As the sun was setting on August 18th 2003, the night fishermen of Hahaya village eased their wooden pirogues off the jagged lava rocks and slid into the water. The ocean off the western coast of Grande Comore was calm and as the half-moon rose, they could see the volcano of Karthala silhouetted against the darkening sky. A few hundred metres offshore, one of the fishermen, a veteran of decades of nights on the dark water, laid his paddles across the boat and prepared a line. He tied two flat black stones above a baited hook, then let the fine filament slip through his fingers until it touched the seabed, deep below.
He was waiting for the nibble and tug of a fish—a snapper or a grouper, perhaps, or if he was lucky, a marlin, which he would take the next morning to sell at the market in Moroni. But this time the tug was unfamiliar, and the old fisherman fought with the line before he managed to pull the fish to the surface.
Deep water at night is ink-black and the first thing he saw was a pair of eyes, glowing pink in the pale moonlight. As they surfaced, he could make out a large fish. He recognised it instantly as a gombessa, or coelacanth (pronounced see-la-kanth). Although rarely caught, it was known to all in the Comoros as their most precious asset, a fish that some said was the ancestor of man.
Only six coelacanths had been caught in the waters off Hahaya since 1966, and none in the previous five years, but the old fisherman knew what to do. He tethered it to the back of the boat and paddled back to the village. He knew there was little time to lose as gombessa live in the ocean depths and had never survived for more than a few hours at the surface. Determined to try, he made a safe water pool, and waited for the sun to rise.
The next morning, his nephew took the first bus into Moroni and went straight to the Centre National de Documentation et de Recherche Scientifique (CNDRS)—a handsome white building off the central roundabout in Moroni, which houses the national museum and archives. He told them about the catch. It was what they had been waiting for since the previous year, when Professor Rosemary Dorrington of Rhodes University in the Eastern Cape had visited the island to talk about a new project—the African Coelacanth Ecosystem Programme (ACEP)—that had been set up in South Africa. She had left behind some equipment and instructions on what to do if a coelacanth was caught. Her point man in Moroni was Said Ahamada, a young environmentalist.
Ahamada was at home when the phone rang. He rushed to the CNDRS, grabbed the collecting kit and then caught a bush taxi to Hahaya. “It was very emotional,” he remembers. “I was very impatient to see the fish. And when I got there it was still moving a little. It was a very big female, close to two metres, and had already turned brown. But its eyes were still shining; it was amazing to see lights coming from its eyes.”
The coelacanth was hauled out of the water and laid on white plastic sacking, where, almost immediately, it died. Following Dorrington’s instructions, Ahamada took blood samples. He paid the fisherman, then heaved the coelacanth, still wrapped in its sacking, into the boot of a red hire car and, clutching the vial of blood, careered back to Moroni.
The fish was laid out on a table at CNDRS. Ahamada cut carefully into its side and extracted samples from all the major organs—liver, heart, blood and gills. He carefully put each of them through a small manual meat-grinder, specially adapted for the task in case a lack of electricity made it impossible to use a blender, to homogenise the tissue. The samples were stored in the CNDRS freezer.
"I was excited that this fish from the Comoros was going to be used for science," Ahamada says. "But at that time I had no idea how important it would be." It wasn’t for another decade—until April this year—that he would find out exactly how important.
Coelacanths are the size of humans. They are slate-blue when alive, with white flecks on the thick scales that cover their bodies. They live in the gloaming, around 200-400 metres below the surface, where light barely penetrates and few creatures venture. They spend their days sheltering in rocky caves in small groups, coming up to feed at night as the water above them cools. Unlike most fish, they give birth to live young—small, perfectly formed baby coelacanths—and when disturbed they lift themselves into headstands, apparently using an electro-sensory organ in their snout to detect the presence of predators or prey.
The handful of people who have seen them in their natural habitat talk of their glowing eyes and their gentle demeanour. They describe coelacanths moving with surprising grace, deploying their fanned fins in a diagonal formation—right fin in front, left trailing behind—that is similar to a lizard walking.
It was those fins that first excited the attention of scientists, nearly two centuries ago. In 1839, the Swiss scientist Louis Agassiz described a fossil fish that had been found in Permian marl slate near Durham in the north of England. He named it Coelacanthus (from the Greek for hollow spine) granulatus (for the tubercular ornamentation on the surface of its scales). Over the decades, similar fossils were found across the world, dating from around 380m years ago to around 70m years ago, when the fossil record disappeared and the coelacanth was assumed to have become extinct.
The Coelacanthus fossils caused a stir in the scientific world, particularly after the publication of Charles Darwin's "On the Origin of Species" in 1859. In the coelacanth’s lobed fins, palaeontologists thought they saw clues to the identity of the "missing link", the first fish that crawled out of the sea to evolve into amphibians, reptiles, mammals and, eventually, man. They postulated that the lobed fins of the fossil coelacanths suggested that they were the ancestor of the first fish that crawled out of the sea. Others put their money on the lungfish, the first living specimen of which had been discovered in the Amazon in the 1830s by Johann Natterer, a Viennese naturalist.
Natterer had returned to Vienna from a collecting expedition with an eel-shaped creature, around two feet long, with both gills and functioning lungs, which he suggested in a monograph was a "new species of animal of the family of fish-like reptiles" (Ichthyodea). A year later an Englishman, Thomas Weir, came back from the Gambia with a similar lungfish, though this one was enclosed in sun-baked clay. (Scientists later discovered that it was common practice for lungfish to go into summer dormancy in the hot, dry season, and then wake up again when the rains came to melt their muddy nests.)
The debate over which of the two—lungfish or coelacanth, the one with the lungs or the leg-like flippers—was most closely related to our ancestor would rage for a century and a half. The evidence seemed to point first to one, then to the other. But without more fundamental information, especially the microbiological data locked inside genomes, it was never going to be conclusive. The lungfish was not going to provide the answers: it has a genome thought to be around 40 times the size of the human genome, and the modern lungfish was a different beast from its fossil ancestor. It was up to the coelacanth to unlock these evolutionary secrets, and, with the specimen that was caught in Hahaya, the bets were once more on the table.
By sheer chance, Said Ahamada was due to fly to South Africa the month after the Hahaya fish’s appearance, to attend the inaugural conference—in East London, on the southern coast—of the coelacanth project that Rosemary Dorrington had initiated. Dorrington and her colleague Greg Blatch had had the idea of trying to sequence the coelacanth genome. The genome is the library of hereditary material that contains both the active genes that determine how a creature looks, works and develops, and the non-coding sequences that include once-active strings of DNA. There, the scientists hoped to find clues to the coelacanth’s past, present and future, what it evolved from and into. It was a mammoth task, particularly for a small lab with basic equipment, but the pay-off would be immense. Written in those long strands of DNA, similar in size to the human genome, there could be the answer to one of the fundamental questions of evolutionary science: how did we evolve from fish?
A year earlier, in 2002, Dorrington and Robin Stobbs, a former technician at Rhodes and a long-time coelacanth fanatic, had flown to the Comoros to try to get some fresh coelacanth tissue to sequence. "I thought it would be easy, but then I realised that no coelacanth had been caught there for five years," Dorrington says. "To a great extent, the fishermen had been persuaded to change their fishing techniques by the Association pour la Preservation du Gombessa, in an attempt to protect the endangered fish—and the chances of catching one by accident were close to zero. It’s really incredible: the monetary value to them of catching a coelacanth used to be close to ten years' income, but they had decided they were not going to do something that would jeopardise the coelacanth. They are really amazing people."
When Dorrington booked her flight back from sabbatical in America to prepare her talk for the ACEP conference, she still didn’t know about the Hahaya fish and was worried that she wouldn’t be able to raise the $100m she estimated the project needed. But Ahamada’s news was a potential game-changer. As soon as he arrived from the Comoros, cradling his precious icebox, Dorrington whisked him off to her lab. She needed to see if the tissues had been harvested and frozen in time to be of use as samples.
With Ahamada at her side, she ran a quick DNA prep to see if there was enough there to work on. "I was nervous about it. But the genomic DNA from the Hahaya animal was of sufficient quality to work with. It was exciting stuff."
On October 29th 2003, following a gala reception in the Marjorie Courtenay-Latimer Hall in East London featuring an African dance performance by the Ngqoko Women’s Group, and reports from marine biologists from along the African coast, Rosemary Dorrington stood up to give a speech. She explained what she had done on her trip to the Comoros, and showed pictures of herself and Stobbs with Said Ahamada and his team outside the museum in Moroni. Then she introduced Ahamada, saying that a fish had been caught in the Comoros and that he had brought samples back with him to Rhodes. With a final flourish she projected a slide showing the cells taken from the Hahaya coelacanth. The raw material was there. The genome project could go ahead.
Watching from a seat of honour in the auditorium named after her was a spry 96-year-old woman with lively black eyes. It was with her that the modern episode in the life of the coelacanth began.
Seventy-five years ago, on December 22nd 1938, Marjorie Courtenay-Latimer was racing to finish a display at the East London Museum, where she was curator, when she received a call from the manager of a fishing fleet. He told her that the trawler Nerine had just docked and that the captain had some specimens that he thought might be of interest. She caught a taxi to the wharf and climbed aboard the boat. There was a pile of fish on the fo'c'sle. "I picked away the layers of slime to reveal the most beautiful fish I had ever seen," she told me when I first met her, 60 years later. "It was five feet long, a pale, mauvey blue with faint flecks of whitish spots and an iridescent silver-blue-green sheen all over. It was covered in hard scales, and it had four limb-like fins and a strange little puppy-dog tail. It was such a beautiful fish—more like a big china ornament—but I didn’t know what it was." The deckhand told her that it had been trawled at a depth of 40 fathoms off the mouth of the Chalumna River in the Eastern Cape, and that it had snapped at the captain’s fingers as he looked at it in the trawl net.
She managed to persuade the taxi driver to put it in his boot and took it back to the museum. Although she didn’t recognise it, a faint bell was ringing in the back of her mind from a school biology lesson about ganoid fish, an ancient group characterised by their scaly armour. "But I thought it couldn’t be a fossil fish because it was still alive." She knew she had to find a way to preserve it. She took measurements and drew a rough sketch while her helper, Enoch, went off to borrow a handcart, and together they set off into town.
They went first to the mortuary and then to East London’s cold storage—the only two refrigeration facilities large enough to accommodate the fish—but, three days before Christmas, there was no room at either inn. In despair, Courtenay-Latimer turned to the local taxidermist, who suggested she preserved the fish in a sheet soaked in formalin until she could find someone to identify it. She borrowed a sheet from her mother and wrapped it up. Then she tried to phone Dr J.L.B. Smith, a chemistry lecturer at Rhodes and honorary curator of fishes for the museums along the south coast. But he was away and when he hadn’t got back to her by the next day, she wrote to him, enclosing her sketch.
For the next few days, she waited for a response. By December 27th, oil was seeping from the fish and the taxidermist was worried that it would begin to decay. So Courtenay-Latimer told him to skin it – but carefully, so as to preserve the scales. They found pure white flesh below, no ribs and, instead of a spine, a flexible, oil-filled tube.
It was 13 long days before she heard from Smith. He was on holiday along the coast in Knysna, where he eventually received her letter and saw the sketch. “I stared and stared, at first in puzzlement,” he wrote in “Old Fourlegs: the Story of the Coelacanth” (1956). “I did not know any fish of our own, or indeed of any seas like that; it looked more like a lizard. And then a bomb seemed to burst in my brain, and beyond that sketch…I was looking at a series of fishy creatures that flashed up as on a screen, fishes no longer here, fishes that had lived in dim past ages gone, and of which only fragmentary remains in rocks are known…What I suspected was so utterly preposterous that my common sense kept up a steady fire of scorn for my idiocy in even thinking of it.”
He sent Courtenay-Latimer a wire urging her to save the fish's innards. "From your drawing and description," he wrote, "the fish resembles forms which have been extinct for many a long year."
On February 16th 1939, Smith finally made it to East London to view the stuffed fish, which lay on the table in Courtenay-Latimer’s small office. A short man, bristling with intellect and not noted for his patience, particularly with the more dilatory students, he circled the coelacanth several times. He peered at it, stroked it, then turned to Courtenay-Latimer and said: "Lass, this discovery will be on the lips of every scientist in the world."
When Smith’s paper on the coelacanth was published in Nature, with the first line “Ex Africa semper aliquid novi” (“there is always something new out of Africa”), it was greeted with great fanfare. Newspapers and magazines around the world were full of the find, which was memorably acclaimed in the Eastern Province Herald as the "Best Fish Story in 50,000,000 years", and in the Illustrated London News as "One of the Most Amazing Events in the Realm of Natural History in the Twentieth Century". Smith named the fish Latimeria chalumnae, in honour of Marjorie Courtenay-Latimer and the area in which it was found.
Smith was frustrated by the loss of the coelacanth’s inner parts which, he believed, would have revealed much about its morphology and provenance—and, more importantly still, provide clues as to how evolution worked. He devoted the next 14 years to the search for another coelacanth. With his wife, Margaret, he scoured the coasts of southern Africa, looking for another specimen, leaving posters with a description and photograph of Latimeria and the offer of a £100 reward.
It was on Christmas Eve 1952 that Smith received the news he had been waiting for. Eric Hunt, the captain of a trading schooner, wired him to say that a fish had been caught off the Comoros, then still a French colony—and that he had better get there smartish to claim it.
Smith reached for a telephone. He tried to contact South Africa’s ministers of defence and transport and the head of the armed services, to no avail. In the vacuum of Christmas, he realised there was only one option: the prime minister, Dr D.F. Malan, the architect of apartheid, an anti-British, deeply religious creationist. With the help of the local MP, Smith telephoned Malan at his cottage along the coast. Mrs Malan picked up the phone and said the pm was in bed and she wasn’t prepared to disturb him. "10.30pm of the 26th December in the year of our Lord 1952," Smith wrote. "It was probably the lowest ebb of my life. The sands of time were running out, fate was screwing me down to the dregs…What on earth was I to do, for now there seemed no more hope?"
Then the phone rang. It was Malan. Smith, in stumbling Afrikaans, summed up the situation and ended with a plea for a plane, so he could fly to the Comoros and bring the coelacanth back to South Africa. "Your story is remarkable," Malan said when he was finished. "First thing in the morning, I shall try to get through to my minister of defence to ask him to locate a suitable aeroplane to take you where you need to go."
Smith sent a telegram to Hunt, the captain of the trawler: HOLD ON STOP GOVERNMENT SENDING PLANE. The following morning he was sitting in the unlined hull of a military Dakota heading towards the Comoros. But he was nervous: the captain had told him that he had not been able to reach anyone on the islands to warn them of their arrival. He hadn’t even been able to establish whether there was a landing strip in the Comoros.
They spent the night in Lourenço Marques (now Maputo) before flying low over the Mozambique Channel towards the Comoros. Smith soon caught sight of a string of thickly vegetated, mountainous islands, fringed by aquamarine water and, beyond, the indigo of the deep. The plane started to descend towards a slender airstrip. Looking out of the window, Smith saw a small boat tethered near a makeshift town. He realised it had to be Hunt’s boat, with the coelacanth aboard.
The plane landed in a tropical downpour. As the door opened, Smith saw Hunt’s face peering in. After he had been taken to meet the governor, Pierre Coudert, crisp in a white tropical uniform, Smith begged to be shown Hunt’s boat. There, lying in a kapok-lined coffin by the mast, was his fish. "God, yes! It was true! I saw first the unmistakable tubercles on the large scales, then the bones of the head, the spiny fins! It was a coelacanth all right. I knelt down on the deck so as to get a closer view, and as I caressed that fish I found tears splashing on my hands and realised that I was weeping, and was quite without shame. Fourteen of the best years of my life had gone in this search and it was true…It had come at last."
Back in South Africa the next day, Smith took the coelacanth, still in its coffin, to show to the prime minister.
"My, it is ugly," D.F. Malan said. "Do you mean to say we once looked like that?"
The coelacanth's allure did not fade with the discovery of the second fish and what was thought—at the time, anyway—to be its ancestral home. Scientists and museums around the world clamoured for a specimen of their own, while the public queued around the block when Smith’s fish was put on display in Grahamstown. A friend of the Smiths', Bee Rennie, recalled the excitement: "People were converging from all directions…From judges to candlestick-makers and heaven knows who. We saw the judge president sort of pushing his way in, next to Helen Campbell the very short hairdresser."
