Annals of the Former World

by John McPhee

  It was on the mechanics of the seafloor that geology’s revolutionary inquiries were primarily focussed in the early days. Harry Hess, a mineralogist who taught at Princeton, was the skipper of an attack transport during the Second World War, and he carried troops to landings—against the furious defenses of Iwo Jima, for example, and through rockets off the beaches of Lingayen Gulf. Loud noises above the surface scarcely distracted him. He had brought along a new kind of instrument called a Fathometer, and, battle or no battle, he never turned it off. Its stylus was drawing pictures of the floor of the sea. Among the many things he discerned there were dead volcanoes, spread out around the Pacific bottom like Hershey’s Kisses on a tray. They had the arresting feature that their tops had been cut off, evidently the work of waves. Most of them were covered with thousands of feet of water. He did not know what to make of them. He named them guyots, for a nineteenth-century geologist at Princeton, and sailed on.

  The Second World War was a technological piñata, and, with their new Fathometers and proton-precession magnetometers, oceanographers of the nineteen-fifties—most notably Bruce Heezen and Marie Tharp at Columbia University—mapped the seafloor in such extraordinary detail that in a sense they were seeing it for the first time. (Today, the very best maps are classified, because they reveal the places where submarines hide.) What stood out even more prominently than the deep trenches were mountain ranges that rose some six thousand feet above the general seafloor and ran like seams through every ocean and all around the globe. They became known as rises, or ridges—the Mid-Atlantic Ridge, the Southeast Indian Ocean Ridge, the East Pacific Rise. They fell away gently from their central ridgelines, and the slopes extended outward hundreds of miles, to the edges of abyssal plains—the Hatteras Abyssal Plain, the Demerara Abyssal Plain, the Tasman Abyssal Plain. Right down the spines of most of the submarine cordilleras ran high axial valleys, grooves that marked the summit line. These eventually came to be regarded as rift valleys, for they proved to be the boundaries between separating plates. As early as 1956, oceanographers at Columbia had assembled seismological data suggesting that a remarkable percentage of all earthquakes were occurring in the mid-ocean rifts—a finding that was supported, and then some, after a worldwide system of more than a hundred seismological monitoring stations was established in anticipation of the nuclear-test-ban treaty of 1963. If there was to be underground testing, one had to be able to detect someone else’s tests, so a by-product of the Cold War was seismological data on a scale unapproached before. The whole of plate tectonics, a story of steady-state violence along boundaries, was being brought to light largely as a result of the development of instruments of war. Earthquakes “focus” where earth begins to move, and along transform faults like the San Andreas the focusses were shallow. At the ocean trenches they could be very deep. The facts accrued. Global maps of the new seismological data showed earthquakes not only clustered all along the ridges of the seafloor mountains but also in the trenches and transform faults, with the result that the seismology was sketching the earth’s crustal plates.

  To Rear Admiral Hess, as he had become in the U.S. Naval Reserve, it now seemed apparent that seafloors were spreading away from mid-ocean ridges, where new seafloor was continuously being created in deep cracks, and, thinking through as many related phenomena as he was able to discern at the time, he marshalled his own research and the published work of others up to 1960 and wrote in that year his “History of Ocean Basins.” In the nineteen-forties, a professor at Delft had written a book called The Pulse of the Earth, in which he asserted with mild cynicism that where gaps exist among the facts of geology the space between is often filled with things “geopoetical,” and now Hess, with good-humored candor, adopted the term and announced in his first paragraph that while he meant “not to travel any further into the realm of fantasy than is absolutely necessary,” he nonetheless looked upon what he was about to present as “an essay in geopoetry.” He could not be sure which of his suppositions might be empty conjecture and which might in retrospect be regarded as precocious insights. His criterion could only have been that they seemed compelling to him. His guyots, he had by now decided, were volcanoes that grew at spreading centers, where they protruded above the ocean surface and were attacked by waves. With the moving ocean floor they travelled slowly down to the abyssal plains and went on eventually to “ride down into the jaw crusher” of the deep trenches, where they were consumed. “The earth is a dynamic body with its surface constantly changing,” he wrote, and he agreed with others that the force driving it all must be heat from deep in the mantle, moving in huge revolving cells (an idea that had been around in one form or another since 1839 and is still the prevailing guess in answer to the unresolved question: What is the engine of plate tectonics?). Hess reasoned also that the heat involved in the making of new seafloor is what keeps the ocean rises high, and that moving outward the new material gradually cools and subsides. The rises seemed to be impermanent features, the seafloor altogether “ephemeral.” “The whole ocean is virtually swept clean (replaced by new mantle material) every three hundred to four hundred million years,” he wrote, not then suspecting that ocean crust is actually consumed in half that time. “This accounts for the relatively thin veneer of sediments on the ocean floor, the relatively small number of volcanic seamounts, and the present absence of evidence of rocks older than Cretaceous in the oceans.” In ending, he said, “The writer has attempted to invent an evolution for ocean basins. It is hardly likely that all of the numerous assumptions made are correct. Nevertheless it appears to be a useful framework for testing various and sundry groups of hypotheses relating to the oceans. It is hoped that the framework with necessary patching and repair may eventually form the basis for a new and sounder structure.”

