Krakatoa: The Day the World Exploded

by Simon Winchester

  The north–south trend of the magnetic stripes suggested that the rocks below were moving outwards, were on each side flowing away from this central point, just like rainwater hitting the ridge-pole of a roof, with some water then going down one side of the roof, some the other.

  And the conclusion to be drawn from this turned out to be the single missing part of the mechanism for continental drift, the one that Alfred Wegener had never been able to imagine. The mid-oceanic north–south axis was a place where – presumably, logically, but suddenly somehow astonishingly – entire tracts of brand new seabed were being created. The mid-ocean axis was a ridge of making, where the world was welling up from inside itself and spilling out on to the floor, to be carried out and away, thus making ever more room for the making of more.

  And as this brand spanking new submarine real estate was being slowly and steadily made over millions of years, and as it spread off away from the deep-sea ridge-pole both to the east and to the west, so the remanent magnetism of its rocks, with the record of the earth's polarity reversing itself every few tens of thousands of years, remained locked there on the sea-floor for all to see. By seeing this, by understanding what it was, scientists were able, at long last, to claim they had evidence of sea-floor spreading that was incontrovertible. Once radiochemical dating techniques had been perfected, then the timing of the spreading, of the making of new seabed, as well as the dating of the continental drift could all be fully established too.

  Now that there was a mechanism and a date, the rest suddenly became almost too obvious. The conclusions began rushing in, filling the intellectual void that had so plagued the science for all the decades since Wegener had died forty years before.

  In 1962, armed with the new evidence from the Pioneer and all the other armadas that dragged schools offish-shaped magnetometers behind them,* Harry Hess decided he would revisit the nagging question that the war had so rudely interrupted. There was now solid new evidence to underpin his notion that convection currents were at work under the earth's crust, and that the continents were indeed being moved about upon it, like gigantic rafts, colliding and bouncing and plunging down back into the earth's molten heart, in a ceaseless supra-terrestrial dance. ‘One may quibble over the details,’ he wrote in a paper that year, which is now a classic,* ‘but the general picture on palaeomagnetism is sufficiently compelling that it is more reasonable to accept than disregard it.’ Continental drift was most assuredly taking place. There could be no doubt – and, indeed, since the publication of Harry Hess's paper, there never has been.

  By curious chance, I came to know, if only slightly, both Harry Hess and Keith Runcorn. The encounter with Hess was to provide the more embarrassing memory, at which I blush to this day.

  It was the early spring of 1966, and I was approaching twenty-two. I had lately – for reasons of Buggins's turn, rather than any innate talent – been elected president of the Oxford University Geological Society. In that capacity, and because I had often visited America – and Princeton University as well, as it happened – I had managed to persuade Professor Hess to speak to the final, celebratory Society meeting of that Hilary Term. Hess was by then a famous figure, and his acceptance of an invitation written by a mere undergraduate was a signal honour for all. Senior academics from most of Oxford's science departments were eagerly gathering to hear the great man speak.

  It is a tradition of the Society that the event's protagonists – including the guest – wear black tie. Accompanied by my lieutenants, I met Hess (he was clothed as requested, though he called it his ‘tux’ and it looked suitably moth-eaten and academic) at Oxford railway station. We had decided to give him (and ourselves – we were spending official OUGS funds) a good feed at a well-known riverside inn out in the country, perhaps ten miles from town. So the five of us – Society secretary, treasurer, vice-president, Hess and me – drove off in my 1935 Morris 8, a venerable car much prized by my friends but, as it turned out, of lamentably timed unreliability.

  We dined early and very well, beside an immense inglenook-fire, in a building composed of ancient thatch and mid-Jurassic corbels, on soup and lamb and a '59 Aloxe-Corton. Our date with the assembled worthies of the university was at eight thirty, in the Museum. We left at ten minutes to eight – ample time to get to the venue. Harry Hess had enjoyed the Corton, and we had opened, as I recall, three bottles. He was a happy man.

  The car, however, was less so. It broke down in some flat and nameless swamp, five miles equidistant from both inn and Oxford. It was foggy, cold and dark. We had no telephone, and no way of finding one. Our only option was to walk, taking a path that our treasurer claimed to recognize and that he laughably described as a short-cut. Well-shone black patent shoes are less than suitable for the Oxfordshire mud. We found a pub, made a telephone call – but the person who answered hadn't heard about any meeting or any museum. Hess had a couple of whiskies.

  We arrived back in Oxford at ten o'clock – muddy, wet, cold and, in the case of Harry Hess, agreeably and pleasurably drunk. Our audience had remained, not knowing what else to do. The speech was a disaster. Someone tripped over, and much of what was said was incoherent. Maps fell down. The projector fused. The front row glared at us, deeming us responsible for the supposed slur on this monarch of high academia.

  I was not asked to convene any further meetings of the Geological Society; but Harry Hess wrote later to say he could not recall a more amusing or satisfying evening in recent years, and hoped that we would remain in contact – which we did, until he died three years later.