The French, aggrieved at having what they thought of as their poisson stolen from under their noses, decreed that, until further notice, only French scientists would be able to study any further coelacanths. For the next few decades, a handful of specimens were caught each year and taken to the laboratory of Dr Jacques Millot in Tananarive (now Antananarivo), Madagascar. Once he felt he had enough coelacanths, they started giving or selling further specimens to museums and research institutions. A coelacanth is on permanent display in the main hall of the Natural History Museum in London, and its counterparts are in most of the world’s major museums. Over 25 years of study, Millot published a highly detailed book in three volumes,"L'Anatomie de Latimeria".
The coelacanth exerted a hold on adventurers and romantics, who were attracted by its rarity—mere hundreds were thought to exist—and inaccessibility. In the 1980s, an East German scientist, Hans Fricke, hand-built two submersibles in which he succeeded in diving to depths of 300-400 metres. Here, after much searching and to his great excitement, he found—and filmed—coelacanths, hiding in rocky caves off the south-western coast of Grand Comore. "I always say it is a creature that doesn’t belong in our marine world,” Fricke declared. “It is a very special fish."
In 1997, a young American marine biologist called Mark Erdmann was on holiday on the Indonesian island of Sulawesi when he saw a coelacanth in the fish market. He took photographs and returned the next year to set up a base in the hope of finding another one.
Like Smith, he visited the local fishermen and put up reward posters. After months of waiting, on July 29th 1998, an Indonesian fisherman from the island of Manado Tua, Om Lameh Sonathon, caught the fish he knew as Rajah Laut, "King of the Sea". He towed it to the next-door island of Bunaken, where Erdmann was living with his wife, Arnaz. For half an hour, Arnaz swam with the coelacanth while Erdmann took pictures. But it was already dying, moving listlessly in the water. Erdmann grabbed his dissecting kit and hauled the coelacanth into a cooler chest to take it back to the main island. A few minutes later, it died.
"I was filled with excitement and adrenaline," Erdmann told me soon afterwards. "But at the same time it was heart-breaking to see it slowly dying, especially having swum with it. At risk of slipping into anthropomorphisms, I had the impression of great gentleness and intelligence. I can honestly say that if it had looked more alive when we had been photographing it, I would have had the impulse to let it go."
Erdmann’s paper on the Indonesian coelacanth was published in Nature, as Smith’s had been. It was greeted with similar excitement, both by the media and by scientists, most of whom were relieved that there was a larger world population of coelacanths than had previously been believed.
But the South Africans were not to be outdone. The first fish had been found in their waters and logic dictated there should be others. Deep-sea divers, using a technically complex method involving a mixture of three gases, tried descending to ever greater depths in the hope of finding a coelacanth. In June 1998, a South African diver died in the attempt. Two years later, three tri-mix divers came face-to-face with a large fish that they thought was a coelacanth, at a depth of 104 metres, off Sodwana Bay, just south of the Mozambique border.
They immediately started planning a return dive and on November 27th 2000 they found three coelacanths ranging from about 1 to 1.8 metres long. After 15 minutes on the bottom, they began their slow ascent. But at around 70 metres, two of the divers, Dennis Harding and Christo Serfontein, had a problem with their equipment. They made a dash for the surface. Harding lost consciousness and, despite his team’s best efforts to revive him, died. Serfontein regained consciousness in time to be taken back down to a depth where he could safely decompress. After 134 minutes in the water, he was taken to nearby Richards Bay, where he spent six hours in a decompression chamber. His companions, however, returned to dive again.
All these coelacanth specimens would reveal more and more about the workings of what was dubbed—like the giant sequoia tree and the horseshoe crab—a living fossil, an extant relic of ancient times. But where precisely it fitted into the evolutionary tree—whether it was indeed the direct descendant of our fishy ancestor—remained unresolved. The answer, it seemed, would only come with a more detailed examination of the coelacanth, at the cellular level. And that was what Rosemary Dorrington and Greg Blatch would set out to do.
After the 2003 ACEP conference in East London, however, Dorrington and Blatch realised that it was time to pass on the baton: their equipment was not up to the monumental task of sequencing the coelacanth genome. They arranged to meet up with Chris Amemiya, a professor of microbiology at the University of Washington in Seattle and another long-term coelacanth fan. He was excited by the idea of unlocking the secrets hidden in the coelacanth’s cells and organised for the samples to be flown to America.
