St Thomas, Danish West Indies, March 16, 1873, 18° 22′ N, 64° 56′ W to Hamilton, Bermuda, 32° 18′ N, 64 ° 48′ W Atlantic Ocean

The Bermuda Triangle is, of course, famous for the unexplained ship and plane disappearances that have occurred there. Even today you will not find it on any map. Broadly speaking, the Bermuda Triangle is defined by three points: the island of Bermuda, the city of Miami, and the island of Puerto Rico. In March of 1873 Challenger was traveling north along the right-hand side of that triangle.

It is difficult to determine the truth or otherwise of the Bermuda Triangle myth. It’s tempting to regard the problem as simply statistical. In the days when radio and radar communications were primitive or nonexistent, planes and ships were lost at sea all the time; it was one of the perils of the job. But the enduring myth of the Bermuda Triangle as a region where this happens rather too often for comfort will not go away. Many explanations, which range from the feasible to the plain daft, have been advanced. An example of the former is that the Bermuda Triangle lies on the 80° meridian, a line that is one of two places on Earth where the magnetic and true north poles are in perfect alignment. Navigators on these so-called agonic lines do not need to make the usual correction to

their compasses. It is just possible that even experienced navigators might be thrown if they failed to re-adjust their compasses when on an agonic line. The other agonic line, in the Pacific Ocean just to the east of Japan, is also known to local mariners as a place of unexplained disappearances.

The Bermuda Triangle is also known as a region where unpredictable weather can develop rapidly. It is, after all, in the region where the Gulf Stream originates and that vast mass of water moving northeast at 4 knots carries a hefty kinetic punch. Small “meso” thunderstorms of ferocious intensity can develop and vanish quite suddenly in the Caribbean.

At the daft end of the scale of Bermuda Triangle explanations is the notion that the area is frequented by roving extraterrestrials who abduct humans and their machines. Indeed Steven Spielberg’s epic, Close Encounters of the Third Kind, opens with the famous loss of Flight 19, when five Grumman torpedo bombers disappeared without trace in 1945, and adds the fantasy of them turning up in the Mexican desert 30 years later.

Yet there is another explanation for the mystery of the Bermuda Triangle. This is the conjecture that the ocean—and indeed the air above it—can suddenly lose the ability to support objects such as ships and aircraft because the density of water and air are suddenly reduced. This idea proposes that the density change is caused by the submarine convulsions of a strange and little-understood material buried underneath the seabed: methane hydrate.

Methane hydrates are part of a group known as the gas hydrates that occur when a gas molecule is surrounded by a “cage” of frozen water molecules. Technically they are clathrates (crystalline solids), similar to ice, except that they are not of uniform molecular composition: Part of their structure is provided by a “guest” molecule. In the case of methane hydrates the guest is a molecule of one of the most flammable naturally occurring substances: methane. The name clathrate comes from the Latin clathratus, which means “enclosed by bars or grating.” They have been known since the

early nineteenth century. Indeed it was Sir Humphrey Davy who first synthesized them, in the form known as chlorine hydrate, in his laboratory, from a chilled mixture of chlorine gas and water. In 1823 Michael Faraday published a paper entitled “On Fluid Chlorine” in the Philosophical Transactions of the Royal Society, a journal to which Wyville Thomson regularly contributed during the Challenger expedition.

Methane hydrate, the most common of the naturally occurring hydrates, comprise a truly vast energy reservoir. It is estimated that existing deposits contain enough energy to fuel the world for the next 350 to 3,500 years. The huge uncertainty of this estimate, though, reflects just how little we know about their distribution. They occur in two principal settings, on land in the high latitudes of the Arctic and the Antarctic, where the intense cold keeps the molecule stable in the permafrost, and in the deep ocean, on the outer continental shelves where the pressure of the overlying water keeps the clathrate stable. These two environments give us an important clue to the properties of methane hydrates: they can exist, or to be more accurate, remain stable, only within a narrow range of pressure and temperature. If the pressure is reduced by a few tenths of a bar or the temperature increased by only a couple of degrees Celsius, the hydrate spontaneously decomposes, releasing vast quantities—more than 160 times their crystalline volume—of flammable methane.

Until recently methane hydrates were regarded as little more than a laboratory curiosity or a nuisance when they caused blockages in gas pipelines. That changed in 1964 when a Russian drilling crew working in the northern Siberian gas field of Messoyakha discovered naturally occurring methane hydrates. Further prospecting revealed vast quantities under the northern Siberian Tundra, but it was not until the 1970s that methane hydrates were found in their other natural habitat: the oceans, where they were discovered by the Deep Sea Drilling Project’s vessel, and HMS Challenger’s namesake, the GLOMAR Challenger.

The ship was drilling off the coast of Guatemala when it unexpectedly penetrated a methane hydrate deposit. Previous cruises had logged the presence of vast seismic reflectors below the seafloor that were so dense that they were often mistaken for the seafloor itself. But these bottom simulating reflectors (BSRs) turned out to be the lower limits of undersea methane hydrate deposits. The extent of these BSRs is a clear indication of just how much methane hydrate is under the ocean floor.

