North Pole, South Pole: The Epic Quest to Solve the Great Mystery of Earth's Magnetism

by Gillian Turner Phd

  He suggested instead that iron particles in the crust were the source of Earth’s magnetic field. In accordance with contemporary notions of a fluid-filled Earth, he reckoned that the solid crust must be relatively thin, but that it was gradually thickening as the fluid beneath solidified on to it, becoming magnetized in the process and so causing local changes in the direction and intensity of the magnetic field. He supported this theory with the observation that the geomagnetic intensity was greatest at the poles where the temperature was lowest, and where you might therefore expect the magnetized crust to be thickest.

  Gauss’s spherical harmonic analysis of the geomagnetic field was published in 1838 in his Allgemeine Theorie des Erdmagnetismus, but by then Gauss was already heavily involved in another project, the establishment of Göttingen Magnetische Verein, a global network of geomagnetic observatories.

  By the late 1820s Alexander von Humboldt had become absorbed in studying the rapid geomagnetic time variations that the London instrument-maker George Graham had first noticed in 1722, and that had won Charles-Augustin de Coulomb the prize of the Paris Académie des Sciences in 1777. The disturbed days, which Celsius and Hiorter had correlated with auroras, interested von Humboldt most. He had studied them briefly in 1806, naming them “magnetic storms,” but now he made it his mission to carry out a more systematic study of these strange high-frequency disturbances by obtaining simultaneous magnetic measurements from as many different locations as he could.

  He began with his own geomagnetic observatory, which he built entirely of non-magnetic materials in the Berlin garden of his friend Abraham Mendelssohn, father of the famous composer, and here he painstakingly collected measurements hour by hour, day and night. He went on to draw all his European colleagues into the project, including François Arago in Paris and Gauss, who coordinated a series of simultaneous measurements at observatories in Germany, Sweden and England. He even persuaded Tsar Nicholas I of Russia to build ten new observatories—from Saint Petersburg to Sitka in Alaska (then in Russian hands), and from Arkhangel in the north to Beijing, where the observatory was erected in the grounds of the Russian Orthodox monastery.

  Next, he petitioned the president of the Royal Society to set up observatories in British territories the world over, particularly near the equator, at high latitudes, and in the southern hemisphere. This was successful: in 1839 Her Majesty’s Government dispatched two ships under the command of James Ross, a naval officer who had located the north magnetic pole—the place where the dip needle came to rest absolutely vertical—on a previous voyage to the Arctic in 1831. On board one of the ships was Edward Sabine, who was charged with overseeing the establishment of magnetic observatories at Toronto, Saint Helena, the Cape of Good Hope and Hobart (then called Hobarton) in Tasmania.

  Magnetic observatories relied on accurate instruments to gather data. This magnetometer was used to measure declination and the horizontal component of intensity of Earth’s magnetic field, particularly in Britain, the British colonies and the United States from the midnineteenth to mid-twentieth centuries. It was designed by Francis Ronalds, director of the Kew Meteorological Observatory in London.

  Sabine’s background and credentials had made him the natural choice to lead the British front in what would become known as the “Magnetic Crusade.” He was now a major of the British Admiralty, and shortly to receive a knighthood. As well as writing a long review of Hansteen’s work, and an even longer report to accompany his charts of global geomagnetic intensity, he had carried out land-based magnetic surveys, first in his native Ireland, then in Scotland and England. He had also been involved in several expeditions in search of the elusive Northwest Passage, which was thought to cut through the icy reaches of the Arctic between the Atlantic and Pacific oceans, and had been the astronomer on Ross’s expeditions.

  The Kew pattern dip circle. Until the mid-twentieth century this instrument was widely used to measure the inclination to the horizontal of Earth’s magnetic field at geomagnetic observatories.

  When the Board of Longitude was finally disbanded in 1828, having long since fulfilled its goal of achieving accurate determination of longitude at sea, the Admiralty still required a number of scientific advisers, and Sabine, along with Michael Faraday and a physicist called Thomas Young, had been appointed. Sabine had also been a prominent player in the 1831 foundation of the British Association for the Advancement of Science.

