Cox, Allan Verne

views updated


(b. Santa Ana, California, 17 December 1926;

d. Woodside, California, 27 January 1987), geophysics, geomagnetic reversal timescale, geomagnetism, paleomagnetism, plate tectonics.

Cox’s most important scientific work concerned the development of the reversal timescale, work that he did with Richard Doell and G. Brent Dalrymple, and work that he accomplished during the half-dozen years after obtaining his PhD in 1959. Cox and Doell, both paleomagnetists, teamed up with Dalrymple, a geochronologist, and developed increasingly accurate reversal timescales. The timescale they and their competitors in Australia, Ian McDougall, Don Tarling, Hans Wensink, and F. H. Chamalaun, produced from 1963 through 1966 played a pivotal role in the confirmation of seafloor spreading.

Early Years . Cox was born on 17 December 1926 in Santa Ana, California, approximately 30 miles southeast of Los Angeles. His father was a housepainter. Although he had no formal education in electronics, he liked repairing radios for friends, and Cox remembered that their garage was always filled with electronic equipment. He had one sibling, a sister, Lois. His mother attended high school while raising her children, and graduated from high school the same year as Allan. After Cox had finished fifth grade at a local grammar school, his parents sent him to a Lutheran parochial school in a nearby town. He left after only three years, later attributing his drifting away from religion to his unrewarding stay at the school. Returning to public school, he attended local high school in Santa Ana, where he was inspired by his chemistry teacher, who had to compress a chemistry class into six weeks because of obligations to the U.S. Navy. Cox found the hard work refreshing, and decided that he wanted to be a chemist. He built a chemistry laboratory in his basement. Cox also read widely during his high school years, both fiction and nonfiction, and most every book in the local library on Sigmund Freud.

Encouraged to attend college by his teachers who recognized his abilities, Cox enrolled at the University of California at Berkeley in summer 1944 immediately after graduating from high school, planning to major in chemistry. However, he left after only one quarter and enlisted in the U.S. Merchant Marines for three years. With time to spare, he continued to read widely in different fields, including philosophy, anthropology, and science. He returned to Berkeley in 1948, thought about majoring in philosophy, but decided to continue majoring in chemistry. Although he enjoyed his chemistry courses, he did not get to know his professors and found the department impersonal. He also was demoralized by the rise of McCarthyism, felt alienated, and his grades suffered.

In winter 1950, during his junior year, Cox was introduced to the geologist Clyde Wahrhaftig, who was studying rock glaciers in the Alaska Range for the U.S. Geological Survey (USGS). Wahrhaftig asked Cox to be one of his assistants for the upcoming summer. Cox accepted. Unlike his experience in Berkeley’s chemistry department, Cox found his coworkers in Alaska open and friendly. Cox also enjoyed the fieldwork, and found the geological problems intellectually stimulating. He began reading geology for pleasure.

Cox continued to major in chemistry, but found it even harder to concentrate on his studies. The 1950–1951 academic year proved to be his undoing, and undergoing a personal crisis, he left school in spring 1951 before completing his courses and spent a month by himself hiking through the Sierra Nevada. As a result, he lost his student deferment, and was drafted by the U.S. Army. Before reporting for active duty in October 1951, he spent another summer with Wahrhaftig and his colleagues studying rock glaciers in Alaska. He decided that he would study Earth science once he left the army. Wahrhaftig continued to send Cox reading in geology during Cox’s army years. Cox spent another summer (1954) working with Wahrhaftig in Alaska, and they eventually coauthored a major paper (1959) on rock glaciers in the Alaska Range.

After his discharge from the army, Cox returned to Berkeley in fall 1954, and obtained a BA degree in geophysics in June 1955. He decided on geophysics instead of geology, partly because he already had taken almost enough courses in mathematics and physics to complete majors in both subjects. He found his professors in the Earth sciences much more personal than his previous professors. He particularly liked Perry Byerly’s class in geophysics. After doing very well on his midterm exam, Cox was asked by Byerly if he had thought about a career in geophysics. Cox later remarked that his decision to become a geophysicist grew out of his talks with Byerly and the personal attention he received.