BSRs exist because methane hydrate deposits have a relatively high seismic velocity; sound waves speed up when they encounter a methane hydrate reservoir. An echogram of the kind pioneered by Maurice Ewing shows this clearly. This velocity difference distinguishes hydrates from overlying sediment as well as the underlying BSRs themselves, which often cap reservoirs of natural gas. Before the GLOMAR Challenger’s cruise, smaller ships had penetrated these BSRs by coring, but they had not retrieved any methane hydrate because, relieved of its imprisoning pressure and temperature, the methane hydrates spontaneously decomposed on the way to the surface, vanishing into thin air before the researchers could even get a look at them. But the GLOMAR Challenger successfully retrieved a 3-foot length of core that contained methane hydrate and shipped it to the Colorado School of Mines, a leading center for energy studies in the United States, where it sparked a storm of interest.

Where do the methane hydrates occur around the world? Mapping their distribution is a complex high-tech task and it is no surprise to learn that their distribution around North America is particularly well known. One of the richest fields discovered so far is on the northern margin of the Bermuda Triangle on the seafloor topographic high known as the Blake Plateau.

With methane hydrate fields being discovered all the time, interest in them as a new form of energy is snowballing. A special leg of the Ocean Drilling Program (the successor to the Deep Sea Drilling Project) was recently commissioned to investigate the methane hydrates under the Blake Plateau. One of the most inter-

esting discoveries made by the scientists who studied the Blake Plateau material was the age of the methane hydrates. Using the radioactive decay of iodine they were able to determine their age as 55 million years old. This age is notable because it is precisely the age of the boundary between the Paleocene and Eocene epochs. For many years the P-E boundary (as it is commonly abbreviated) was not considered an important epoch boundary. But in the late 1980s Leg 113 of the Ocean Drilling Program drilled the Maud Rise near Antarctica and discovered a major change in the proportion of oxygen isotopes in sediments that spanned the P-E boundary. Put simply, a carbonate shell (such as a foram) growing in warm water incorporates more of the light isotope of oxygen (oxygen-16) than a foram shell growing in colder water, which incorporates relatively more of the heavy isotope (oxygen-18). Since the 1950s this approach has been developed into a fantastically sensitive paleo-thermometer that can now recognize temperature differences of less than 1°C.

Lowell Stott and James Kennett, working at the University of California at Santa Barbara, were stunned to find that the temperature record for the period across the P-E boundary in the deeper core was warmer than the record in the shallower core. This was quite the opposite of what they expected, because the temperature of the oceans generally decreases sharply with depth. How could deeper waters be warmer than shallower waters? Their explanation was that during this one critical interval of the Cenozoic (the last 65 million years of Earth history) deep waters were not forming in the high latitudes (as is the case today) but rather in the low latitudes and flowing away from the tropics, eventually arriving in the Antarctic and warming the deep waters there.

In the present day, as in much of the Cenozoic, deep waters are formed near the poles by surface waters that cool and sink as their density increases. These then flow out from the high latitudes, circulating cool water throughout the world’s ocean basins. What could have caused deep waters to form in the low latitudes and

disrupt the usual system of deepwater circulation so drastically? Stott and Kennett suggested that the answer was massive global warming, so intense that it was able to increase the density of low-latitude waters through enhanced evaporation, which concentrated it into something approaching brine. But this merely pushed the question back, for what could have caused such massive global warming? This time the answer lay in the carbon isotope record in the core. Stott and Kennett discovered that at exactly the same part of the core where the oxygen isotope anomaly had been found there was a large negative anomaly in carbon isotope values. There are two stable isotopes of carbon, carbon-12 and carbon-13, which differ only in the number of neutrons in their atomic nuclei. The former has 6, the latter 7. Conventionally the proportions of the two are expressed as a ratio normalized against a known standard. Where there is more of the light isotope of carbon the ratio is more negative, where there is more of the heavy isotope of carbon the ratio is more positive.

Stott and Kennett discovered that the normal ratio of carbon isotopes in the natural world was quite suddenly swamped by a preponderance of the lighter of the two isotopes—carbon-12. Normally the proportion of the two isotopes of carbon is controlled by the rate of photosynthesis in plant tissues. When photosynthesis proceeds more rapidly, proportionally more of the light isotope of carbon, carbon-12, is squirreled away into protoplasm. When this occurs in the ocean, the surrounding water is enriched in the heavy isotope of carbon and this is reflected in a higher ratio of carbon-13 in the hard parts of shell-secreting animals such as forams. But the carbon isotope ratio in the shells of the forams that Stott and Kennett analyzed was vastly more negative than could reasonably be explained even if there had been a total collapse in the rate of photosynthesis. Also, the shift was found to be equally large in forams that came from deep waters, which proved conclusively that the light carbon must have come from a reservoir outside the pool normally used by photosynthesis.

Methane has a carbon isotope composition about 60 times more negative than the isotopic composition of carbon in the ocean and the biggest reservoir of methane in the world is methane hydrate. If that methane were suddenly liberated from the hydrate form, the impact on the carbon isotope record would be immediate and severe. Indeed, modeling experiments have shown that it would be exactly the same as the record that was found in the deep-sea cores from the Maud Rise.

