Domestic Dread
The Mimic in the Range
I landed in Providenciales at about midday on Tuesday, 10 March 2026. The contract had come through the National Oceanography Centre in Southampton on 17 February, and I had said yes before reading the whole thing. That was usually a mistake. But the header had BATEC printed across the top, and I assumed anything that specific had proper funding behind it.
I landed in Providenciales at about midday on Tuesday, 10 March 2026. The contract had come through the National Oceanography Centre in Southampton on 17 February, and I had said yes before reading the whole thing. That was usually a mistake. But the header had BATEC printed across the top, and I assumed anything that specific had proper funding behind it.
The airport smelt of sun cream and diesel. I collected my holdall and a hard case containing the hydrophone kit, then found the man with the clipboard near baggage reclaim. He wore a faded polo shirt bearing the logo of a charter company I did not recognise.
He did not ask my name or request any identification. He simply checked my face against whatever photograph he had, then nodded towards the exit. A twelve-metre supply launch was waiting at a jetty about twenty minutes south.
It was a small vessel with a white hull and rust stains around the waterline. Two other passengers were already aboard, both wearing civilian contractor passes on lanyards.
We did not speak. I stowed my kit below, then came back up and stood near the stern while the engine turned over. The crossing to South Caicos took about three hours. The water stayed flat for most of the journey.
It was that Caribbean blue-green which does not look real until you are sitting in the middle of it. I watched the depth sounder climb as we cleared the shelf. One of the other passengers was sick over the rail about an hour into the crossing. The wind picked up after that, though not enough to matter.
I liked this sort of work. A clean end date, clear deliverables: turn up, validate some equipment, write a report and go home. The National Oceanography Centre used me for support contracts fairly often because I did not ask many questions and did not cost much. That suited me.
I preferred working aboard other people's vessels anyway. You saw the installation, learnt the systems and left before the politics began.
This job was straightforward on paper. The British Atlantic Trials and Evaluation Centre needed someone to inspect a spread of old undersea range equipment and confirm that it still met its installation specifications. The network dated back to the 1970s, although parts of it had been overlaid with modern fibre-optic cabling. The tasking document called the work "documentation and validation".
That meant several days of tests, serial-number checks and photographs. Easy work, provided the weather held. Dr Helen Mercer's name appeared on the contract as supervising researcher. I had worked under her twice before, both times on short-notice equipment inspections.
Helen ran things efficiently and did not micromanage, which I appreciated. Her emails were always direct. This one had contained three lines.
"Martin, BATEC validation contract. Two weeks maximum. Confirm by Friday."
I had accepted on Thursday, 19 February 2026.
We reached the BATEC jetty shortly before 16:00. The establishment sat low and blocky against the coast, a mixture of prefabricated buildings and older concrete structures with an unmistakably military appearance. There were several antenna arrays, a water tower and not much else. The man with the clipboard handed me over to someone in Royal Navy working dress, who checked my identification again and issued me with an access card and a site map printed on ordinary paper.
My room was in the civilian contractor quarters near the northern end of the station. It had a narrow bed, a desk and a window overlooking the car park. I dropped my holdall and opened the hydrophone case to make certain nothing had cracked during the flights and crossing. Everything looked intact.
The unit was a simple portable acoustic test set, useful for baseline checks and little else. Its hydrophone could transmit test pulses only when immersed or coupled to a suitable test transducer. For dry work, it also had a small airborne calibration speaker and a contact transducer that could be clamped to a pipe, casing or bulkhead. A selectable wideband calibration-noise mode swept broadband energy across the working range when checking receivers and filters. It was not subtle, but it was useful for finding faults.
I had brought the set because Helen's tasking sheet mentioned preliminary signal validation, and I intended to make several ambient recordings before touching any part of the permanent array.
I ate in the station mess at about 18:00. Grilled snapper, peas and rice. The only other people there were two women in civilian clothes who did not look up when I sat down.
Afterwards, I returned to my room and read the rest of the tasking document. It listed six pieces of equipment by serial number, all hydrophones or transponders and all supposedly still active.
My job was to confirm that they powered up, measure their frequency response and verify that their positions matched the installation records. Two weeks seemed generous, but I was not going to complain.
I could not sleep that night. That happened sometimes in unfamiliar places.
At about 22:00, I put on my jacket and boots and walked down to the pier with the acoustic set in a rucksack. The water was dead calm. There was no vessel traffic and no illumination except the lamps along the jetty and a handful of lights in the main building.
