include the Troodos Ophiolite (Cyprus), the Oman Ophiolite, and the Bay of Islands Complex (Newfoundland). These exposures provide rare opportunities to study the structure and composition of oceanic lithosphere in situ. The classic "Penrose" ophiolite model, from bottom to top, consists of tectonized peridotite (mantle), layered ultramafic and gabbroic cumulates, massive gabbro, sheeted dike complex, pillow lavas, and pelagic sediment.

Triple junctions and their stability

A triple junction is the point where three plate boundaries meet. McKenzie and Morgan (1969) analyzed the geometry and kinematics of triple junctions, showing that a triple junction is stable if its position relative to all three plates remains constant through time. Stability depends on the boundary types and their relative orientations.

Of the sixteen possible combinations of boundary types (R = ridge, T = trench, F = transform fault) at a triple junction, only certain configurations are stable. The Ridge-Ridge-Ridge (RRR) triple junction is always stable regardless of geometry. The Bouvet triple junction in the South Atlantic (Africa, South America, Antarctica) is an RRR example. Ridge-Trench-Trench (RTT) and Trench-Trench-Trench (TTT) junctions may be stable depending on the relative orientations and velocities. An unstable triple junction must evolve by changing the boundary types or by adding new plate boundaries.

Kinematics of plate circuits and Euler poles

The instantaneous motion of any rigid plate on a sphere is described by an angular velocity vector $\mathbf{\omega }$ (the Euler vector) passing through the center of the sphere. The velocity of a point at position vector $\mathbf{r}$ on plate B relative to plate A is $\mathbf{v}={\mathbf{\omega }}_{AB}\times \mathbf{r}$. The relative angular velocity between plates A and C can be obtained from a plate circuit: ${\mathbf{\omega }}_{AC}={\mathbf{\omega }}_{AB}+{\mathbf{\omega }}_{BC}$.

Closure of a plate circuit provides a consistency test. The NUVEL-1A model of DeMets et al. (1994) determined Euler vectors for twelve plates using data from magnetic anomalies, transform fault azimuths, and earthquake slip vectors. Circuit closure residuals are typically less than 3 mm/yr, confirming plate rigidity to high precision.

True polar wander versus apparent polar wander

Apparent polar wander (APW) is the apparent motion of the Earth's rotation pole as seen from a single continent, reconstructed from paleomagnetic pole positions of known age. APW can result from either plate motion or reorientation of the entire solid Earth. True polar wander (TPW) is the latter: a rotation of the whole lithosphere (and mantle) relative to the spin axis, driven by mass redistribution.

TPW is detected by comparing APW paths from different continents. If all continents show the same synchronous shift in their apparent pole positions, the cause is TPW rather than differential plate motion. Estimates suggest TPW rates of 0.2 to 1 degree per million years at various times in Earth history, with several episodes of rapid TPW possibly triggered by changes in mantle density structure.

Wilson cycle timescales and supercontinent history

The periodicity of the Wilson cycle, roughly 400 to 500 Myr, reflects the time required for a supercontinent to assemble, insulate the underlying mantle, build a thermal anomaly, and rift apart. The geological record preserves evidence for multiple supercontinent cycles.

Nuna (also called Columbia) assembled approximately 1.8 to 1.5 Ga. Rodinia assembled approximately 1.1 Ga and fragmented approximately 750 to 600 Ma. Pannotia (or Vendia) assembled briefly approximately 600 Ma. Pangaea, the most recent supercontinent, assembled approximately 335 Ma and began fragmenting approximately 200 Ma. The causes of supercontinent breakup are debated: the insulation hypothesis attributes breakup to mantle overheating beneath the continental lid, while the external forcing hypothesis links breakup to changes in plate boundary forces propagated from subduction zones on the other side of the globe.

Evidence for past supercontinents comes from multiple independent sources. Paleomagnetic data constrain the latitude and orientation of each continent through time, enabling reconstruction of past configurations. Faunal provincialism — the distribution of distinctive fossil assemblages on different continents — records the timing of continental separation and collision. Matching geological provinces (orogenic belts, sedimentary basins, igneous provinces) across now-separated continents provide structural evidence for former connections.

Connections Master

Plate boundaries and earthquake hazards

The type of plate boundary controls the character and magnitude of earthquakes. The largest earthquakes (moment magnitude greater than 9) occur exclusively at convergent boundaries, where long fault segments can accumulate and release enormous strain. The 1960 Chile earthquake (Mw 9.5) and the 2011 Tohoku earthquake (Mw 9.0) both occurred at subduction zones. Transform boundaries produce frequent moderate-to-large earthquakes (up to approximately Mw 8). Divergent boundaries on land produce smaller earthquakes related to normal faulting in rift zones.

Seafloor spreading and ocean chemistry

Hydrothermal circulation at mid-ocean ridges exchanges heat and chemical elements between the ocean and the oceanic crust. Seawater percolates through the hot basalt near the ridge axis, leaching metals and other elements from the rock. The heated fluid returns to the ocean through hydrothermal vents (black smokers) at temperatures exceeding 350 degrees Celsius, enriched in iron, manganese, zinc, copper, and hydrogen sulfide. This process buffers ocean chemistry on geologic timescales and supports chemosynthetic ecosystems independent of sunlight.