At about the same time that a young American geologist named Jerry Dickens was developing these insights, more ocean drilling was bringing up other material of P-E boundary age, from the mid-Pacific as well as the Blake Plateau area itself. All three records showed about the same magnitude of carbon isotope shift at the P-E boundary. At the same time, work on land led by the Swedish geologist Birger Schmitz showed an anomaly in land sections of P-E age around the Mediterranean region. The evidence was unequivocal: There had been explosive outgassing of methane hydrate deposits at the P-E boundary, enough to cause a global warming of up to 8^oC. As time went by, the details of the event were elaborated. Santo Bains and I, in our Oxford Laboratory, in conjunction with Richard Norris of the Woods Hole Oceanographic Institution, showed that there was not one event but several—a cascade of methane hydrate outgassing events, probably with each one triggering the next. We found, however, that the initial triggering event was nothing more sinister than an underwater earthquake, the kind of thing that happens all the time, especially close to the mid-ocean ridges.

But what had stopped this runaway greenhouse event once it started? Again it was detailed work led by Santo Bains that provided the answer: Bains analyzed the concentration of the element barium in sediments deposited across the boundary and showed that it increased markedly and then dropped back down again in perfect synchrony with the carbon and oxygen isotope records. The explanation was devastatingly simple. Vast amounts of methane had been

pumped into the atmosphere, causing greenhouse warming. But methane rapidly decomposes into carbon dioxide, which was then consumed by this vastly increased mass of photosynthesizing tissue greedily sucking all the liberated carbon dioxide back down into the ocean again.

What is the significance of the age of the Blake Plateau methane hydrates? They could not have spontaneously dissociated to plunge the world into the closest analogue we know to our own greenhouse future. Methane hydrates take eons to form, as dead organic material is cooked and compressed into raw methane gas and then locked up in a cage of enclosing water-ice crystals. The hydrates that let go at the P-E boundary must have predated it by millions of years. Is, then, the fact that the Blake Plateau hydrates are precisely the age of the P-E boundary a coincidence? No, it seems likely that the methane hydrates of the Blake Plateau today are themselves the record of the CO₂ that was released at the P-E boundary and that was eventually drawn down by the enhanced photosynthesis suggested by Bains and co-workers. The clue to the day the oceans boiled is buried under the Bermuda Triangle, and who can say that these methane hydrates are not too awaiting the day when they will once again be unleashed?

Since the hypothesis was first put forward in the early 1990s, methane hydrates have been implicated in mass extinctions elsewhere in the fossil record. For example, at the Cretaceous-Tertiary boundary (more usually associated with a probable meteor impact), 65 million years ago when the dinosaurs died out, the sudden climate change at the Cenomanian-Turonian boundary (90 million years ago), the climatic oscillations of the most recent series of ice-ages (2.5 million years ago to the present day) as well as the biggest series of glaciations of all time: those that characterized the late Neoproterozoic eon (600 million years ago).

Methane hydrates are not the only clathrates that can form in the deep ocean environment. Carbon dioxide hydrates can be induced to form in deep waters and the current thinking is that, if

artificially induced, these deposits might be able to serve as a storage medium for humanity’s excess CO₂. Peter Brewer of the Monterey Bay Aquarium Research Institute and his associates have successfully induced CO₂-hydrate formation at depths greater than 3 kilometers. At such depths, CO₂ hydrates are metastable, that is, that as long as the pressure and temperature remain unaltered they remain in this state forever. Brewer reasoned that if CO₂ from power plants were piped directly to the deep sea at the appropriate water depth, it could be stored there indefinitely in the form of hydrate so that it would not contribute to global warming.

Of course this is still experimental and much work must yet be done to determine this plan’s feasibility. But even if the idea is sound, there remains the question of whether we should do it. If there is one thing that we have learned from the legacy of the seafloor’s fossil record—and particularly the sediments that straddle the P-E boundary—it is that hydrates are not necessarily stable. They are subject to periodic earthquakes and volcanic activity that can destabilize them. It is likely that their eventual dissociation is simply a matter of time and statistics. Carbon dioxide hydrates stored in reservoirs on the seafloor would probably require protection as strong as that of a nuclear waste reprocessing facility to prevent an accident that would plunge the world into a climatic crisis dwarfing the one that took place at the P-E boundary.

Finally, returning to methane hydrates, there is no question that their energy potential is enormous. With confidence in nuclear power plummeting and environmental concerns about conventional oil and gas extraction growing daily, methane hydrates offer a comparatively cheap and easy form of new energy. But how is it to be extracted? Some think that the approach will be similar to that used to recover petroleum from difficult sites: Steam or hot water could be pumped down a drill hole to melt the hydrate, which would be collected from another drill hole. The resulting methane gas could then be piped ashore.

The problem with this method is that dissociating methane out

at sea could easily destabilize the continental shelf that supports the rig and, as we have seen, the larger scale consequences of a methane hydrate “event cascade” could be catastrophic. This hazard has led some scientists to suggest that retrieving the hydrate intact and then liquefying it on ships or drilling platforms is a better approach. One suggestion is to burn the hydrate to form hydrogen and carbon monoxide and then use a catalyst to convert the mixture into a liquid hydrocarbon, which could be readily transported by ship. The downside, though, is a 35 percent loss of energy.

Another approach is being considered by Roger Sassen of Texas A&M University, who envisions production on the ocean floor. Extracted methane could be recombined with water to form a new hydrate uncontaminated by mud and rock. Submersibles would then tow the hydrate in special storage tanks to shallower areas where it could be more safely decomposed into water and fuel.