  In 1963, Drummond Matthews and Fred Vine, of Cambridge University, published an extraordinary piece of science that gave to Hess’s structure much added strength. Magnetometers dragged back and forth across the seas had recorded magnetism of two quite different intensities. Plotted on a map, these magnetic differences ran in stripes that were parallel to the mid-ocean ridges. The magnetism over the centers of the ridges themselves was uniformly strong. Moving away from the ridges, the strong and weak stripes varied in width from a few kilometres to as many as eighty. Vine and Matthews, chatting over tea in Cambridge, thought of using this data to connect Harry Hess’s spreading seafloor to the time scale of paleomagnetic reversals. The match would turn out to be exact. The weaker stripes matched times when the earth’s magnetic field had been reversed, and the strong ones matched times when the magnetic pole was in the north. Moreover, the two sets of stripes—calendars, in effect, moving away from the ridge—seemed to be symmetrical. The seafloor was not only spreading. It was documenting its age. L. W. Morley, a Canadian, independently had reached the same conclusions. Vine and Matthews’ paper was published in Nature in September, 1963, and became salient in the development of plate tectonics. In January of the same year, Morley had submitted almost identical ideas to the editors of Nature, but they were not yet prepared to accept them, so Morley then submitted the paper in the United States to the Journal of Geophysical Research, which rejected it summarily. Morley’s paper came back with a note telling him that his ideas were suitable for a cocktail party but not for a serious publication.

  Data confirming the Vine-Matthews hypothesis began to accumulate, nowhere more emphatically than in a magnetic profile of the seafloor made by the National Science Foundation’s ship Eltanin crossing the East Pacific Rise. The Eltanin’s data showed that the seafloor became older and older with distance from the spreading center, and with perfect symmetry for two thousand kilometres on either side. All through the nineteen-sixties, ships continued to cruise the oceans dragging magnetometers behind, and eventually computers were programmed to correlate the benthic data with the surface wanderings of the ships. Potassium-argon dating had timed the earth’s magnetic reversals to apparent perfection for the last thre
e and a half million years. Geologists at Columbia calculated the rate of seafloor spreading for those years and then assumed the rate to have been constant through earlier time. On that assumption, they extrapolated a much more extensive paleomagnetic time scale. (Improved radiometric dating later endorsed the accuracy of the method.) And with that scale they swiftly mapped the history of ocean basins. Compared with a geologic map of a continent, it was a picture handsome and spare. As the paleomagnetist Allan Cox, of Stanford University, would describe it in a book called Plate Tectonics and Geomagnetic Reversals, “The structure of the seafloor is as simple as a set of tree rings, and like a modern bank check it carries an easily decipherable magnetic signature.”

  Meanwhile, geophysicists at Toronto, Columbia, Princeton, and the Scripps Institution of Oceanography were filling in the last major components of the plate-tectonics paradigm. They figured out the geometry of moving segments on a sphere, showed that deformation happens only at the margins of plates, charted the relative motions of the plates, and mapped for the first time the plate boundaries of the world.

