'Do we know how advanced the process is?' asked Lund.

'We've studied the data from the Thorvaldson expedition a few weeks ago,' said Bohrmann. He was trying to sound reassuring. 'That was when you first discovered the worms. The readings were normal then. They must have started rising since.'

'So what's the deal?' asked Stone. 'Is it getting warmer down there or isn't it?'

'It's not. The stability field is unchanged. The fact that methane's escaping must he due to processes occurring deep in the sediment. Deeper, in any case, than the worms could burrow.'

'What makes you so sure?'

'We've already proved-' Bohrmann broke off. 'With the help of Dr Johanson we've already proved that these creatures can't survive without oxygen. They can only burrow a few metres deep.'

'All you've proved is what happens in a tank,' said Stone, disparagingly. He seemed to have selected Bohrmann as his new arch-enemy.

'If the water isn't getting warmer, then maybe the seabed is,' suggested Johanson.

'Volcanic activity?'

'It's just an idea.'

'Well, it makes sense – but not in this region.'

'Can the dissociated methane get into the water?'

'Not in sufficient quantities, no. For that the worms would need to reach a gas pocket, or be capable of melting hydrates.'

'But they can't possibly have reached a gas pocket,' Stone insisted stubbornly.

'No, like I said-'

'I know exactly what you said. Now it's your turn to listen to me. Each one of those worms is radiating heat, the same as any living creature does. And the warmth they're creating is melting the ice. It only melts a few centimetres on the surface, but it's enough to-'

'The body temperature of a deep-sea creature matches that of its environment,' said Bohrmann, smoothly.

'But, even so, if-'

'Clifford.' Hvistendahl placed a restraining hand on Stone's arm. It looked like a friendly gesture, but Johanson sensed it was a warning. 'Why don't we wait for the next set of readings?'

'Bugger that!'

'You're not helping, Cliff Drop it.' There was silence again.

'What happens if the methane keeps escaping?' asked Lund.

'There are various possible scenarios,' said Bohrmann. 'Methane fields have been known to disappear. The hydrates can dissociate within a year. That could be what's happening here, and it's conceivable that the worms have triggered the process. If that's the case, large quantities of methane will be released into the air above Norway.'

'Just like fifty-five million years ago?'

'No, there isn't enough for that, and we really shouldn't speculate. Having said that, I don't see how the process can continue without a decrease in pressure or an increase in temperature, and there's no evidence of either. In the coming hours we'll send down the video grab. Maybe that'll clear things up. That's all for the moment.' And with that he left the room.

JOHANSON EMAILED Lukas Bauer in the Greenland Sea. He was starting to feel like a biological detective. Have you seen this worm? Can you describe it to me? Could you pick it out from five other specimens in an identity parade? Is this the worm that stole the lady's handbag? All relevant information will be noted in evidence.

First he wrote a few friendly lines about their meeting in Oslo, then enquired whether Bauer had detected unusually high levels of methane in the area where he was working. He'd deliberately left this point out of his other emails.

When he returned to the deck, he saw the video sledge dangling from the arm of the crane while Bohrmann's geologists inspected it. They were hauling it in. Not far away, outside the repair room, a group of sailors sat talking on a large chest filled with scrubbing brushes. Over the years, it had established itself as a lookout and living room combined. Draped in a threadbare cloth, it was known by some as 'the couch'. It was the ideal place to sit and poke fun at the unsteady movements of the research assistants and scientists, but there were no jokes today. The tension was affecting the sailors too, most of whom knew what the scientists were up to: there was something wrong with the continental slope, and everyone was worried.

From now on everything had to happen as quickly as possible. Bohrmann had asked for the ship to be slowed right down so that they could investigate a site he'd identified using data from the multi-beam echo-sounder and the video-sledge. Beneath the Sonne there was a large field of hydrates. Taking a sample meant releasing a monster that appeared to belong to the Jurassic age of deep-sea science. The video-guided grab – a pair of metal jaws weighing several tonnes – was scarcely the most sophisticated piece of technology. In fact, it was probably the crudest, yet most reliable way of wresting a chunk of history from the seabed. Opening its maw, it bit into the sediment, and tore out hundreds of kilos of silt, ice, fauna and stone, which it then deposited at the feet of the scientists. The sailors had named it T. Rex. As it dangled from the A-frame, jaws agape, ready to plunge into the sea, the similarity was striking. A monster in the service of science.

However, as with all monsters, the grab was powerful, but lumbering and dumb. Inside its jaws were floodlights and a camera, enabling its handlers to see where it was heading before they let it off the leash. That was impressive. But the dim-witted T. Rex was incapable of stealth. No matter how carefully you let it down – and there were limits, since it took force to penetrate the seabed – it created a bow wave that frightened away most creatures. As soon as their finely tuned senses detected it, worms, fish, crabs and any other organism capable of rapid movement escaped before it pounced. Even the more up-to-date instruments gave advance warning. The bitter words of a frustrated American scientist summed up the situation: 'There's plenty of life down there. The trouble is, it sees us coming and steps aside.'

The grab was lowered from the A-frame. Johanson wiped the rain off his face and entered the control room. A crewman was operating the joystick that moved the grab up and down. He'd spent the last few hours steering the video sledge, but he still seemed focused. He had to be: staring at hazy pictures of the seabed for hours on end had a hypnotic effect. A moment of carelessness, and a piece of equipment that cost as much as a brand new Ferrari would be lost forever.

Inside the control room the light had been dimmed. The monitors cast a pale glow on the watching faces of the people sitting and standing in front of them. The rest of the world no longer existed: there was only the seabed, whose surface they studied like a coded landscape in which every detail held a message.

Outside the cable slid over the winch.

The water looked as though it was going to spurt out from the monitors, then the metal jaws passed through a shower of plankton. The screens turned blue-green, then green, then black. Bright dots – tiny crabs, krill and other creatures – sped away like comets. Watching the voyage of the grab was like seeing the opening credits for the original Star Trek series, but now there was no music. It was deathly silent in the lab. The figures on the depth gauge were changing all the time. Then the seabed flashed into view, looking like a lunar landscape. The cable stopped.

'Minus seven hundred and fourteen metres,' said the man at the controls.

Bohrmann leaned over. 'Don't do anything yet.' The monitor filled with mussels. They liked to colonise hydrates, but now they were hidden by a mass of wriggling bodies. Johanson had a strange feeling that the worms weren't just burrowing in the ice but were eating the mussels in their shells. He could see jaws shooting out and ripping off chunks of mussel flesh, which vanished into the tube-like bodies. There was no sign of the white methane ice under the siege of worms, but they all knew it was there. Bubbles rose up from the bottom, with tiny shimmering fragments – splinters of hydrate.