The Wilson cycle and evolutionary biology

The opening and closing of ocean basins controls the geographic isolation and reconnection of terrestrial and shallow-marine populations. The breakup of Pangaea isolated dinosaur populations on separate continents, driving independent evolutionary trajectories. The closure of the Isthmus of Panama approximately 3 Ma created a land bridge that allowed the Great American Biotic Interchange: North American mammals migrated south and South American mammals migrated north. Faunal provincialism in the fossil record serves as both evidence for past continental configurations and a driver of speciation.

Plate boundaries and mineral deposits

Specific plate boundary settings concentrate particular mineral deposits. Porphyry copper deposits form above subduction zones where magmatic-hydrothermal fluids concentrate copper and molybdenum. Volcanogenic massive sulfide deposits form at divergent boundaries on the seafloor. Mississippi Valley-type lead-zinc deposits form in carbonate platforms on passive margins (divergent boundary interiors). The plate tectonic context is a primary exploration criterion for mineral resources.

Connections to mantle convection (Unit 27.01.03)

The forces at plate boundaries are ultimately driven by mantle convection. Understanding the force balance (slab pull, ridge push, basal drag) at boundaries motivates the deeper question of how the mantle convects, what its thermal and compositional structure looks like, and whether convection is layered or whole-mantle in style. These questions are addressed in the next unit.

Historical notes Master

Wilson and the transform fault concept (1965)

J. Tuzo Wilson, a Canadian geophysicist at the University of Toronto, introduced the concept of transform faults in 1965. Before Wilson, geologists were puzzled by the offsets in mid-ocean ridges. The fractures that displaced ridge segments appeared to be strike-slip faults, but the sense of motion was opposite to what the offset direction suggested. Wilson realized that the active fault segment between two offset ridge crests had the opposite sense of motion from the apparent offset. The fracture zone only appeared offset because the ridge had spread since the fault formed. This insight unified the geometry of plate boundaries and was a critical step toward the formal theory of plate tectonics.

Vine, Matthews, and Morley (1963)

In 1963, Fred Vine, a graduate student at Cambridge, and his supervisor Drummond Matthews published their analysis of magnetic surveys over the Carlsberg Ridge in the Indian Ocean. They observed that magnetic anomalies formed a symmetric pattern about the ridge axis and proposed that the anomalies recorded geomagnetic reversals frozen into the spreading ocean floor. Lawrence Morley independently reached the same conclusion from surveys over the Juan de Fuca Ridge but was initially unable to publish his paper. The hypothesis was met with skepticism until 1966, when Vine used the Jaramillo polarity event to confirm the predicted anomaly pattern, converting most remaining skeptics.

McKenzie and Morgan: triple junctions (1969)

Dan McKenzie and W. Jason Morgan published their analysis of triple junction stability in 1969, providing the kinematic framework for understanding how plate boundary networks evolve. They showed that the stability of a triple junction depends on the geometry and velocity vectors of its three arms, and that unstable triple junctions must evolve toward stability by changing their configuration. This work formalized the geometric constraints on global plate tectonics and demonstrated that the plate system on Earth is self-consistent.

Hess and seafloor spreading (1962)

Harry Hess of Princeton University proposed the concept of seafloor spreading in his 1962 paper "History of Ocean Basins." Hess suggested that mid-ocean ridges are sites of mantle upwelling where new oceanic crust is created, and that ocean trenches are sites where old crust is returned to the mantle. His hypothesis was based on the observed thinness of oceanic sediment, the discovery of guyots (flat-topped seamounts), and the absence of ocean floor older than the Mesozoic. Hess described his paper as "an essay in geopoetry," reflecting its speculative nature, but it proved remarkably prescient.

Dietz (1961) and the term "seafloor spreading"

Robert Dietz of the U.S. Navy Electronics Laboratory independently proposed a seafloor spreading model in 1961, actually coining the term "seafloor spreading" before Hess's paper appeared. Dietz emphasized the role of the mantle in driving the process and predicted the age progression of the ocean floor. Both Hess and Dietz are credited with the hypothesis, though Hess's formulation is more widely cited.

Bibliography Master

1. Tarbuck, F. K. and Lutgens, E. J., Earth Science, 15th ed. (Pearson, 2018), Ch. 2.

2. Kearey, P., Klepeis, K. A. and Vine, F. J., Global Tectonics, 3rd ed. (Wiley-Blackwell, 2009), Ch. 2–3.

3. Wilson, J. T., "A new class of faults and their bearing on continental drift," Nature 207 (1965) 343–347.

4. Wilson, J. T., "Did the Atlantic close and then re-open?," Nature 211 (1966) 676–681.

5. Vine, F. J. and Matthews, D. H., "Magnetic anomalies over oceanic ridges," Nature 199 (1963) 947–949.

6. McKenzie, D. P. and Morgan, W. J., "Evolution of triple junctions," Nature 224 (1969) 125–133.

7. Hess, H. H., "History of ocean basins," in Petrologic Studies: A Volume in Honor of A. F. Buddington (Geological Society of America, 1962) 599–620.

8. DeMets, C., Gordon, R. G., Argus, D. F. and Stein, S., "Effect of recent revisions to the geomagnetic reversal time scale on estimates of current plate motions," Geophys. Res. Lett. 21 (1994) 2191–2194.

9. Condie, K. C., Earth as an Evolving Planetary System, 3rd ed. (Elsevier, 2016), Ch. 9–10.

10. Dietz, R. S., "Continent and ocean basin evolution by spreading of the sea floor," Nature 190 (1961) 854–857.