Cox obtained a Stanford Oil Company of California scholarship in geophysics, and became a graduate student at Berkeley in geophysics in September 1955. Although he found Byerly’s subject of seismology interesting, he wanted to find something less mainstream or “Big Science.” John Verhoogen offered him just such an opportunity. Verhoogen’s former student, Richard Doell, had set up a paleomagnetics laboratory at Berkeley, and Verhoogen identified Cox as Doell’s successor. Doell was completing his PhD work when Cox was a senior. Doell showed him how to use the magnetometer, find appropriate rocks in the field, and prepare them for measurement on the magnetometer. Cox agreed to work in paleomagnetism. With Doell and Cox, Verhoogen was doubly blessed with two wonderful students. Doell completed his PhD in spring 1955, left Berkeley for a one-year position in geophysics at the University of Toronto, and Cox took over the magnetics laboratory.

Work on Paleomagnetism . When Cox began working in paleomagnetism, practitioners were working on two central problems. The first was continental drift and polar wander. A group of scientists at Cambridge University—S. Keith Runcorn, Edward Irving, K. M. Creer, and Jan Hospers—had begun using paleomagnetism to test continental drift and polar wander. Creer developed what came to be called an apparent polar wander (APW) path for Great Britain based on his own findings and those of Irving and Hospers. He argued for polar wander and continental drift, invoking the latter to explain discrepancies between his APW path and his reworking of John Graham’s 1949 finding from North America. Graham, a U.S. paleomagnetist, had begun working in the subject in the late 1940s at the Carnegie Institution in Washington.

Irving found paleomagnetic support that India had drifted from the Southern to the Northern Hemisphere as Alfred Wegener and others had proposed. Runcorn collected samples in North America and found discrepancies between the pole positions of North America and Great Britain that could be explained in terms of continental drift, but he attributed them to systematic errors and opted for polar wander without drift. Irving moved to the Australian National University in 1954, set up a paleo-magnetic laboratory, began to develop a polar wander path for Australia, and argued for continental drift and polar wander, given discrepancies in APW paths or pale-opoles for North America, Great Britain, India, and Australia (1956).

Runcorn, who moved to Newcastle, changed his mind (1956) and adopted polar wander and continental drift. P. M. Du Bois, one of Runcorn’s students, collected samples from North America and argued (1955) in favor of continental drift. Graham also continued working in paleomagnetism, sampling rocks from the Grand Canyon during the summer of 1954. Creer soon joined Runcorn at Newcastle, and, in 1956, began paleomagnetic work in South America. A. E. M. Nairn, recruited by Runcorn, started working in Africa in 1956. Another group of pale-omagnetists from Great Britain, begun and overseen by Patrick M. S. Blackett, and run on a daily basis by John A. Clegg, began supporting continental drift in 1954. They confirmed Irving’s Indian results (1956), added more results from Spain and France, and argued in favor of continental drift. Back in the United States, Graham (1965) and Doell (1956) independently obtained samples from the Grand Canyon. Both tentatively endorsed polar wander but rejected continental drift.

The other problem was that of reversals. About one-half of the rocks that had been paleomagnetically studied had normal polarity and the other half had reversed polarity. Reversals were interpreted in either of two ways; either as reversals of the geomagnetic field or as self-reversals. Graham (1953) favored self-reversals as a general explanation for reversed rocks. Having found what he thought were reversed rocks, he asked Louis Néel, future Nobel laureate for his theoretical work on ferromagnetism and antiferromagnetism, if there were ways for rocks to undergo a self-reversal. Néel (1951) suggested four possible mechanisms, and in 1951, Seiya Uyeda found a self-reversing rock from a volcano in Japan that confirmed the existence of one of Néel’s mechanisms. However, Hospers and Antonio de la Roche from France defended field reversals during the early 1950s, and most paleomagnetists agreed with them. Hospers and Roche had begun to develop reversal timescales. Runcorn, Irving, and Creer also found reversed rocks, and preferred field reversals.