  If it was altogether true, as Hess had claimed, that with relative frequency “the whole ocean is virtually swept clean,” then old rock should be absent from deep ocean floors. Since 1968, the drill ships Glomar Challenger and JOIDES Resolution have successively travelled the world looking for, among other things, the oldest ocean rocks. The oldest ever found is Jurassic. In a world that is 4.56 billion years old, with continental-shield rock that has been dated to 3.96, it is indeed astonishing that the oldest rock that human beings have ever removed from a seafloor has an age of a hundred and eighty-five million years—that the earth is twenty-five times as old as the oldest rock of the oceans. In 1969, it seemed likely that the oldest ocean floor would be found in the Northwest Pacific. The Glomar Challenger went there to see. Two Russians were aboard who believed that rock older than Jurassic—rock of the Paleozoic, in all likelihood—would be discovered. They took vodka with them to toast the first trilobite to appear on deck. Trilobites, index fossils of the Paleozoic, came into the world at the base of the Cambrian and went out forever in the Permian Extinction—sixty-five million years before the age of the oldest rock ever found in modern oceans. As expected, the oceanic basement became older and older as the ship drilled westward from Hawaii. But even at the edge of the Marianas Trench, the Russians were disappointed. No vodka. Ah, but there might be older rock on the other side of the trench, in the floor of the Philippine Sea. The ship pulled up its drill pipe and moved across the trench. This time the rock was Miocene, more or less a tenth as old as the Jurassic floor. The Russians broke out the vodka. A toast! Neil Armstrong and Edwin Aldrin were walking around on the moon.

  “In the old days we would have called this North America,” Deffeyes said, sinking another clear tube into the ground. “We now think of plates. The plate-tectonics revolution came as a surprise, with very little buildup to it. There was none of that cloud that precedes a political revolution. In the nineteen-fifties, when I was a graduate student, nearly all the faculty at Princeton thought continental drift was sheer baloney. A couple of years later, Harry Hess broke it open. I had thought I would go through my career without anything like it. Oil and mining seemed enough of a contribution to keep one going. But now something had come along that was so profound that it took the whole science with it. We used to think that continents grew like onions around old rock. That was overturned by plate tectonics. And we could see now how amazingly fast you could put up a mountain range. A continent-to-continent collision was a hell of an episode at a limited place. After the Appalachians and the Urals were recognized as continent-to-continent sutures, people said, ‘O.K., where’s the suture in California?’ Geologists kept jumping up and saying, ‘I’ve got the suture! I’ve got the suture!’ It turned out, of course, that there were at least three sutures. In each instance, a great island had closed up a sea and hit into America—just as India hit Tibet, just as Kodiak Island, which is a mini-India, is about to plow into Alaska. Fossils from the mid-Pacific have been found here in the West, and limestones that lithified a thousand miles south of the equator. Formations in California have alien fossils with cousins in the rock of New Guinea. For a while, people were going around naming a defunct ocean for every suture. The first piece, coming in from the west, was the one that rode up onto North America about forty miles, not a trivial distance, in Mississippian time. That was the action that first tipped the rock in the Carlin unconformity. The old name for it was the Antler Orogeny. In the early Triassic, the second one arrived—the Golconda Thrust—and rode fifty miles over the trailing edge of the first one, and in the Jurassic the third one came in, sutured on somewhere near Sacramento, and more or less completed California. I have read that two geologists have found in Siberia a displaced terrane that was taken off of North America. The Lord giveth and the Lord taketh away.”

  I mentioned that I had read in Geology that one out of eight geologists does not accept plate-tectonic theory.

  He said, “There are still a few people dragging their feet. They don’t want to come into the story.”

  I asked him if he thought the Uinta Mountains could be explained in terms of plate theory. The Uintas are a range in the Rockies, seven hundred miles from the sea, and they run east-west, unlike virtually all other ranges for thousands of miles around them. If the Western cordilleras were raised by colliding plates, how did the Uintas happen to come up at right angles to the other mountains?

  He said, “You must have been talking to a Rocky Mountain geologist.” He said nothing else for a time, while he tapped at the earth I had uncovered and captured a perfect sample. Then he said, “The north side of the Uintas is a spectacular mountain wall. Glorious. You come upon it and suddenly you see structurally the boundary of the range. But you don’t see what put it there. The Uintas are mysterious. They are not a basin-range fault block, yet they have come up nearly vertically, with almost no compression evident. You just stand there and watch them go up into the sky. They don’t fit our idea of plate tectonics. The Rockies in general will be one of the last places in the world to be deciphered in terms of how many hits created them, and just when, and from where.”