I lowered the hydrophone over the edge and ran its cable back to the recording unit, then secured the case to the jetty railing so it could not slide into the water.
The unit had a small screen showing the live input as a waveform with a frequency readout. I switched it on and watched the baseline settle. There was little beyond a low mechanical hum, probably from an onshore pump or generator, and a few transient clicks that could have been anything.
I let it run for a couple of minutes and began thinking about returning to my room. Then I felt the sound before I properly heard it.
A clean, whistle-like tone rose through the water and struck me in the chest. It was low, perhaps 60 or 70 Hz, near-infrasonic and strong enough to register through my body before my hearing fixed on it.
The waveform on the screen spiked into a smooth, symmetrical shape. It held steady for about four seconds, then stopped.
I assumed it was a calibration signal. BATEC operated an underwater trials range, so test transmissions were probably routine.
I checked the duration and frequency. Both were clean. There was no modulation and no drift, only a pure tone.
Then it happened again.
The same frequency, the same duration: four seconds, perhaps fractionally less. I waited for a third transmission, but the water remained quiet. I took out my mobile, opened the notes application and recorded the time and approximate frequency.
There was no clear bearing, which bothered me. Even with a basic hydrophone installation, there was usually some indication of direction. This signal seemed to come from everywhere and nowhere. Reflections could do that, particularly over a flat, hard seabed, but not usually so completely.
I coupled the test set's transmitter to the immersed hydrophone and sent a standard test sequence: three short pulses with regular spacing, the sort of signal used to determine whether a transponder was awake and listening.
The water returned it to me.
Not immediately. The reply came perhaps ten seconds later. But it was my sequence, stretched and wrong.
The three pulses were present in the same order, yet their spacing was uneven and the tone wavered as if someone had copied a recording through a damaged loudspeaker. I played it back twice to be certain.
It was unquestionably my sequence. It was also unquestionably distorted.
I logged the delay, timing and distortion, then packed the equipment and carried it back to my room.
I did not know what it meant, but it felt like something that ought to be reported.
The following morning, Wednesday, 11 March, I found Chief Petty Officer Callum Reed in the station operations building. He was the Royal Navy liaison named on my paperwork. His office was a small room with a desk, a metal filing cabinet and a window overlooking the water. He was typing when I knocked.
"Yes. Come in."
"Martin Shaw, National Oceanography Centre contractor."
He looked up. He had dark eyes, close-cropped hair and the squared military posture of someone who could sit upright without appearing to make any effort.
"Right. You're the validation technician."
"Yes. I ran some baseline checks last night and picked up something unusual."
He said nothing, merely watched me.
I took out my mobile and showed him my notes.
"A clean whistle tone, repeated twice. Then I transmitted a test sequence and received it back with distortion. There was no clear bearing on any of it."
Callum looked at the screen for perhaps three seconds, then handed the mobile back.
"Probably a fault in your set."
"It isn't the set."
"Then it was a reflection from the shelf. We get that sometimes."
I did not press him. He plainly did not want to discuss acoustic anomalies, and I was not there to argue about them. I only wanted the event on record in case it happened again.
"Stick to the programme," Callum said. "You've got the serial list?"
"Yes."
"Good. Start with the North Range array. We'll have a boat ready at 08:00."
That was all.
I returned to my room and prepared the kit. Callum met me at the jetty at 07:50. He was already sitting in a five-metre rigid-hulled workboat with an outboard motor and a canvas kitbag wedged beneath the forward bench.
"Morning," I said.
He nodded. "We're not going to the North Range array."
I stopped halfway into the boat.
"The programme says-"
"Change of plan." He did not look up from the painter he was coiling. "I need you to inspect an older node first. Visual confirmation."
I sat on the aft bench and placed my rucksack between my feet. It contained the portable acoustic set, a handheld GPS receiver and a backup data logger I had packed out of habit.
"Which node?"
"SR-4. It's near the exclusion-buoy line."
He started the outboard. It caught on the second pull and settled into a rough idle.
"It went silent the previous night. Probably corrosion. Maybe a failed seal."
That sounded routine enough. Nodes failed all the time. Seawater found every weakness eventually.
I assumed we would recover it, document the damage and replace it if Callum had a spare unit in the canvas bag.
"Do you have the serial specifications?" I asked.
"On my mobile."
He backed us away from the jetty, then turned south along the coast. The water was flat and there was almost no wind. For the first ten metres or so, I could see the seabed clearly: pale sand, scattered coral heads and patches of dark seagrass. The sun was already warm against the back of my neck.