Whatever the eventual extraction method, it is clear that methane hydrates are the fuel of the future. A harder question to answer is whether methane hydrates are responsible for the myth of the Bermuda Triangle.

Hamilton, Bermuda, April 4, 1873, 32° 18′ N, 64 ° 48′ W

It is one of the ironies of the Challenger expedition that despite the many phenomena and organisms that it did discover—the calcite compensation depth, the mid-Atlantic Ridge, manganese nodules, to name but a few— there were also spectacular secrets of the silent landscape that it did not uncover and one of these was the existence of methane hydrates. So, unaware of what they had missed, the voyagers arrived in Bermuda on the evening of April 4, 1873, sliding to anchor at Grassy Bay with the aid of a local pilot standing at the foretop and directing the four men at the wheel. The narrows through the reefs were treacherous and a knowledgeable pilot was a requirement. There were dozens of barely submerged corals all around them and Joe Matkin could see on many of them the impaled remains of less fortunate ships.

The island of Bermuda was strategically important to Victoria’s navy and the whole of the British North American Fleet was stationed there under the command of Governor Major General Lefroy. Because Bermuda was so important, the main town of Hamilton had the most elaborate docking facilities in the North Atlantic. These facilities included an enormous floating iron dock that had been recently towed all the way from Britain by four men o’ war, among them HMS Warrior. Built in 1861, Warrior was the jewel in the British navy’s crown. She was the ultimate expression of British naval supremacy, more than 400 feet long and armed with 26 muzzle-loading 68 pounders and 10 breech-loading 110 pounders. With a hull composed of iron plate she was almost single handedly responsible for keeping the peace with the French throughout the 1860s despite much saber rattling. But by the time Challenger arrived in Bermuda in 1873, Warrior had already been decommissioned from front line service. She had been rendered obsolete by the development of the fully steam powered turret warship based on the fearsome Monitor design pioneered by the Americans during their own civil war less than a decade before.

The harbor at Grassy Bay was only eight miles from the Bermudan capital, Hamilton, and all on board found the location convivial. William Spry wrote, “Nothing could have been more romantic than the little harbor stretched out before us: the variety and beauty of the islets scattered about; the clearness of the water; the quantity of boats and small vessels cruising between the islands, sailing from one cedar grove to another, made up as charming a picture as could well be imagined.” It was a welcome relief to the tedium of dredging and nothing short of blissful to have something to take their minds off the death of young Stokes only 10 days before. It was almost unbelievable, therefore, that amidst such tranquility and beauty, death should once again visit Challenger on their first night in port.

This time the victim was one whose continued well-being was central to the comfort and education of the youngest member of

the ship’s company. Joe Matkin had just retired for the night when he was woken by the ship’s writer Richard Wyatt in a state of great agitation. In the quiet of the crew’s mess, among the snores and grunts of a score of sleeping men, Wyatt heard strangled gasps coming from the hammock of ship’s schoolmaster Adam Ebbels, the man charged with the education of Captain Nares’s young son until they reached Australia. When Wyatt went to investigate he found the schoolmaster lying in his hammock, hands clawed, ashen-faced, and quite, quite dead. An inquest held by the ship’s surgeon in the morning found that he had died of apoplexy, or what we today would call a stroke. He was buried in a small nearby cemetery less than seven hours after his death.

There was more bad news. The mail brought news that the steamer Atlantic, one of the original five vessels of the White Star Line (the shipping line whose name would become forever associated with marine tragedy when Titanic sank several decades later) had sunk near Halifax on April 1st, running aground during one of the storms that plagued the region, with the loss of 560 passengers. The news triggered a rush of morbid sympathy below decks. Then, as now, there was a shared sense of peril among those who made their living at sea. This tragedy, coupled with the recent deaths of William Stokes and Adam Ebbels, was a potent reminder to all aboard of the dangers inherent in their enterprise. For Joe Matkin that feeling was compounded after Sunday church when he walked in the dockyard graveyard. He was struck by the number of naval men buried there. “You would be surprised to see what a quantity of seamen and Naval Officers have been buried there during the last 80 years,” he wrote in a letter to his cousin Tom. “There are 80 or 90 different Ship’s gravestones; each ship has a large stone with the names of all the officers, seamen and marines, that they have buried engraved thereon. Drowning, falling from Aloft, and Yellow Fever appear to have caused the most deaths. . . .” It was another reminder that a three-and-a-half-year voyage round the world, even in the year 1873, was no picnic.

Bermuda, April 21, 1873, 32° 18′ N, 64 ° 48′ W, to Station 43, May 1, 1873, 36° 45′ N, 71° 90′ W

HMS Challenger left Hamilton on April 21 and soundings confirmed almost immediately that Bermuda was indeed a vast seamount, towering 4 kilometers high from its base on the seafloor. She then shaped a course north and west toward the Canadian coastline under Admiralty orders to investigate the strange anomaly in the thermal structure of the North Atlantic known as the Gulf Stream.