  The establishment of magnetic observatories proceeded apace. The British East India Company quickly established another four—in Madras, Bombay, Singapore and the Himalayas—and by 1841 Göttingen Magnetische Verein could boast that a total of fifty stations had made observations during one or more of Gauss’s selected time intervals. Of these, thirty-five were in Europe, six in Asia, two in Africa, three in North America and four in and around Australia and New Zealand. Curiously, there were no observatories in Central or South America, where von Humboldt had begun his geomagnetic studies.

  Early on, Gauss had advocated recording only declination, considering the accuracy of inclination and intensity measurements to be too poor, but by the time Edward Sabine set up the British observatories all three elements were being routinely recorded. In his 1857 report to the Royal Society “On what the Colonial Magnetic Observatories have accomplished,” Sabine was able to claim that these measurements represented “not partially, but completely, the whole of the magnetic affections experienced at the surface of the globe.” The data would, he said, be useful in studying two aspects of geomagnetism: “the actual distribution of the magnetic influence over the globe, at the present epoch, in its mean or average state;” and “the history of all that is not permanent … momentary, daily, monthly or annual change and restoration; or in progressive changes … continually accumulating in one direction.”

  Sabine became immersed in his observatories and the data pouring out from them. In 1841 the Royal Society took over from Göttingen Magnetische Verein the responsibility for collating all the geomagnetic observatory data, and Sabine took on the job of analyzing it. After several years he eventually began to see a pattern in the frequency of magnetic storm activity. He noted that at Toronto, an observatory that consistently produced high-quality results, the disturbances decreased markedly between 1841 and 1843, but from 1843 to 1848 they increased in number again, by a factor of about three.

  At the same time, Samuel Heinrich Schwabe, a German pharmacist and amateur astronomer—he had an astronomical observatory on the roof of his house—had been watching the little dark patches that appeared on the face of the sun from time to time. He had, in fact, been observing these sunspots for a great many years while pursuing his real mission of searching for an inner planet, “Vulcan,” which he believed would show up as a spot passing across the face of the sun. Schwabe never found Vulcan of course, but he did notice that the number of sunspots followed an eleven-year cycle, with the greatest number occurring in 1828, 1837 and 1848, and the fewest in 1833 and 1842.

  No one had yet explained what sunspots were, but to Sabine the correlation with terrestrial magnetic storms was unmistakable: the sun and its cycle of spots seemed to be somehow responsible for von Humboldt’s magnetic storms. But how could this be—across 150 million kilometers of empty space?

  During the first half of the nineteenth century the silent voice of the magnetic needle had certainly yielded huge amounts of data. Voluminous reports and treatises had been written, and complex mathematical representations constructed. Yet by the end were scientists really any closer to understanding the internal workings of the Earth? Gauss was convinced that the seat of Earth’s magnetism was internal. But Sabine’s latest results reinforced those of Celsius and Hiorter, and once again seemed to point to an influence from the heavens. Von Humboldt had commented:

  The phenomena of Earth’s magnetism, in its three forms of variation, dip and intensity, have of late years been examined with great care, in the most different zones, by the united efforts of many travelers; and
there is scarcely any branch of the physical knowledge which, in so small a number of years, so much has been gained towards an acquaintance with its laws, though not perhaps with its causes.

  How right he was. The application of science to the study of the Earth’s interior was only just beginning. An answer to the puzzle of magnetism would necessitate solving many of the planet’s inner secrets and assembling the solutions, one by one, like the pieces of a jigsaw.

  The Core of the Matter

  Ancient civilizations could only speculate about the nature and shape of the world they lived on … Early in the twentieth century it became evident that the interior of the Earth has a … structure like that of an onion.