We travelled offshore for about fifteen minutes. The coastline shrank behind us, and the water deepened to the blue-black colour that marked the edge of the shelf.
Callum steered with one hand and held his mobile in the other, checking it every minute or so. I did not ask what he was studying. I assumed it was a navigation display or the node's recorded position.
Ahead, a line of orange Ministry of Defence exclusion buoys stood roughly sixty metres apart. Each carried black BATEC markings and trailed a narrow cable into the water.
"That's the line," Callum said.
He reduced the throttle.
"We're going just inside it."
I took out my GPS receiver and marked our position. The exclusion line did not appear on any of the maps I had been issued, but that was hardly surprising. A joint Ministry of Defence and Royal Navy trials establishment was not going to publish all its infrastructure.
Callum brought the outboard down to idle and allowed the workboat to drift. He searched the water as if looking for something specific. A minute passed, then another.
"There," he said, pointing to a patch of water about forty-five metres away.
I saw nothing at first. He opened the throttle and steered towards it. As we drew closer, I caught a metallic glint approximately three metres below the surface.
A cylindrical housing about the size of a fire extinguisher was fixed to a frame bolted into the seabed. It had no buoy or surface marker. The node simply stood in open water.
"That's SR-4," Callum said.
He cut the engine. The silence was abrupt. There was no wind and no noticeable chop, only the faint creak of the hull and the occasional slap of water against it.
"Do you want me to free-dive to it?" I asked.
"Not yet. I want to make a radio call first. Let operations know we're on site."
He reached beneath the bench and took out a handheld VHF radio. He keyed the microphone twice, then spoke.
"BATEC Operations, this is Reed. On site at SR-4 for visual inspection. Confirm receipt."
Static hissed from the loudspeaker.
He waited five seconds and tried again.
"BATEC Operations, Reed. Radio check."
More static followed, then a faint crackle but no voice. He tried a third time with the same result. Frowning, he adjusted the squelch and keyed the microphone again.
The static cut out. A clean whistle came from the radio's loudspeaker, low and steady at roughly 60 Hz. It lasted half a second, then broke into three rapid bursts.
The pitch and smoothness were identical to the signal from the previous night. Then the channel went silent.
Callum stared at the radio.
"That's the tone I heard from the jetty," I said.
He did not answer. He keyed the microphone again. The static returned, but beneath it I could hear a faint whistle, as though it were bleeding through from another channel.
He lowered the radio and clipped it to his belt.
"We'll call from the platform."
"What platform?"
He pointed east.
I followed his gaze and saw a small square structure about four hundred metres away. It looked like a maintenance platform, with solar panels on its roof and a metal ladder on one side.
"Engine trouble?" I asked.
"Radio trouble."
He reached for the starter. The outboard coughed once, spluttered and died. Callum tried again. It turned over but would not catch. He adjusted the choke and pulled the cord three more times without success.
I examined the fuel line. It was connected, with no obvious kinks or leaks.
"When did you last run it?"
"Yesterday."
He continued pulling the cord. On the sixth attempt, the engine coughed again and died at once.
That was when I noticed the water.
It was rolling. Not breaking or chopping, but moving in long, slow swells that lifted the workboat perhaps thirty centimetres and lowered it just as gently.
The cycle was perfectly regular: five seconds up, five seconds down.
I looked at the sky. It was clear. There were no clouds and no wind. Nothing offshore suggested a storm.
There was no reason for the sea to move in such a rhythm. It felt mechanical.
Callum stopped pulling the cord. He was watching the water as well.
"Current?" I asked.
"No current moves like that."
The next swell raised us. My stomach dipped as we crossed the crest and began the smooth descent down the other side.
Too smooth.
"We're drifting towards the platform," Callum said. His voice was flat and controlled. "We'll use the landline there."
I did not argue. The workboat was already travelling east, carried by the swells. Callum took the oars from beneath the bench, fitted them into the rowlocks and began rowing to increase our speed. He said nothing else.
I kept watching the water. The swells continued: five seconds up, five seconds down. I counted six cycles, then stopped. My chest felt tight.
The platform grew larger. It was older than I had expected, with rust streaking the support legs and barnacles crusted around the waterline. The deck was perhaps two and a half metres square, enclosed by a waist-high railing, with a metal equipment cabinet bolted into one corner.
The solar panels above it were intact. The ladder looked sound. Callum rowed alongside and caught one of the support legs to steady us.
"Hold us there."