This narrow band of surface water, originating near the Gulf of Mexico and flowing northeast toward the Newfoundland Grand Banks, was known to be much warmer than the surrounding waters. As Challenger traversed it, the crew dredged and sounded in the manner to which they had become accustomed, finding that the bottom here was more than 1,500 fathoms (3 kilometers) deep. With their marvelously intricate thermometers and water samplers, they discovered that the width of the stream in that area was fully 60 miles and at least 8°F (4°C) warmer than the waters on either side

The Gulf Stream had been discovered more than 200 years earlier by the Spanish explorer Don Juan Ponce de Leon. Ponce de Leon had been born into the royal court of Aragon and had begun his naval career as a crew member on Christopher Columbus’s second expedition to the New World. In 1508 he settled in Puerto Rico and made that island his base for continuing his explorations of the western Atlantic. On March 27, 1513, while searching for a miraculous fountain reputed to confer the gift of eternal youth, he landed on the peninsula that we now call Florida and claimed it for Spain. At the time he did not realize that he had reached the mainland of North America, thinking instead that he had merely found another island. Even so, he named the new land Florida in celebration of its discovery at Easter time (the Spanish name for this festival is Pascua Florida). Returning to Spain, he secured the governorship

of Florida and Bimini Island and returned there in 1521. But he was never able to enjoy his newfound dominion, because shortly thereafter he was wounded by a Seminole Indian arrow and died in Cuba. But during his extensive explorations of the Caribbean region he discovered the Gulf Stream and by 1844 the United States Navy had already begun the task of mapping its complex course systematically.

We know today that the name “Gulf Stream” is a misnomer— it implies a simplicity of structure that is not the case. In fact, the Gulf Stream is complex network of surface currents that shift course over time, sometimes disappearing entirely and reappearing elsewhere. The Gulf Stream is an example of a western boundary current, a discrete and narrow surface-water current that is confined to the western boundary of the ocean basins by the interaction of three factors: the general shape of the ocean basin; the rotation of surface waters (clockwise in the northern hemisphere, counterclockwise in the southern) in large-scale circulating features known as gyres; and the action of the wind over large fetches of open sea. The corresponding current in the Pacific Ocean is the Kuroshio, which flows north to about the latitude of central Japan before turning eastward for the open Pacific. In the southern hemisphere (in the Indian Ocean) it is the Agulhas Current. Because of the counterclockwise circulation of Southern Hemisphere gyres the Agulhas Current flows southward rather than northward.

The Gulf Stream is also the surface limb of the current system known to scientists today as the “North Atlantic conveyor” by which warm waters from the Caribbean and the Gulf of Mexico are transported northeastward and then return cool at depth in a current known as “North Atlantic deep water.” Nearly all the water that enters the Gulf Stream has been driven westward across the Atlantic by the northeast trade winds.

In the Caribbean and Gulf of Mexico, this stream of surface water is funneled between the continental shelves where its velocity increases to about four miles per hour. This “Florida current” then

turns north between the Florida peninsula and the Bahamas and flows north along the line of the North American continental shelf. By the time the current, now the Gulf Stream proper, reaches Cape Hatteras, its velocity has slowed to about one mile per hour, yet its average temperature can be as much as 11°C higher than the surrounding water. The fact that for a thousand miles south of the Gulf Stream the sea surface temperature changes by only 6°C puts this 11°C contrast into perspective.

About 1,500 miles north of Cape Hatteras, the Gulf Stream meets the southward flowing Labrador Current and the confluence of the hot and cold currents causes some of the most widespread fogs in the world. Further out into the Atlantic, the Gulf Stream separates into several different meandering currents that move in the general direction of Europe. In the center of the North Atlantic the Gulf Stream separates into two new currents, the Canary Current that heads south toward Spain and the coast of North Africa, and the North Atlantic Drift that brings warmth and moisture to Great Britain and northern Europe.

The importance of this effect should not be underestimated. In winter the air over Norway is more than 22°C warmer than the average for that latitude at that time of year. Contrast this with the eastern coast of Canada, where the prevailing winds blow out to sea and the Gulf Stream has little effect. Here Halifax, Nova Scotia, a thousand miles further south than Bergen, Norway has an average temperature of −5°C during its coldest month while Bergen remains above freezing.

The lower limb of the North Atlantic conveyor, North Atlantic deep water, is powered not by wind-induced currents but by density-induced sinking. The warm surface waters of the Gulf Stream lose a great deal of moisture through evaporation as they travel north, which has the effect of increasing their density. On top of this, they eventually reach some of the coldest oceans in the world—the Greenland Sea off the eastern coast of Greenland, the Norwegian Sea in an area north of a point midway between Iceland

and Scotland, and the Labrador Sea to the west of Greenland. In winter particularly, these regions are very cold. As the once balmy waters of the Gulf Stream take on the Arctic chill, their density increases further and eventually they sink. This newly formed North Atlantic deep water then starts to flow south again at a depth of more than two-and-a-half kilometers. The North Atlantic conveyor can be thought of as a giant engine for transporting heat from the tropics to the poles, and the Gulf Stream is a vital part of that engine.

It is a paradox that, as we face the risk of global warming, Great Britain and Northern Europe might be cooling. It is already known that subtle shifts in global climate have affected the course of the Gulf Stream, diverting it further south in the direction of Spain. The fear now is that as we warm the atmosphere through emissions of greenhouse gases we might be starting to prevent the northern-latitude waters of the Greenland, Labrador, and Norwegian Seas from becoming cool enough to sink. If that happens, the bottom limb of the North Atlantic conveyor will cease to function properly and the Gulf Stream will once again be diverted south, shifting the climate of Northern Europe at a stroke into the temperature regime more normal for those latitudes.

The Gulf Stream is the thermal blanket that keeps Great Britain warm, and even in 1873 Challenger’s crew were well aware of its role in warding off the frigid climate typical of western Canada. As William Spry put it, “Had our shores been without its warming influence, and the British Isles compelled to subsist on their own geographical allowance of heat, we should be left in the same condition.” Today we must appreciate that importance again lest the ice arrive and catch us unawares.

Halifax, Nova Scotia, May 9, 1873, 44° 38′ N, 63 ° 35′ W

Challenger arrived in Halifax, Nova Scotia at noon on May 9, 1873, on a last-minute decision by Captain Nares. He had planned to put in at New York instead, but the Narrative notes, “the usual dirty

weather was experienced on the passage towards New York: occasional strong winds, amounting sometimes to a gale, with light breezes intervening, and after crossing the Gulf Stream thick fogs, with rain, until close in to land.” Immediately after they finished the Gulf Stream soundings, the weather deteriorated even further, so Nares headed instead for Halifax. As Challenger steamed gently into harbor, Joe Matkin found himself noticing how very much like an English coastal city Halifax was, with its surrounding lighthouses, forts, and batteries. The weather, too, was more English than they had become accustomed to in recent months, cold yet bracing; a welcome change from the stifling heat of the Caribbean and Bermudas. For Joe Matkin, though, the most impressive sight was at night, when he watched the northern lights flicker and shimmer in great shifting veils across the vast Canadian sky.

Halifax was the principal naval station in the British Dominion of Canada and the dockyard was extensive and well equipped. Challenger had not sighted the wreck of Atlantic on the run in but the crew soon discovered that it was the principal topic of conversation in the town. Bodies were still being brought in for identification and burial, and the walls of the town’s predominantly wooden buildings were covered with posters, some describing victims still requiring identification, others offering rewards for information on those still missing.

Despite the grim business of clearing up in the aftermath of the tragedy all on board enjoyed Halifax. The food was excellent, “the fish market more plentifully supplied than any other known, cod, salmon, halibut and mackerel are very abundant and lobsters only 1d each,” wrote Joe Matkin. Taking advantage of this fecundity the ship’s steward, Alfred Taylor, and his assistant, Joe Matkin, took aboard six months’ worth of provisions and 200 tons of coal.

While Challenger was in Halifax, a mail steamer arrived from England, bringing welcome correspondence for all on board and fresh news of the home media sensation of 1873, the trial of Arthur Orton, the so-called “Tichborne Claimant.” A few years previously

Orton had claimed to be Sir Roger Tichborne, eldest son of the tenth Baronet of Tichborne and heir to the Tichborne Estates. Sir Roger was believed to be have been lost at sea off the coast of Brazil in 1854 but his mother, who had never reconciled herself to her loss, had become convinced that her long-lost son had by some miracle survived. The rest of the family was more skeptical and the case was brought to trial in 1871. It lasted more than three months and in the end Orton was found to be an impostor and sentenced to trial for perjury.

In May of 1873 the start of Orton’s second trial was only four months away and the media were in a frenzy of speculation as to whether the original judgment was correct and, in the event of a second conviction, how long he would be sent down for. Feelings on the issue ran high both at home and, as Matkin reported, on Challenger, too. Fresh news, especially when fuelled by unregulated shore-based alcohol, was apt to re-ignite fistfights among bluejackets with differing opinions. In time the case of the Tichborne Claimant would become infamous as one of the longest and most expensive trials in British legal history.

Challenger left for Bermuda again on May 19, 1873, having spent only 10 days in Halifax. The expedition was three weeks behind schedule and the Admiralty, which had calculated the duration of Challenger’s voyage so precisely that all aboard already knew that they were due back at Spithead in April or May of 1876, were anxious that they should make up the lost time before they reached the Cape of Good Hope. Several of the senior officers, including the aristocratic Lord George Campbell, were in New York, having made the 400-mile overland trip by train, and many would not rejoin the ship until she was ready to leave Bermuda for the Azores, several weeks hence.

But there was some compensation for those left behind in the warmth of the welcome they had received in Halifax. As the time for departure came near, the ship was overrun with visitors and their departure from Halifax seemed even livelier than their fare-

well to Portsmouth five months before. As they steamed slowly out into Halifax Bay to the stirring strains of Auld Lang Syne from HMS Royal Alfred’s brass band, the dockyard walls were crowded with cheering people. Alfred had supplied extra men to make up Challenger’s manpower deficiencies: Five seamen had deserted in Halifax, lured by the promise of instant wealth in the great territories of the United States, while another had been discharged and yet another hospitalized.