  I had known Keith Runcorn well during my time as a reporter in Newcastle-upon-Tyne, where he was professor of geophysics. He had taken me under his wing, seeing me as a local reporter interested in science who might help him publicize his research on deep-ocean tides. He had an arrangement with Cable & Wireless to make use of one of their disused trans-Pacific telegraph cables, and had set up monitoring equipment on Fanning Island, in what is now the Republic of Kiribati, to measure the tiny electrical impulses generated in the cable by the deep-water tidal movements in mid ocean. For readers in the grim and grisly winters of Tyneside, newspaper stories about the blue waters and eternal sunny skies of a South Pacific atoll made a welcome change; I wrote about him often, liked him enormously, and only wished that my paper had the budget to let me go out, as he had often suggested, to spend a season on Fanning, measuring the currents, soaking up the Polynesian Way.

  Once I left the north-east of England, Keith Runcorn and I lost touch – one might rightly say that we drifted apart. From time to time I would see his name on various celebratory papers, or on programmes for conferences called to note this anniversary or that, and all to do with the theories of continental drift of which, along with Hess and a small army of others, he was now world-renowned as an architect.

  And then, in December 1995, I read that he had been found, brutally murdered, in a hotel room in San Diego. He had been on his way to the annual meeting, in San Francisco, of the American Geophysical Union. He was to give a paper about his new interest, the magnetic fields that were then being measured on the surface of the moon.

  4. Definition

  It turned out to be but a short step from a wholesale acceptance of the mechanics of continental drift to the creation of the new theory that would be christened plate tectonics – that set of abiding planetary principles that today is generally acknowledged to be the first properly global theory to have been adduced and accepted in the history of earth science.

  The crucial first mention of the theory came on 24 July 1965, in the British journal Nature, under the authorship of the genial Canadian who is now most generally accepted as the ‘father’ of plate tectonics. Many hundreds of scientists had already spent thousands of man-years working away with their gravimeters and magnetometers, their polarizing microscopes and their hammers, to create a vague and pointillist image of the emerging theory. But it was not fully recognized until that summer's morning in 1965, when the t
hick, red-covered and always in those days infuriatingly tightly rolled magazine thudded on to desks in geological laboratories across the world. And then, though scores had been involved, no scores seemed to need settling: there was no contest for this particular moment of glory. It was the bluff, genial, approachable and down-to-earth University of Toronto professor J. Tuzo Wilson who, by opening his four-page essay with the following declarative paragraph, in essence created the new science:

  Many geologists have maintained that movements of the Earth's crust are concentrated in mobile belts, which may take the form of mountains, mid-ocean ridges or major faults… This article suggests that these features are not isolated, that few come to dead ends, but that they are connected into a continuous network of mobile belts about the Earth which divide the surface into several large rigid plates.

  His realization was essentially two-fold: the one discovery announced in this seminal 1965 Nature paper; the other, inextricably linked to the first and made two years beforehand, when the reacceptance of continental drift was well under way, but in need of as much hard evidence in support as could be found.

  The first discovery came when J. Tuzo Wilson looked at the Hawaiian Islands. After thinking about what he saw, he came up with a piece of evidence that added mightily to all the news about tiger-striped remanent magnetics and basalt conveyor-belts on the bed of the ocean. In the tradition of good science – close observation followed by prescient deduction, with in this case the addition of the basilisk glare of his own geophysical insight – he saw something, something that the ancient Hawaiians themselves had suspected for centuries.

  The Hawaiian Island chain stretches for more than 2,500 miles, across thirty degrees of Pacific longitude. Most of the islands one would not think of as Hawaiian. Nihoa, Necker, Tern, Disappearing, Laysan, Lisianski, Kittery and Seal Islands, for example, are almost unknown: only Midway and Ocean Islands, at the north-western end of the chain, are household names, and those mainly because of wars and naval battles. What is currently (but technically wrongly) thought of as the Hawaiian Islands is the 400-mile line of just nine bodies of rock and palm reaching from the outermost pinnacle of Kaula, via the north-westernmost (and still privately owned) island of Niihau, to that great chunk of basalt at the south-east called Hawaii, which is known by most non-Polynesian visitors as the Big Island.

  Hawaiian legend has long acknowledged what casual visitors may notice too: that the dusty, half-dead island of Niihau looks much wearier and older than the feisty, bubbling and fiery island of Hawaii. The dank black Waialeale swamp at the summit of Kauai – the wettest place on earth, the locals say – looks prehistoric; the fresh crags of Diamond Head on Oahu (the ‘diamonds’ being glittery and new-looking olivine crystals) look young.

  To the ancient Polynesians, who closely examined the islands' soil erosion and vegetation, the difference in age was also self-evident. They incorporated the difference in age into their stories. Pele, the goddess of volcanoes, once lived in Kauai. But then she was attacked by her older sister Namakaokahai, the goddess of the sea. And so she fled, south-east, to Oahu. Namakaokahai attacked again, and Pele moved south-east once more, to Maui. And then a third time, whereupon Pele moved yet again, this time to the Halemaumau Crater of Kilauea, on the summit of Hawaii itself. She had moved 300 miles south-eastwards, hopping from island to island, as one volcano after another exploded and died behind her.