I reached up and gripped the railing while he secured the painter to a cleat welded to the nearest leg. The boat struck the platform twice, then settled. Callum climbed up first. I passed him his canvas bag, then followed with my rucksack.
The deck was smaller than it had appeared from the water, perhaps two metres by two and a half. The solar panels provided some shade, but the air beneath them was hot and still. I could smell rust, old paint and something damp and organic, like seaweed left too long in the sun.
The equipment cabinet occupied the eastern corner. It was a weatherproof glass-fibre box with a hasp and padlock, except that the door hung open.
Callum saw it when I did. He walked over and stopped thirty centimetres away. I joined him and looked inside.
The cabinet was not merely damp. It was wet. Water lay two or three centimetres deep in the bottom, and the interior glistened as though someone had just emptied a bucket into it. The radio transceiver, battery pack and junction box inside were soaked. The rapid corrosion forming on the terminals suggested salt water.
"That didn't come from spray," I said.
Callum crouched and examined the hasp. It was undamaged. The padlock was missing, but there were no pry marks and no sign that the door had been forced.
I stepped back and examined the deck. That was when I noticed the footprints.
They were faint, pressed into a thin layer of gritty residue, probably dried salt and algae, which covered much of the platform. Five bare footprints formed an uneven trail from the ladder to the open cabinet.
Callum came over while I knelt beside the clearest impression.
It had five toes, but they were elongated and splayed at unnatural angles. The big toe sat too far from the others, and the jointing looked wrong. Most of the pressure had fallen across the ball and outer edge of the foot, with almost no heel impression, as though whoever made it had walked on the front of the foot.
The print was about twenty-five centimetres long and unusually narrow. I took two photographs while Callum stood beside me, one close view of the clearest impression and another showing the line of prints between the ladder and cabinet.
"You see that?" I asked.
He studied them for several seconds, then straightened.
"Someone came out to inspect the equipment. Probably maintenance."
"Barefoot?"
He did not answer. He took the VHF radio from his belt and tried it again. The loudspeaker hissed, then fell silent. There was no whistle this time, only dead air.
I looked back at the footprints. They made no sense. BATEC personnel did not perform offshore maintenance barefoot, and feet did not normally have toes and joints shaped like those. The gait was toe-heavy, the proportions wrong.
Callum clipped the radio back to his belt.
"The boat's drifting away. We'll get aboard, row to shore and report to operations."
He was right. The workboat had moved about five metres from the platform and was still being pulled by the slow, regular swells, held only by the painter.
But Callum did not move towards the ladder immediately. He walked instead to the southern railing and stared out across the water. His search was not casual. He divided the sea into sections, checking each quadrant in turn, beginning at the platform legs, working out towards the buoy line and then beyond it to open water.
His jaw was tight. He gripped the railing with both hands.
"What are we looking for?" I asked.
He did not answer.
I stood beside him. The swells were still moving in their exact five-second rhythm. Up, down, up, down, with perfect spacing.
The water was clear for perhaps six metres around the platform legs before falling away into blue-black emptiness. There were no shapes or obvious movement, only barnacles and several strands of weed swaying with a current that should not have existed.
"Callum, we should go."
He did not look at me.
"Agreed."
Yet he remained at the railing. His knuckles had whitened around it.
I waited through three more swell cycles. The tightness in my chest had not eased, and standing there was making it worse. I wanted to leave the platform. I wanted to be ashore. I wanted to be anywhere that did not contain malformed bare footprints with elongated toes and impossible jointing.
At last, Callum released the railing.
"Let's move."
He climbed down first. I waited until he was in the workboat, then swung one leg over and followed. The ladder was slick, and rust flaked beneath my palms.
Halfway down, I glanced at the water and saw nothing. Only the swells. Only blue.
I dropped into the boat. Callum was already fitting the oars into the rowlocks. He did not try the outboard again, but began rowing without a word.
We were perhaps ten metres from the platform when I remembered the radio. It remained clipped to his belt.
He had tried it twice while we were on the platform: once beside the open cabinet and once near the railing. Both attempts had produced dead air. No static, no whistle, only silence.
I do not know why I asked him to try it again. Perhaps the silence was worse than the whistle. Perhaps I wanted proof that the radio was genuinely faulty and not simply selective about when it worked.
"Try it again," I said.
Callum stopped rowing. He looked at me, then at the radio. He unclipped it and held it between us.
"I've already tried."
"I know. Try again."
He pressed the transmit button for two seconds, then released it. Nothing happened.
