H.F. Reid, The elastic-rebound theory of earthquakes, Univ. Calif. Pub. Bull. Dept. Geol. Sci., 6, 413-444, 1911. This paper was the transcription of the first of the Hitchcock Lectures delivered at the University of California, Berkeley, in the spring of 1911.


For example, the word “strength” in the first proposition would be modified to “frictional strength.” Reid clearly understood that the strength of faults was governed by the friction across fault surfaces, rather than the strength of intact rocks. In his 1911 paper, he discussed the role of friction in experiments on elastic rebound using jelly sheets (p. 422), recognizing that in nature “the surface rocks on opposite sides of the fault are not identical as is the jelly” (p. 430); he considered the role of fault friction in the generation of slip irregularities that cause strong ground motions (pp. 435-436), and in discussing slickensides on the limbs of folds, he states, “It seems quite certain that, as the rocks were being folded by horizontal pressure, the friction would at first prevent any such slipping of the strata; but as the elastic forces become stronger, slipping would occur suddenly with an elastic rebound of the adjacent strata, which would constitute an earthquake” (p. 437).


Although Reid’s formulation of the elastic rebound hypothesis preceded plate tectonics by more than 50 years, notions about the horizontal mobility of the Earth’s crust were already in the air. Reid sought to provide some geological mechanism for the “slow displacements” required by his theory in a footnote on p. 28 of his report: “Mr. Baily Willis, on account of the forms of the mountain ranges bordering the Pacific Ocean, has concluded that the bed of the ocean is spreading and crowding against the land. He thinks in particular that there is a general sub-surface flow towards the north which would produce strains and earthquakes along the western coast of North America.”


The story of the 1906 San Francisco earthquake presented here is based on modern reconstructions that have been updated using geophysical analysis techniques and geological knowledge that were unavailable at the time. An example is the estimation of the event’s nucleation point (hypocenter) and origin time. H.F. Reid’s original determination, given in his 1910 report, used local, imprecise estimates of the beginning of shaking to fix the origin time at 13:12:28 GMT and place the hypocenter at a depth of about 20 kilometers between the town of Olema and the southern end of Tomales Bay (i.e., north and west of San Francisco). Fortunately, this event was one of the first large earthquakes to be recorded by a global network of continuously recording seismometers. The network was sparse and poorly distributed by today’s standards, but the State Earthquake Investigation Commission was able to collect copies of seismograms or arrival times of seismic waves from 96 observatories, 83 of which were outside the conterminous United States. In 1906, the structure of the Earth’s interior was still too poorly known to predict the travel times of distant seismic waves, which is why Reid did not use them in his estimates. However, from an analysis of the teleseismic and local records reproduced in Reid’s report, B.A. Bolt (The focus of the 1906 California earthquake, Bull. Seis. Soc. Am., 58, 457-471, 1968) found that the epicenter was similar to that of the small March 22, 1957, earthquake (37.67°N, 122.48°W), from which he obtained an origin time of 13:12:21 GMT. This position, near the point on the San Francisco peninsula where the San Andreas goes out to sea, is consistent with further studies of the archived seismograms.


The Milne seismograph grew out of a collaboration of British scientists (Ewing, Gray, and Milne) in Japan in the early 1880s. They solved the problem of how to obtain a long-period response from a physically short pendulum by inclining it so that the gravitational restoring force is reduced. The Milne seismographs used a horizontal bracket pendulum to attain a period of about 12 seconds, and they were recorded photographically. The early history of seismometry is discussed by J. Dewey and P. Byerly (The early history of seismometry—Up to 1900, Bull. Seis. Soc. Am., 59, 183-227, 1969).


The Jesuits established seismographic stations at their educational institutions in Europe, North and South America, Asia, Africa, and Australia, instrumenting them first

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement