Sir Harold Jeffreys is a world authority in theoretical geophysics. He was born in Northumbria (the northeast of England) and educated at Armstrong College (now the University of Newcastle-upon-Tyne) and Cambridge University. He is now a Senior Fellow of St. John's College, Cambridge. He has published over 300 scientific papers and is the author of 7 books, including Theory of Probability and Mathematical Physics (with his wife, Lady Bertha Swirles Jeffreys). Sir Harold has made innumerable theoretical contributions to seismology. Many of these are documented in his book The Earth, which has been published in six editions. His papers have recently been collated by Gordon and Breach (Publishers) into six volumes, Collected Papers of Sir Harold Jeffreys on Geophysics and Other Sciences. Some idea of the breadth of this research can be seen from the individual volume titles: "Theoretical and Observational Seismology," "Observational Seismology," "Gravity," "Dissipation of Energy and Thermal History," "Astronomy and Geophysics," and "Mathematics, Probability and Miscellaneous Other Sciences."

Sir Harold has also served as Director of the International Seismological Summary (ISS). Among the honors bestowed upon him by various scientific organizations are Fellow of the Royal Society, Royal and Copley Medals of the Royal Society, Foreign Associate of the U.S. National Academy, Bowie Medalist of the American Geophysical Union, Medalist of the Seismological Society of America, and the Vetlesen Prize.

In the citation for the Medal of the Seismological Society of America, his colleague Dr. Ralph Lapwood noted that Sir Harold still rides his bicycle in nearly all weathers and maintains his lifelong hobby of natural history.

HS: How did you become interested in seismology?
JEFFREYS: I suppose I had been interested in astronomy since I was about 10. When I was an undergraduate, my interest in geophysics was aroused by a popular book on the tides by Sir George Darwin (Charles Darwin's son). The great majority of my work has arisen directly or indirectly from this book and from his collected papers. My first paper dealing with the Earth resulted from reading E. J. Routh's book, Advanced Rigid Dynamics. I came across a section on a difficulty in the theory of the figure of the Moon. The point he made was that the Moon's dynamical ellipticities are about 16 times what they would be if the Moon had the hydrostatic form for its present distance. This made me think about strength and elasticity. In 1914, the only evidence I had to go on was some information on the velocity of Rayleigh waves and some Earth models with a homogeneous shell and core put forward by E. Wiechert in Germany. Anyhow, I worked out some possible structures for the figure of the Earth. I decided that the Moon might have solidified at a time when it was considerably nearer the Earth than it is now and might have retained the form it then had.

HS: When did you become involved with geology?
JEFFREYS: My first paper on mountain building was stimulated by Arthur Holmes, who introduced me to a hypothesis in Suess' Das Antlitz der Erde for a granitic layer resting on a basic one. This idea was confirmed by Dorothy Wrinch and me in the paper dealing with seismic waves from the explosion at Oppau in Bavaria in 1921. It was the starting point for a long series on seismology. The standard travel times of seismic waves that were being used then (especially in the ISS) were those of Zoppritz of 1907. Considerable corrections to the tables had been made by A. and S. Mohorovicic and by B. Gutenberg; the central core was discovered by R. D. Oldham. Later studies by H. H. Turner, P. Byerly, and Father J. B. Macelwane, S. J., agreed fairly well, although in some respects they differed appreciably from the earlier work. A succession of papers that I, at first alone and later with K. E. Bullen, published led to the Jeffreys-Bullen tables of 1940.

HS: Which do you regard as the most significant of your contributions to seismology?
JEFFREYS: Three, I would say. First, the definite statement in 1926 that the Earth's core is liquid. Gutenberg was near such a solution, but he still maintained at that time that transverse waves might get into the core and be heavily damped. Second, the Jeffreys-Bullen tables of travel times of seismic waves that were published in collaboration with K. E. Bullen in 1940. [These give the travel times of all the main seismic phases from foci at different depths to surface stations at different distances.] Third, the studies of travel times of primary (P) waves traveling short distances on the Earth [from near earthquakes at distances of] up to 10 degree degrees or so. These established that there were substantial differences in the structure of the Earth's interior under Europe and Japan. They have since been greatly expanded by the work of E. Arnold, M. L. Gogna, and M. Shimshoni.

HS: What particular recollections do you have of your term as Director of the ISS?
JEFFREYS: I don't think I did much except administration for the ISS; J. S. Hughes was responsible for the actual calculations. At that time, many observations of additional phases were published and were very useful, but, because of the expense involved and the increasing emphasis on speed of preparation, it was difficult to justify continuing to list them. Speed may be desirable for its own sake but, in the long run, contributes little to the advancement of a subject. The reduction of reports on the less studied phases is a serious loss; these are precisely those that may lead to more knowledge about the structure of the Earth.

Many of the great advances in seismology have been made by "special studies" in which the research worker collects the original records from the stations and reads them himself. Others have relied on the routine observations submitted to ISS. Both methods have their advantages. The chief advantage of the special study is that the investigator knows what he is looking for; if it is present, he will find it, and, if he is wise, he will equally be able to recognize when it is not. The chief advantage of using the ISS is that the most laborious parts of the work have already been done by the observatory and the ISS staff. In addition, the ISS observations are more numerous and permit better determinations of the epicenter.

When I was Director, about 600 earthquakes were analyzed every year. Up to my time, the calculations had all been done with four-figure logarithmic tables and three-figure tables of squares. I lent ISS an ancient Marchant multiplying machine for a while. Whereas these ordinary machines might take a day to calculate a solution, modern calculating machines may take a few seconds. Nevertheless, plenty of problems remain, for it takes weeks to write a program and sometimes days to get access to a high-speed machine.

HS: Since you published the first edition of The Earth, what do you consider have been some of the major advances in seismology?
JEFFREYS: A most important advance was H. H. Turner's discovery of deep-focus earthquakes in 1922. The evidence was not altogether satisfactory, but much better evidence was produced in 1931 by R. Stoneley and F. J. Scrase. Recording of P waves has greatly improved, and the number of recording stations and the quality of the instruments have much increased. Unfortunately, observations of secondary (S) waves, which are more interesting in general geophysics, have become worse. The use of explosions for obtaining travel times (something I had recommended in 1924) and especially for determining the structure of the ocean floor has led to some big advances. Synthetic seismograms have been constructed for known earthquake sources, and they're very much like actual records. And, of course, much more has been learned about the structure of the Earth's core.

HS: Which of the early figures in seismology do you recall?
JEFFREYS: I never knew Sir George Darwin personally, although I think that nearly all of my work has been inspired by him. When I was taking walks on Coe Fen near Cambridge, I often used to see a man practicing archery. I found out much later that Darwin was an archer, and his home was nearby. So I must have seen him without knowing who he was.

I did know R. D. Oldham. He first identified a true S wave from records of the Assam earthquake of June 12, 1897. Although P and S waves had been predicted theoretically by Poisson in 1829, it was not until the end of the century that they were clearly distinguished in actual records. The problem was due partly to the low magnification and inadequate damping in the early instruments. It was also due partly to the fact that earthquakes send out other waves that travel over the surface, and, at large distances, these are much larger than those predicted by Poisson. Oldham constructed the first useful travel-time tables of P and S, and, in 1906, he published his discovery of the Earth's inner core [in the Journal of the Geological Society of London]. He was mainly a geologist and claimed that he was not really interested in seismology. He was the only man I have known that did first-rate work in a subject that did not interest him. I also knew Gutenberg and Byerly well.

HS: What areas of seismology should be emphasized more?
JEFFREYS: Imperfections of elasticity in the Earth. There's a great deal of evidence for them, but the best data seem to forbid any estimates to within a factor of 2. More work is needed on accurately measuring S waves and observing oceanic earthquakes. More attention to probability theory is needed. It is still the exception for estimates to be given with any proper statement of uncertainty. Most studies pay no attention to statistical significance for which fairly simple tests are available.

HS: Your writings have a lively humor. Are scientists too stuffy?
JEFFREYS: I can't really say. Many of them don't take readily to new ideas, and others, a bit too readily.

Abridged from the Earthquake Information Bulletin. Vol. 12, No. 2, March - April, 1980.

Sir Harold Jeffreys died 18 March 1989.
More Harold Jeffreys links