  —WILLIAM LOWRIE, 1997

  Since 1968, when the Apollo 8 astronauts beamed back the first breathtaking views of Earth, a shimmering globe of blue, green and white, no one can have seriously held on to the notion that our world is flat. Interestingly, though, evidence of the shape of the Earth had been there for all to see since the beginning of civilization, and had certainly been recognized by Aristotle around 350 BC and possibly by Pythagoras 200 years earlier. During a lunar eclipse, as the Earth passes directly between the sun and the moon, a circular arc of shadow is seen to creep across the bright face of the full moon until the lunar disc is completely masked and turns a dull red color. The circular shadow can mean only one thing—the Earth, like the sun and the moon, is spherical.

  A hundred years after Aristotle, Eratosthenes, the head librarian at the great library of Alexandria in Egypt, became the first person to estimate the size of the Earth. He employed a clever, sundial-like technique. He knew that at Syene, which lay on the tropic of Cancer near the modern city of Aswan, 5000 stadia due south of Alexandria, the sun was exactly overhead at noon on midsummer’s day. By measuring the length of a shadow cast at Alexandria at noon on midsummer’s day, Eratosthenes estimated that this distance of 5000 stadia represented one-fiftieth of the circumference of the Earth. If Eratosthenes was using the common Attic stadion, which equals about 185 meters, his result translates to a radius a little over 7000 kilometers. Some scholars think he may have been using the less common Egyptian stadion of 157.5 meters, which would give a radius of 6320 kilometers, or within one percent of the actual value. Either way, the result was remarkable for such a simple experiment.

  Later, thanks to Copernicus, Galileo, Kepler and Newton, came the realization that Earth was just one of a number of planets revolving in almost circular orbits around the sun, while at the same time rotating on their own axes. In Newton’s time only six planets were recognized—Mercury, Venus, Earth, Mars, Jupiter and Saturn—the same six “wandering stars” that had been known to ancient Greek astronomers. Later, following the development of the astronomical telescope, Uranus and Neptune would be discovered, and in 1931 the American astronomer Clyde Tombaugh would detect Pluto (whose status as a planet was rescinded in 2006).

  Compared with what could be seen looking out into space, even by the beginning of the twentieth century little was known of what lay within planet Earth. The ancient Greeks had held that the universe comprised four fundamental elements—earth, water, air and fire—all of which they could see in abundance on the surface of the planet. But although they apparently distinguished different materials on the basis of their physical properties—lodestone and amber, for example—they do not seem to have made much progress in investigating the Earth’s underlying constitution.

  This was hardly surprising: by 1600 William Gilbert was still commenting on the inaccessibility of Earth’s interior to direct sampling and experimentation. Even today, scientists have drilled a mere twelve kilometers into the outermost layers of the planet; the rest remains the realm of science-fiction writers and remote-sensing methods of investigation.

  By the beginning of the twentieth century, however, geophysicists had begun to realize that gravity, geomagnetism and earthquakes all yielded valuable clues to the make-up of the planet if only these clues could be deciphered and interpreted. In the eighteenth century, Edmond Halley had envisaged Earth’s interior as a series of shells that nested inside one another like Russian matryoshka dolls and were scaled in harmony with the heavens. Although this model never really caught on, the basic idea of a concentric layered structure, comprising an inner nucleus or core and a rigid outer shell or crust, would be the starting point for almost all later models. However, intense debate was to develop about the composition and physical state of the various layers.

  The first indication that Earth might not be a homogeneous lump of rock had appeared in 1798, when Henry Cavendish, the reclusive professor of physics at Cambridge University, announced that “the density of the Earth comes out 5.48 times greater than that of water.” This meant that, on average, one cubic meter of Earth had a mass of 5480 kilograms. (Cavendish had actually made an arithmetical mistake: his figure should have been 5450.) By the same calculation, the mass of the whole Earth was about six million million million million (6 x 1024) kilograms.

  Cavendish’s result—that Earth’s density was five and a half times that of water—came as a huge surprise since the densities of rocks found at the Earth’s surface averaged only about two and a half times that of water, and were rarely more than three times the density.