Verhoogen, learning of Irving’s determination of an Eocene paleopole for India, encouraged Cox to obtain an Eocene-aged paleopole for North America to compare it with the Indian paleopole. Parke D. Snavely Jr., who had mapped and named the Siletz River volcanics of northwestern Oregon in 1949, suggested that Cox sample them, in part, because they could be dated paleontologically. Cox began collecting samples in 1955, and submitted his results in a note to Nature in January 1957. He studied eight flows spread through about 4,000 feet. He found normal and reversed rocks. He applied appropriate field tests and magnetically cleaned his samples. His data were clearly reliable. He calculated a paleopole at latitude 37° north and longitude 50° west. But there was a problem, at least for those who supported mobilism: the pole position for the Siletz River volcanics was very close to Irving’s similarly aged Indian paleopole. Cox opted for rapid polar wander, rejected continental drift, and noted that if polar wander is very rapid, it would be very difficult to obtain robust paleomagnetic evidence for continental drift. Indeed, Cox decided to no longer actively work on continental drift.

Work on Reversals . Cox decided to turn to the problem of reversals. He found the problem intrinsically interesting, and had found reversals among his Siletz River volcanics. At the time Verhoogen was inclined toward self-reversals. At first Cox agreed, but began to question the idea of self-reversals after reading Hospers’s papers. With the help of Harold Malde of the USGS, he collected samples from the Snake River Plain of Idaho. He found normal and reversed rocks, and tested various self-reversal hypotheses, including those developed by Néel. None applied, and he also discovered that the stratigraphic position of the normal-reversed horizon he found agreed with the position found by others working on different continents. This led him to favor field reversals in his 1959 PhD thesis, “The Remanent Magnetization of Some Cenozoic Volcanic Rocks.”

Doell, who had moved from Toronto to the Massachusetts Institute of Technology, returned to Berkeley in 1958 to teach a summer class. He and Cox discussed recent work in paleomagnetism, found that they were in general agreement, and decided they would work together on a long-range project at the USGS in Menlo Park, California. They began their project in April 1959, and were greatly helped by James Balsley, then chief of the Geophysics Branch of the USGS. Cox already had decided to work at the USGS. Balsley, who had worked in paleomagnetism and favored self-reversals, knew both Cox and Doell, and wanted to get the USGS involved in rock magnetism.

Balsley was also instrumental in getting Cox and Doell to write a lengthy review of paleomagnetism. Balsley was asked by the Geological Society of America to write such a review, but he declined and recommended Cox and Doell. They finished the review in 1960. Unlike almost every major paleomagnetist except for Graham, they were unwilling to favor continental drift. Although some of its paleomagnetic support, in particular Irving and Green’s 1957 APW path for Australia, was highly suggestive of continental drift during the PermoCarboniferous, they argued that more work was needed before acceptance of continental drift was warranted. Cox also continued to maintain that proximity of the contemporaneous paleopole positions of the Siletz River volcanics and India weakened the support for continental drift. Irving (1959) had proposed a clockwise rotation of the Oregon region that would place the paleopole of the Siletz River volcanics in line with other North American pale-opoles and away from the Indian paleopole. Cox rejected Irving’s alternative explanation. Although Cox and Doell (1961) were slightly more inclined toward mobilism in a second review, they still were unwilling to accept continental drift. Indeed, Cox (1965) continued to resist continental drift during the next few years as evidenced by his review of Irving’s 1964 textbook on paleomagnetism, in which he suggested that Irving had exaggerated the paleo-magnetic support for continental drift.

Cox and Doell decided to determine the time intervals between successive reversals of the geomagnetic field during the past several million years, which would accomplish by tracking the recent history of reversals through the study of temporally overlapping lava flows. This meant that they would need someone trained in radio-metric dating of young rocks. They tried without success to get Jack Evernden and Garniss Curtis at Berkeley; both were working on other projects. Learning about Dalrymple from Wahrhaftig, they successfully recruited him at the end of 1962. Dalrymple, who received his PhD in 1963 under Evernden, began working on the project in early 1963. However, even after the National Science Foundation granted the USGS $120,000 to support Cox, Doell, and Dalrymple’s project, they were refused permission to build their own mass spectrometer until early 1964. Indeed, they threatened to resign unless they received permission. Once they managed to get their spectrometer fully operational in August 1964, they were able to begin dating their samples much more quickly than when dependent on the use of other instruments.