  The article in Geology was based on a questionnaire that was circulated toward the end of the nineteen-seventies. The results indicated that forty per cent of geologists had come to feel that plate theory was “essentially established,” while a roughly equal number preferred to qualify a bit and say that it was “fairly well established.” Eleven per cent felt that the theory was “inadequately proven.” Seven per cent said they had accepted continental drift before 1940. Six per cent thought plate tectonics would be “still in doubt” in the late nineteen-eighties. And one geologist predicted that the theory would eventually be rejected.

  “At any given moment, no two geologists are going to have in their heads exactly the same levels of acceptance of all hypotheses and theories that are floating around,” Deffeyes said. “There are always many ideas in various stages of acceptance. That is how science works. Ideas range from the solidly accepted to the literally half-baked—those in the process of forming, the sorts of things about which people call each other up in the middle of the night. All science involves speculation, and few sciences include as much speculation as geology. Is the Delaware Water Gap the outlet of a huge lake all other traces of which have since disappeared? A geomorphologist will tell you that, in principle, the idea is O.K. You have to deal with partial information. In oil drilling, you had better be ready to act shrewdly on partial information. Do physicists do that? Hell, no. They want to have it to seven decimal places on their Hewlett-Packards. The geologist has to choose the course of action with the best statistical chance. As a result, the style of geology is full of inferences, and they change. No one has ever seen a geosyncline. No one has ever seen the welding of tuff. No one has ever seen a granite batholith intrude.”

  Since I was digging his sample pits, I felt enfranchi
sed to remark on what I took to be the literary timbre of his science.

  “There’s an essential difference,” he said. “The authors of literary works may not have intended all the subtleties, complexities, undertones, and overtones that are attributed to them by critics and by students writing doctoral theses.”

  “That is what God says about geologists,” I told him, chipping into the sediment with his broken shovel.

  “You may recall Archelaus’s explanation of earthquakes,” he said cryptically. “Earthquakes were caused by air trapped in underground caves. It shook the earth in its effort to escape. Everyone knew then that the earth was flatulent.”

  Deffeyes said he had asked his friend Jason Morgan—whose paper “Rises, Trenches, Great Faults, and Crustal Blocks” defined the boundaries of the plates—what he was going to do for an encore. Morgan said he didn’t know, but possibly the most exciting thing to do next would be to prove the theory wrong.

  That would be a reversal comparable to the debunking of Genesis. I remembered Eldridge Moores, of the University of California at Davis, telling me what it had been like to be in graduate school at the height of the plate-tectonics revolution, and how he had imagined that the fervor and causal excitement of it was something like landing on Guadalcanal in the middle of the action of “a noble war.” Tanya Atwater, a marine geologist who eventually joined the faculty of the Massachusetts Institute of Technology, was then a graduate student at the Scripps Institution of Oceanography. In a letter written to Allan Cox at Stanford, she re-creates the milieu of the time. “Seafloor spreading was a wonderful concept because it could explain so much of what we knew, but plate tectonics really set us free and flying. It gave us some firm rules so that we could predict what we should find in unknown places … . From the moment the plate concept was introduced, the geometry of the San Andreas system was an obviously interesting example. The night Dan McKenzie and Bob Parker told me the idea, a bunch of us were drinking beer at the Little Bavaria in La Jolla. Dan sketched it on a napkin. ‘Aha!’ said I, ‘but what about the Mendocino trend?’ ‘Easy!,’ and he showed me three plates. As simple as that! The simplicity and power of the geometry of those three plates captured my mind that night and has never let go since. It is a wondrous thing to have the random facts in one’s head suddenly fall into the slots of an orderly framework. It is like an explosion inside. That is what happened to me that night and that is what I often felt happen to me and to others as I was working out (and talking out) the geometry of the western U.S … . The best part of the plate business is that it has made us all start communicating. People who squeeze rocks and people who identify deep-ocean nannofossils and people who map faults in Montana suddenly all care about each other’s work. I think I spend half my time just talking and listening to people from many fields, searching together for how it might all fit together. And when something does fall into place, there is that mental explosion and the wondrous excitement. I think the human brain must love order.”