He clipped the radio back to his belt and reached for the oars.
The radio clicked once.
The sound was sharp, metallic and loud enough that we both heard it. Callum froze with one hand halfway to an oar. I stared at the radio. It was still attached to his belt. His hands were nowhere near it.
Then it began transmitting.
There was no button press and no gradual rise of static, only a clean, open channel.
The sound was not the whistle. It was worse.
It resembled a calm, level human voice, but there was a wetness in its tone that made my teeth ache. Static lay beneath it, along with a layered quality, as though two people were speaking almost, but not quite, in synchronisation.
"21.474, minus 71.521," it said. "Approach."
The transmission ended. The radio went silent.
I looked at Callum. His expression had emptied. He did not appear confused or surprised. It was the blankness of someone working very hard not to react.
"Do you know those coordinates?" I asked.
He picked up the oars and began rowing harder.
"Callum."
"Yes," he said. "I know them."
"Where are they?"
He completed another three strokes before answering.
"A disused shore repeater chamber. Roughly one and a half kilometres south of the main jetty."
"What is a shore repeater chamber?"
"Concrete access structure. Old infrastructure. It connects to the undersea network."
He continued rowing without looking at me.
"Through tunnels?"
That caught his attention. He stopped and stared at me.
"What tunnels?"
"You said it connects to the undersea network. That means tunnels, cable conduits, whatever term you prefer."
He looked away and resumed rowing.
"It's sealed. Has been for years."
"Then why would something tell us to go there?"
"I don't know."
I did not believe him. His voice was too flat, too carefully controlled. It was the kind of control people used when they knew they were lying badly.
I leant forwards.
"Callum, what is this actually about?"
He rowed another ten strokes. The platform receded behind us and the coastline drew closer. I could see the main jetty in the distance, a small grey line against the green shore.
Finally, he shipped the oars and allowed the boat to drift. For a moment he sat with his hands on his knees, staring at the floor.
"This isn't only validation work," he said.
I waited.
"We lost a diver last year. Petty Officer Owen Price. He disappeared on 18 September 2025 while carrying out maintenance on the South Range array. He didn't surface. We conducted a four-day air-and-sea search. All we recovered were his cylinder, fins and surface marker, found near the shore repeater chamber."
"And you didn't mention this because?"
"Because it isn't your job. Your job is to inspect the equipment and confirm that it matches the installation specifications. That's all."
"Then why am I here?"
He looked up.
"Because the equipment began behaving incorrectly after Owen disappeared. Signals looping. Whistle tones where there shouldn't be any. The range answering transmissions."
He paused.
"The Ministry of Defence wants confirmation and containment. It wants to know what's down there and whether it stays there."
I sat back. The tightness had returned to my chest.
"You're telling me this is a recovery mission."
"No. Owen is gone. This is a containment mission."
"Containment of what?"
"I don't know."
There was the same flat voice and the same poor lie.
I looked towards the shore. We had drifted another six metres. The swells remained regular and wrong.
"We need to return to the station," I said. "Get a full team and proper equipment. Do this in daylight with a support crew."
Callum shook his head.
"We can't wait."
"Why not?"
"Because if the range is answering, someone else may be listening. If someone else is listening, we lose containment before we understand what we're containing."
"Who do you mean by someone else?"
"Foreign monitoring. Commercial salvage. It doesn't matter. What matters is that this remains internal."
I stared at him.
"You want to respond to that transmission now because you're worried about eavesdroppers?"
"Yes."
"That's insane."
"That's the job."
I looked at the radio on his belt. It had made no further sound. It sat there, silent and harmless-looking.
"I'm a contractor," I said. "I inspect hydrophones. I don't conduct containment operations. I don't do search and rescue, and I certainly don't do whatever this is."
Callum picked up the oars.
"You signed the contract. You're here, and at present you're the only technical resource I have."
He began rowing towards the shore and towards the coordinates.
"So you're doing containment work."
I neither agreed nor argued. I sat there watching the coastline approach, trying to calculate how badly I had compromised myself by accepting the contract on a Thursday night without reading the fine print.
Callum seized the starter cord and pulled hard, with the sort of effort that was half mechanical technique and half refusal to accept failure. The outboard coughed, caught and died.
He pulled again. The same thing happened.
On the third attempt, the engine turned over and remained running. Its idle was rough and it spluttered every few seconds, but it held.
Callum did not look at me. He shipped the oars, took the tiller and opened the throttle.