The effect of Challenger’s departure was somewhat spoilt, however, when von Willemoes Suhm, the German naturalist, found that he had mislaid his manservant, and Captain Nares had to send the steam pinnace ashore to find him. The junior officers somewhat irreverently dubbed von Willemoes Suhm “the Baron,” and Herbert Swire, the most disrespectful of all Challenger’s diarists, summed up the general amusement and irritation when he wrote:

After the cold of Halifax Challenger was soon once again in the balmy embrace of the Gulf Stream and “the weather was so warm that all the iron in the ship was dripping with damp and the change was considered very unhealthy, a great many are even now on the Sick List with Rheumatics and low fever,” Joe Matkin wrote. It would be five months before they saw any cold weather again, as they approached the Great Ice Barrier of the Southern Ocean

around Antarctica. To many on board, as they sweltered, that day could not come soon enough.

En route to Bermuda, the Challenger company celebrated the Queen’s birthday. All hands were issued one-third of a pint of sherry, a drink that was listed wryly in the ship’s manifest as “extra surveying stores.” For Joe Matkin the day was memorable for another reason: It was his father’s birthday. Charles Matkin was still languishing at home in deteriorating health and so the young man drank a quiet toast to his father instead, an offence that, if known, was punishable by death. On May 31,1873, a pilot took them back through the dangerous reefs that surrounded Hamilton’s harbor and they slipped quietly to anchor once again. Wine was issued to all hands again that night. It had been a difficult day and all on board were grateful for it.

No ship in Victoria’s navy was as well supplied with wine, pickles, and preserves as Challenger. Welcome as they were though, these extra supplies were not given through any spirit of altruism on the part of the Admiralty. No other ship would see so many changes of climate on its voyage, and the Lords in London knew only too well the value of preserving morale. There was, too, the omnipresent threat of scurvy, which could be combated only by a regular intake of vitamin C from citrus fruits. As Matkin wrote, “Captain Cook’s expedition round the world about 100 years ago lost nearly half their officers and men from Scurvy.” Even in the days of Victoria’s sail-to-steam navy no risks would be taken with the sailor’s age-old scourge.

The expedition had one final piece of unfinished scientific business to address before it could leave the western Atlantic for the last time. On June 9, 1873, Wyville Thomson and a party of the Scientifics and officers left the ship early in the morning aboard the steam pinnace. Stopping only to pick up the Governor, General

Lefroy, from his mansion on the heights above Hamilton, they headed northwest toward one of Bermuda’s most enigmatic natural wonders, the remote, forbidding Walsingham Caves. After an hour’s steaming they came to a channel between two narrow necks of land through which the tide rushed with the frenzy of the Severn Bore. When it had ebbed the pinnace slipped quietly through the channel and floated on a vast, glassy sheet of water that glittered like blue glass in the early morning sunlight. They were in the enclosed inland sea known as Harrington Sound, a most peculiar basin, rectangular in shape and about two miles long by a mile wide, that makes up much of the northernmost tip of Bermuda. To the south, a narrow strip of land protects it from the Atlantic while the other three sides rise in richly wooded ridges that form the highest ground in the islands. Completely landlocked, Harrington Sound is a perfect natural harbor and not many years earlier the Lords of the Admiralty had actually considered abandoning their Bermudan base on Ireland Island in favor of it. However, they eventually dismissed the idea on grounds of cost and convenience.

As the pinnace approached the northern shore of the sound, the party could see that the cliffs were pockmarked with a series of low caves that extended all the way to the water line. Such was the stillness of the water that it was impossible to tell where reality ended and the reflection began. In the still, clear air they could see the mats of green algae that covered the cave roofs reflected in the water’s quicksilver surface.

The caves are relicts from the last series of ice ages when sea level was 125 meters below its current level because so much water was locked up in polar ice sheets. At that time, beginning about one million years ago, the entire Bermuda platform was dry and the island’s land mass some 20 times greater that it is today. Although Bermuda today has no natural freshwater reserves of its own, at that time, in the mid-Pleistocene, a substantial body of fresh ground-water made up the bulk of the subsurface. This water eroded the natural limestone of the island into a series of natural caves that

became gradually drowned as the ice-sheets melted and the sea level rose.

Today the caves of Bermuda consist of a network of submerged passageways and larger hollows at an average depth of 20 meters below sea level. But it was one of the un-submerged caves, known as Admiral’s Cave, that Wyville Thomson had come to see. Fifty-four years before, in 1819, Sir David Milne, at that time commander in chief of the North American and West Indian Station, had discovered a very fine stalagmite growing from the floor of the cave. It measured 11 feet in length, was fully 2 feet in diameter, and weighed three-and-a-half tons. Milne removed it from the cave and eventually it found its way to the Natural History Museum at the University of Edinburgh, the institution where Wyville Thomson was now professor.

In 1863 Sir Alexander Milne, son of Sir David, First Lord of the Admiralty and like his father before him, also commander in chief of the North American and West Indian Station, visited the cave and examined the stump of the stalagmite. He was amazed to see that the stump was regenerating! As he watched he could clearly see drops of water falling from the roof of the cave onto the stump and where they landed two small knobs of calcareous matter were already being formed. He estimated the total bulk of these two small knobs, formed in the 44 years since his father’s visit, at about 5 cubic inches. The discovery set him thinking, and he realized that here was, in principle, a perfect way to estimate the age of the original stalagmite itself and from that the age of the Admiral’s Cave.