When they began their project, they, like everyone else including their competitors at the Australian National University (ANU), mistakenly assumed that field reversals were periodic instead of episodic. Both groups began to realize that reversals might be episodic by 1964, but Cox, Doell, and Dalrymple were the first to incorporate the idea into a reversal timescale. In their third timescale, which was published in June 1964, they distinguished between long-term reversals, which they called epochs, and short-term reversals, which they called events. Epochs lasted from 0.9 to 1.4 million years; events were about one-tenth as long. Deciding to name reversal-epochs after deceased geophysicists who had increased the understanding of Earth’s magnetism, they called the current normal epoch the Brunhes, the first reversed epoch the Matuyama, and the third epoch the Gauss. The Brunhes, they claimed, began 0.9 million years ago; the Matuyama lasted for 1.4 million. Naming events in terms of the location of the rocks used in their discovery, they identified two events: the normal Olduvai within the Matuyama, and the reversed Mammoth in the Gauss.

Continuing to refine their timescale, Cox collected samples in the Galapagos Islands and on Nunivak Island off the coast of Alaska in the Bering Sea, and Doell and Dalrymple returned to the Jemez Mountains in New Mexico to obtain more samples. The samples from the Jemez Mountains included three reversed ones with ages ranging from 0.71 and 0.73 million years, an intermediately magnetized one aged 0.88 million years, and a normally magnetized one aged 0.89 million years. Doell and Dalrymple (1966) concluded that they had found another event, which they named the Jaramillo. They had two options. They could keep the boundary between the Brunhes and Matuyama at 1.0 million years and place the new event within the Brunhes as a reversed event, or they could redefine the boundary between the Brunhes and Matuyama at 0.7 million years, and place the event within the reversed Matuyama as a normal event, which ended 0.89 million years ago. They chose the latter because they thought it would be stratigraphically more useful. McDougall, H. L. Allsopp, and Chamalaun (1966) of the ANU group independently discovered the same event, but their paper did not appear until after Doell and Dalrymple’s paper. The ANU group found three normally magnetized samples aged 1.01 million years, which established when the Jaramillo event began.

Dalrymple told Fred Vine about the discovery of the Jaramillo event at the Geological Society of America’s November 1965 meeting. Vine, who had obtained changing seafloor spreading rates using their previous timescale that did not include the Jaramillo, immediately realized that the new timescale would give him constant spreading rates. Thus, their timescale played an important role in confirming seafloor spreading, and thereby continental drift. Ironically, Cox had decided not to use paleomagnetism to test continental drift when he turned to the problem of reversals, but ended up providing unexpected paleomagnetic support for it after all.

Although Cox’s most significant scientific achievement was the work he did with Doell and Dalrymple, he continued to contribute to the understanding of secular variation, nondipole features of Earth’s magnetic field, and the origin of reversals, and he discovered a geomagnetic excursion. He also worked in tectonics, even returning to the Oregon Coast range. With the triumph of plate tectonics, Cox accepted Irving’s rotational analysis of the mountain range, and he and his students developed more and more refined reconstructions of the region. Cox wrote more than a hundred papers, and two textbooks. The first, Plate Tectonics and Geomagnetic Reversals (1973), contained many of the important early papers surrounding the development and early application of plate tectonics and development of the reversal timescale. The second, Plate Tectonics: How It Works (1986), coauthored with Robert Brian Hart, remains one of the best introductions to the working of plate tectonics.

Cox spent much of his career at Stanford University. Securing a position as research associate at Stanford in 1961, he became professor of geophysics in 1967, and dean of its School of Earth Sciences in 1979, a position that he continued to hold at his death in 1987. Cox was a popular teacher with both graduates and undergraduates. In recognition of his working with undergraduates, Stanford University established the Allan Cox Medal for Faculty Excellence Fostering Undergraduate Research. The Geological Society of America also established the Allan V. Cox Student Research Award, which is given annually in recognition of outstanding student research involving application of geophysical principles and techniques.