  Like many legends, this old yarn has its basis in fact. The sea attacks volcanoes – the waters and the waves erode the fresh-laid rocks. And this is why Pele herself moved, shifting always to the younger and newer volcanoes, and relentlessly away from the older and worn-out islands of the north-west, and down towards their more recently created and unspoiled cousins in the south-east.

  It was in 1963, in the cold of faraway Toronto, that J. Tuzo Wilson first looked at these stories, examined the geological maps, peered at the records of the various eruptions and lava flows on the Hawaiian main chain, and came up with an idea. Volcanoes had existed in this area of the world for what seemed a very long period of time. So it seemed likely that, for some reason, there was at this particular point in the earth's mantle a deep and stationary hot spot (geologists have since elided the phrase to form today's hotspot).

  The heat from this fiercely hot zone partly melted the rocks of the upper mantle lying above it. The magma created by this melting was lighter than the surrounding solid rock; it burst through the remainder of the mantle into and through the crust; and then it erupted through the ocean floor to produce either a sea-mount that would remain always below the surface, or an undersea volcano that would grow until it breached the surface of the sea and emerged as an island.

  That part seemed simple and logical. But what Wilson then deduced, both from the Polynesian legends and from his own observations, was that while the hotspot remained static, the upper mantle and crust above had moved, carrying one island at a time away from the hotspot and abruptly stopping the volcanism there. After a while, magma erupted again at a new site that lay in the wake, as it were, of the first moving island, and another fresh island was made altogether. It was rather like moving a baking tray full of batter slowly away from a single intense gas burner, creating a line of ripples and partially cooked batter – the batter cooked at one end of the tray, uncooked at the other, with all the stages in between.

  Here in the Pacific, on a grand scale, was a 400-mile-long trail of volcanoes – in various stages of cooking. The older rocks were in this case at the north-western end of the trail, the newer rocks at the south-eastern. If the radiometric dating confirmed this, then it could be incontrovertibly proven that the mantle and crust along the Hawaiian chain was itself moving north-westwards over a hotspot that was now and long had been sited directly under where the Big Island is today.

  And the dating of the rocks proved it, precisely. The Kauai basalts were shown to be 5.5 million years old; those on Hawaii itself were on average less than 0.7 million years old; and in the craters where Pele is said now to be lurking, they are still being created, millions of tons of hotspot-induced basaltic melt baked afresh every day.

  The dating clinched the idea. Wilson was shown to be exactly right. Science could now add to the evidence of the tiger-stripe magnetic anomalies off Seattle this second indication of the undersea geological conveyor-belt. The magnetic evidence had been found below the surface; this from Hawaii was by contrast all on the surface. Together they presented science with a sort of gigantic rock-hewn tape-recorder that showed not only the mid-Pacific movement that had been taking place over the last five million years but also the expansion of the sea-floor, with the continents on each side of it, undeniably and indisputably, drifting.

  The second realization – a true Eureka! moment of epiphany – came after Wilson starting experimenting not with rocks and magnetic traces or radioactive rubidium, but with paper and a set of scissors. The title of his famous Nature paper, ‘A New Class of Faults and Their Bearing on Continental Drift’, suggests why he was indulging himself so.

  John Dewey, who went on from Cambridge to become professor of geology at Oxford, remembers the moment. It was

  ... one early autumn morning in 1964, I was sitting in my room in the Sedgwick Museum in Cambridge… when Toronto's Tuzo Wilson, on sabbatical leave, sauntered in clearly bursting to tell anyone who would listen about his new ideas. He had discovered that I was the new lecturer in structural geology and said: ‘Dewey, I have discovered a new class of fault.’ ‘Rubbish,’ I said, ‘we know about the geology and kinematics of every kind of fault known to mankind.’ Tuzo grinned, and produced a simple coloured folded paper version of his now famous ridge/transform/ridge model, and proceeded to open and close, open and close it with that wonderful smile on his face. I was transfixed both by the realization that I was seeing something profoundly new and important, and by the fact that I was talking to a very clever and original man.

  Wilson's achievement was to show what would happen
if a geological process caused a fault – in which one body of rock breaks and slips alongside another – across a ridge that, for some reason, happened to be spreading open. He was interested because new bathymetric survey data – the soundings of the sea – was showing, to everyone's astonishment, that the great ridges that ran down the centres of both the Atlantic and the Pacific Oceans were incised by dozens of deep gashes. Geologists initially thought these gashes were spectacular versions of nothing more than classic faults, tearing the ridge asunder as they sheared

  J. Tuzo Wilson's famous transform fault structure, which finally and definitively demonstrated exactly what happens when sea-floor spreading occurs at a mid-ocean ridge.

  to the left or to the right as random happenstance directed.

  But Tuzo Wilson looked at the bathymetric pictures of these gashes again and said to himself no! – if the ridge in the centre happened itself to be opening up, for some wholly unconnected reason, then the directions of the fault slippage would be exactly the opposite of those that might happen across a line that was totally stable. He gave a name to the new phenomenon: a normal left-to-right or right-to-left fault was known as a transcurrent fault, but he named the fault he was now predicting, which had never been seen or described before (because no one had ever imagined a ridge spreading itself open), a transform fault.