It was his brother, though, David Milne Home, who performed the calculation and presented a paper before the prestigious Royal Society of Edinburgh, estimating the age of the stalagmite and the cave system as some 600,000 years, a figure that sounds plausible given that modern estimates of the cave system’s age suggest a maximum of 1,000,000 years. And now, in 1873, 10 years after Sir Alexander Milne’s visit, Wyville Thomson, chief scientist of the greatest scientific expedition ever organized, was anxious to

examine the stump to see for himself how much more it had grown. But when the Scientifics examined the stump, they found that the tiny calcareous lumps had not perceptibly grown since Sir Alexander Milne’s visit.

Despite this disappointment, all agreed that the caves of Bermuda were magical. In all the caves they were surrounded by beautifully fluted and fretted columns whose pure white frosted surfaces shone out like beacons in the harsh magnesium light of their lanterns. But it was Painter’s Vale Cave that induced their greatest awe. At the foot of a bank of debris lay a pool of deep clear water, perfectly still and reflecting the roof like a mirror. Clambering down the slope, as their eyes became accustomed to the dark, they could see that the lake stretched far back into the gloom. A little punt was moored near the shore and, lighting candles, Nares rowed Governor Lefroy back into the darkness. From the narrow shore the party watched as the dim light of their candles receded and dimmed, their voices became hollow and distant, and they became denizens of the underworld—troglodytes.

They were not alone in enjoying this subterranean existence because Bermuda’s caves support a diverse fauna specially adapted to a lightless existence. Today we know that the fauna of Bermuda’s caves represents what is known as a biodiversity hotspot. Biodiversity hotspots are defined as small areas with exceptional concentrations of unique species. Bermuda’s caves qualify as such because of a very rich community of cave-dwelling animals known formally as anchialine organisms. Two orders, one family, and fifteen genera of crustacea—more than 60 species in all—from the caves were recently identified and they are all new to science.

These animals, often without eyes or pigmentation, neither of which is needed in the lightless world of the caves, are highly endemic; that is, they are restricted to a narrow strip of land between Harrington Sound and Castle Harbor; the north shore that Wyville Thomson and the Scientifics visited that long ago day in 1873. Many of the cave-dwelling animals are even more restricted, some

being unique to individual caves. They get their nutrition from planktonic organisms brought in on the sluggish tides where the caves connect with the ocean. These tides slowly replenish the deep fully marine waters with oxygen and nutrients while the surface waters, containing as they do a component from rainfall, remain brackish. Strangely enough, although it was not immediately apparent to the Scientifics, the caves of Bermuda hold one of the very things that had motivated the Challenger expedition in the first place—living fossils.

Some of the cave’s animals have organs and structures that are known only from the fossil record. On top of this, many of Bermuda’s cave-dwelling species are very similar to those known from Europe and Asia and must have arrived on the island before the spreading of the mid-Atlantic Ridge separated Bermuda from those continents.

Other animals in Bermuda’s caves seem to be close relatives of cave dwellers known from isolated islands elsewhere in the Atlantic and, surprisingly, in the Pacific, too. Still others seem to be related to species known only from the deep sea. The question of how animals from such diverse habitats arrived in the caves of Bermuda is currently the subject of intense scientific research that is helping to answer broader questions about the evolution of oceanic species.

Despite the great antiquity of Bermuda’s cave system and its inhabitants, its ecology is extremely delicate and currently under severe threat from that most rapacious of all predators—man. This is worrying because of the uniqueness of the ecology. The huge impact of man on the caves, particularly in the last century, has been due mainly to construction activities and water pollution. The U.S. military has poured raw sewage into the pool of Bassett’s Cave, which as long ago as 1837 was known to be the most geologically interesting of all the caves. Other threats come from dumping, littering, and, tragically, from deliberate vandalism. Quarrying operations have already destroyed many caves and have ramifications further afield when the water of adjoining caves becomes polluted.

In such cases the death of the fauna is inevitable and because many animals are cave-specific this means extinction—the final and absolute removal of a species from our planetary ecology.

For Herbert Swire, the caves were only moderately interesting. He found it hard to share the Scientifics’ enthusiasm and treated the cave expedition as an excuse for a picnic. But for Swire the perfect picnic included a crucial ingredient, one that was sadly lacking that day at the Walsingham Caves: there were no ladies. Swire was a young man and felt the lack of female companionship especially keenly. He was also incurably romantic and not long after wrote in his diary, “I have just finished reading the late Lord Lytton’s last work, published since his death, Kenelm Chillingly. What a sad ending to a noble book! I think I have never been so near blubbering as I was when I came to the chapter in which Kenelm receives from Mrs. Cameron, Lily’s last message, in the shape of a note written before she died. I wonder has such a truly noble girl ever lived in reality upon this earth?” For Swire, though, as for the others, it would be many months before he had the female companionship he so desired, because the Azores, the Cape Verde Islands, and the vast South Atlantic beckoned.

Their final act before leaving Bermuda was to erect the grave-stone they had brought back from Halifax in the naval cemetery there. It was a large marble cross and bore the inscription “This stone is erected by the Officers and Crew of HMS Challenger, to the memory of Adam Ebbles, Naval Schoolmaster who died at Bermuda, April 4th: also to Wm H. Stokes, 1st-Class Boy, who was killed March 25th 73, off the West India Islands. In the midst of life we are in death.”