Cox was elected to the National Academy of Sciences in 1969 and to the American Philosophical Society in 1984. He was awarded the John Adam Fleming Medal of the American Geophysical Union (1969), the Arthur L. Day Medal of the Geological Society of America (1975), the Arthur L. Day Prize of the National Academy of Sciences (1984), and the Vetlesen Prize from Columbia University, which he shared with Doell and Runcorn (1971). He also served as president of the American Geophysical Union (1978–1980).

Cox continued to display a wide range of interests in the arts. He loved ballet, theater, music, and visual arts. He even became the artist Walter De Maria’s geological consultant for some of his outside installations, such as The Lightning Field in New Mexico. He loved the outdoors, and spent part of most summers backpacking in the Sierra Nevada.

Cox took his own life the night of 27 January 1987. An expert cyclist, he rode his bicycle off a descending mountain road before it started to curve, crashing head-on into a large tree. The parents of a boy with whom Cox had allegedly had repeated sexual contact since the child was fourteen had recently told him that they had reported him to the police. The child’s father had studied for his PhD under Cox. This crisis seems to have precipitated Cox’s suicide: he had mentioned suicide to others, and several books on how to make wills were found at his home the day after his death.


Cox’s papers are housed at the Stanford University Archives.


“Remanent Magnetization of Lower to Middle Eocene Basalt Flows from Oregon.” Nature179 (1957): 685–686.

With Clyde Adolph Wahrhaftig. “Rock Glaciers in the Alaska Range.” Geological Society of America Bulletin 70 (1959): 383–436.

With Richard R. Doell. “Review of Paleomagnetism.” Geological Society of America Bulletin 71 (1960): 645–768.

———. “Paleomagnetism.” Advances in Geophysics 8 (1961): 221–313.

With Richard R. Doell and G. Brent Dalrymple. “Reversals of the Earth’s Magnetic Field.” Science 144 (1964): 1537–1543.

Review of Paleomagnetism and Its Application to Geological and Geophysical Problems by E. Irving. Science 147 (1965): 494.

With Richard R. Doell and G. Brent Dalrymple. “Time Scale for Geomagnetic Reversals.” In The History of the Earth’s Crust, edited by Robert A. Phinney. Princeton, NJ: Princeton University Press, 1968.

Plate Tectonics and Geomagnetic Reversals. San Francisco: W.H. Freeman, 1973.

With Robert Brian Hart. Plate Tectonics: How It Works. Palo Alto, CA: Blackwell Scientific Publications, 1986.


Banerjee, Subir K. “The Scientific Work (1957–1987) of Allan Cox.” Journal of Geophysical Res earch 93(B10) (1988): 11563–11568.

Coe, Rob, and Brent Dalrymple. “Allan Cox 1926–1987.” Eos 68, no. 19 (1987): 513–514.

Doell, Richard R., and G. Brent Dalrymple. “Geomagnetic Polarity Epochs: A New Polarity Event and the Age of the Brunhes-Matuyama Boundary.” Science 152 (1966): 1060–1061.

Glen, William. The Road to Jaramillo: Critical Years in the Revolution of Earth Science. Stanford, CA: Stanford University Press, 1982.

Irving, E. “Palaeomagnetic Pole Positions.” Geophysical Journal of the Royal Astronomical Society2 (1959): 51–79.

Krauskopf, Konrad B. “Allan V. Cox.” Biographical Memoirs, National Academy of Sciences 71 (1977): 17–31.

McDougall, Ian, H. L. Allsopp, and F. H. Chamalaun. “Isotopic Dating of the Newer Volcanics of Victoria, Australia, and Geomagnetic Polarity Epochs.” Journal of Geophysical Research 74 (1966): 6107–6118.

Netzer, Baie. “Officials Say Cox’s Death Was Suicide.” Stanford Daily, 30 January 1987.

Henry Frankel