One of the keys to plate tectonics was the discovery that the Earth's magnetic field has reversed its polarity 170 times in the last 80 million years. As new basaltic material is squeezed up into the mid-ocean cracks and solidifies, it is magnetized according to the polarity of the Earth's magnetic field. If the field reverses its polarity, the strip of new material is magnetized in an opposite sense. As the oceanic floor continues to spread, the new strips of rock are carried away on either side like a conveyer belt.
Using these magnetic strips as evidence of movement, it became obvious that the Earth's surface consisted of a mosaic of crustal plates that were continually jostling one another. If the Earth was not to be blown up like a balloon by the continual influx of new volcanic material at the ocean ridges, then old crust must be destroyed at the same rate where plates collide. The required balance occurs when plates collide, and one plate is forced under the other to be consumed deep in the mantle.
We now know that there are seven major crustal plates, subdivided into a number of smaller plates. They are about 80 kilometers thick, all in constant motion relative to one another, at rates varying from 10 to 130 millimeters per year. Their pattern is neither symmetrical nor simple. As we learn more and more about the major plates, we find that many complicated and intricate maneuvers are taking place. We learn, too, that most of the geological action - mountains, rift valleys, volcanoes, earthquakes, faulting - is due to different types of interaction at plate boundaries.
How are earthquakes connected with plate tectonics? In 1969, Muawia Barazangi and James Dorman published the locations of all earthquakes that occurred from 1961 to 1967. Most of the earthquakes are confined to narrow belts and these belts define the boundaries of the plates. The interiors of the plates themselves are largely free of large earthquakes, that is, they are aseismic. There are notable exceptions to this. An obvious exception is the 1811-1812 earthquakes at New Madrid, Missouri. Another is the 1886 earthquake at Charleston, South Carolina. As yet there is no satisfactory plate tectonic explanation for these isolated events; consequently, we will have to find alternative mechanisms.
Plate tectonics confirms that there are four types of seismic zones. The first follows the line of mid-ocean ridges. Activity is low, and it occurs at very shallow depths. The second type of earthquake associated with plate tectonics is the shallow-focus event unaccompanied by volcanic activity. The San Andreas fault is a good example of this, so is the Anatolian fault in Northern Turkey. In these faults, two mature plates are scraping by one another. The friction between the plates can be so great that very large strains can build up before they are periodically relieved by large earthquakes. The third type of earthquake is related to the collision of oceanic and continental plates. One plate is thrust or subducted under the other plate so that a deep ocean trench is produced. This type of earthquake can be shallow, intermediate, or deep, according to its location on the downgoing lithospheric slab. Such inclined planes of earthquakes are known as Benioff zones. The fourth type of seismic zone occurs along the boundaries of continental plates. Within this zone, shallow earthquakes are associated with high mountain ranges where intense compression is taking place.
In summary, plate tectonics is a blunt, but, nevertheless, strong tool in earthquake prediction. It tells us where 90 percent of the Earth's major earthquakes are likely to occur. It cannot tell us much about exactly when they will occur. For that, we must study in detail the plate boundaries themselves. Perhaps the most important role of plate tectonics is that it is a guide to the use of finer techniques for earthquake prediction.