  Had Cavendish made a mistake? It seemed unlikely. Thirty years earlier the Astronomer Royal, Nevil Maskelyne, had also tried to estimate Earth’s density using a quite different method. He had measured the deflection of a plumb line towards a cone-shaped Scottish mountain, Schiehallion, and obtained a density value four and a half times that of water. Although Maskelyne’s and Cavendish’s values were different, both were high enough to indicate there was more to the Earth than just its surface rocks.

  Although often dubbed his attempt to “weigh the world,” Cavendish’s experiment was actually designed to test Newton’s law of gravitation, then more than a century old, and to determine the still unknown universal gravitational constant, G. To do this he had set out to compare the force of gravity between two massive lead spheres and the gravitational force that the Earth exerted on each of them—in other words, their weight.

  He had inherited the design for this experiment from Reverend John Michell, a minister in the Church of England and amateur scientist, who had died before being able to complete the work himself. It comprised a huge torsion balance similar to those that Cavendish and Coulomb had used to investigate the electrostatic and magnetic forces, but much bigger: the main beam was nearly two meters long. At each end of the beam was fixed a two-inch diameter, 1.6-pound lead sphere, and about nine inches from each of these Cavendish suspended a twelve-inch, 348-pound sphere. He then meticulously measured the tiny forces acting between each pair of spheres. To minimize extraneous disturbances, the whole apparatus was sealed in a room of Cavendish’s house, with Cavendish operating from outside and observing through a tiny window.

  Henry Cavendish, the Cambridge University professor of physics whose experiment to “weigh the world” led to the conclusion that the Earth has a heavy, iron-rich core.

  Before long the Cavendish experiment had become the standard way to measure gravitational forces in the laboratory, and numerous repetitions revealed no error in Cavendish’s original result. It was clear there was some extremely dense material within the Earth.

  Around this time scientists began to question the physical and chemical properties of the rocks found on the Earth’s surface, and the processes by which these rocks had been formed, deformed and reformed through time. This would be the starting point of a new scientific discipline: geology. To explain the fluid magma spewed out from volcanoes, earthquakes and other geothermal activity, some geologists argued that Earth must have a molten interior, and only a thin solid crust. This was consistent with the observation— for example, in deep mine shafts—that temperature increased with depth, leading to the conclusion that the melting points of common rock-forming materials might be reached just a few tens of kilometers dow
n.

  Many physicists opposed this theory. Ampère, for example, argued that during the course of its orbit around the Earth the moon would raise tides in the magma, as it did in the waters of the oceans, and that these tides would tear apart such a thin crust. Poisson disputed whether such a crust could form over a predominantly molten interior in the first place. He believed that very early in Earth’s history, when the whole planet was still molten, cooler blobs of fluid would have sunk because of their higher density, so any solidification would have taken place from the inside out, rather than the outside in.

  Two schools of thought emerged. The first favored a largely molten Earth with a thin rigid crust. The second supported a model more like Halley’s, in which Earth had a substantial, solid nucleus and a relatively thin fluid layer underneath a rigid crust.

  One of the first scientists to address the problem objectively was William Hopkins, a Cambridge University mathematician. Academically, Hopkins was something of a late starter. Having tried his hand at farming and not enjoyed it, he was already twenty-nine and married for the second time when in 1821 he entered Peterhouse at Cambridge to study mathematics. Age proved no barrier: he excelled and graduated seventh “Wrangler”—seventh of the first-class honors students in the year’s Mathematics Tripos.

  Being married, Hopkins now found himself ineligible for a college fellowship. However, he gained a lectureship in mathematics and also became the university’s Esquire Bedell, a largely ceremonial position, the responsibilities of which included carrying the university mace when accompanying the vice-chancellor on official occasions, and carving the roasted joint for him at banquets. Hopkins tutored many of Cambridge’s top mathematics scholars, including William Thomson (later Lord Kelvin), Gabriel Stokes and James Clerk Maxwell, and soon became known as “the wrangler-maker.”