"As a little kid, I had read ‘Old Fourlegs’ and was fascinated," Amemiya says. He knew that sequencing the entire genome was going to be a long haul. Fish genomes had been sequenced before; the puffer fish was the first in 2002. But the coelacanth was of a different order of difficulty—and importance. In 2003, there were only a handful of places in the world that could do that type of work. One of them was the Broad Institute in Boston, which jumped at the chance to be involved. One of its research scientists, Jessica Alföldi from the Vertebrate Genome Biology Group, was given responsibility for running its end of the project. Together, Amemiya and Alföldi wrote white papers to raise the grants and, bit by bit, the funding and aims of the project started to come together.
By the time the tissues from the Hahaya coelacanth started being run through the Broad Institute’s bank of state-of-the-art DNA sequencing machines, a team of 91 scientists from 40 institutes in 12 countries on all six inhabited continents was in place, waiting for the data to emerge. "The coelacanth genome consists of approximately 3 billion base pairs," Alföldi explained to me. "Each chromosome contains 50m to 250m base pairs. The machines we use can only sequence 100 base pairs at a time."
It was an intricate process of cutting and stitching, involving geneticists and computer scientists, all working at the technology’s frontline. Eventually they had a draft genome assembly with which to work. At that point, the biology began. The data had to be analysed, the interesting genes identified and isolated and then compared to similar genes in fish, mammals, humans—and lungfish.
The prize was the seat on the evolutionary tree at the fork where the fish branch met the tetrapods—the first four-limbed vertebrates and their descendants, including humans. "The coelacanth is evolutionarily a fantastic organism," Amemiya says. “Before our study, people had been using more conventional methods to determine the coelacanth’s phylogeny. The lungfish seemed to show higher affinities to tetrapods – but those data sets are less clear-cut in determining relationships and not everyone subscribed to that point of view. That is why we had to use the molecules to get that data.
“We knew before the sequencing started that it was going to be big—that we would find all sorts of stories in the genome.”
The team homed in on 251 genes—a far greater number than had been looked at before—from a wide range of different creatures, then compared one with another to determine where exactly the modern coelacanth sat on the tree of life. So it was that in April 2013, 174 years after Agassiz described the first coelacanth fossil, 75 years after Marjorie Courtenay-Latimer saw her beautiful fish, 74 years after Smith’s first paper was published in Nature and almost ten years after the old fisherman caught his gombessa in the waters off Hahaya, the coelacanth was once again the cover star of Nature. In their report, the result of close to a decade of work, Amemiya’s team concluded that "the genetic analysis strongly supports the conclusion that tetrapods are more closely related to lungfish than to the coelacanth".
J.L.B. Smith’s Old Fourlegs, it seems, is not our several-million-times great-grandmother after all, but rather our great—many greats—aunt. "But the coelacanth is more closely related to us than it is to a salmon or a shark," Alföldi told me. And because of the unmanageability of the lungfish’s genome, the coelacanth’s genes provide the best chance to understand how life emerged from the waters to colonise the land.
The answers are appearing already. By comparing the coelacanth to land-living creatures, Amemiya and his team are starting to learn how genes changed, which were lost and which adapted, how we came to be able to breathe, smell, excrete, and walk on land. The coelacanth, for example has the same structure in its fins as we do in our arms: a stylopod (upper arm) and two zeugopods (corresponding to our radius and ulna). And unlike other fish, which have no fingers, it has the sequences in its genes to make autopods (fingers). Because it is impossible to study a living coelacanth, the scientists are now taking those autopod sequences and inserting them into the relevant place in mice embryos. Their experiments show that a mouse with those genes is able to make the proteins to grow fingers—a vital stage in the evolution of land animals.
And it’s just the beginning. In September Amemiya was putting the finishing touches to a further 11 papers, to be published at the end of this year, which will reveal more about our ancient forebear and what it was about it that made us who we are. The coelacanth genome has been published—something the team were determined to do from the beginning—and is now available for any scientist to use. The fish caught that night in Hahaya has already started to answer some of the biggest questions in evolutionary science, and should continue to bear scientific fruit for decades to come.
In the Comoros, few fishermen use the old techniques any more. "It is only the old men who have the patience during the night, when the coelacanths come up to feed," Said Ahamada says. In the deeper waters of southern Africa and Indonesia, our ancient ancestors are being left in peace again, to swim and to breed as they have done, virtually unchanged, for nearly 400m years.