I did not ask what had changed. Perhaps nothing had. Perhaps the engine had always intended to start and simply required spite as fuel.
We turned south and travelled parallel to the coastline. The swells continued in their rhythm: five seconds up and five seconds down. I stopped counting because it did not help, and because my chest tightened whenever I thought about the regularity for too long.
Callum steered with one hand. The radio remained clipped to his belt. He did not try it again, and neither did I.
The coastline passed beside us: scrub, a limestone shelf and patches of mangrove where the rock dipped close enough to the water. There were no buildings or markers, only the sort of empty restricted shore that told you the Ministry of Defence owned it and did not want you there.
I kept the rucksack between my feet and watched the water. Near the shore it was clear enough to see the bottom: sand, scattered rock and clumps of weed anchored in the shallows. The depth was perhaps four metres, five at most.
Nothing moved except the weed, and even that moved incorrectly. Every strand tilted in the same direction at the same time, synchronised with the swells.
After about ten minutes, Callum reduced the throttle. We drifted towards a low outcrop where the scrub thickened and the limestone rose above the tide line. He nosed the workboat in close, cut the engine about six metres from shore and allowed our momentum to carry us towards a level section of rock.
I caught the edge and steadied the boat while Callum tied the painter to a rusted cleat that someone had drilled into the limestone years before. The cleat held, though only just.
Callum climbed out with the canvas bag. I followed with my rucksack.
The rock was dry but deeply pitted, its surface covered in salt crust and bird droppings that crunched beneath my boots. I smelt dead seaweed and something faintly chemical underneath it, sharp enough to catch in the nose. Old fuel, perhaps, or treated discharge.
Callum walked inland without waiting. I followed.
About ten metres from the water, beyond tangled scrub and a low limestone ridge, a hatch lay flush with the rock. It was roughly a metre square, made of heavy steel and painted naval grey decades earlier. Most of the paint had vanished beneath rust and salt stains.
A stencil across the top read "MOD PROPERTY" in faded white. Beneath it was an older Royal Navy stores marking and a confined-space warning triangle bleached almost colourless by the sun.
The hatch had a recessed handle on one side and a hasp on the other. Its padlock had not gone missing or broken through corrosion. Someone had cut cleanly through the shackle with bolt cutters or a hacksaw. The two halves still hung through the hasp, as though whoever had opened it wanted the breach to be noticed.
I crouched and touched the cut surface. It was smooth, and recent enough that rust had not begun to bloom across the exposed steel.
"When was this last inspected?" I asked.
Callum stood over me, staring at the lock.
"Six months ago. Possibly longer."
"Was the lock intact?"
"It should have been."
I stood and searched the surrounding ground. There were no prints in the salt crust, though the limestone was too hard to retain much detail. I saw no rubbish, discarded tools or other disturbance. Only the cut lock.
"Someone staged this," I said.
Callum did not answer. He bent, pulled the severed padlock from the hasp and threw it into the scrub. Then he grasped the recessed handle and lifted.
The hatch opened with a metallic groan, its counterweight taking most of the load. It swung upright and locked into place, exposing a dark square and the top of a steel ladder descending into it.
The smell reached me at once: damp concrete, salt, old battery acid and something organic decaying slowly in darkness.
I took a head torch from my rucksack and switched it on. Callum was already wearing his.
"Stay close," he said, and started down.
The ladder descended about four metres into a short tunnel. The floor and walls were concrete. The low ceiling carried conduit along its length. My light picked out water stains and white mineral deposits where moisture had seeped through the walls.
Callum moved ahead without stopping. I followed a metre behind, sweeping my light over the walls and ceiling from habit.
The tunnel turned left after about six metres and opened into a wider chamber. It was an equipment room, perhaps two and a half metres by three, with just enough headroom to stand without brushing the overhead conduit.
Old and new systems occupied the same walls. Thick industrial-grey conduit from the 1970s analogue hydrophone network ran beside modern fibre-optic bundles clipped into plastic mounts. Water pipes sweated condensation. A junction box the size of a briefcase was bolted to the rear wall. Its cover was missing, exposing the components inside.
There were terminal strips, cable connectors and white-green corrosion across half the contacts.
The smell was stronger in the room: batteries, salt, wet concrete and something I could not identify. Mildew, perhaps, or algae growing somewhere beyond the reach of our lights.
Callum stopped in the centre and made a slow turn, using his head torch to inspect every corner. I did the same. There were no footprints, no rubbish and no obvious evidence that anyone had entered, apart from the hatch having been opened from outside.
"This connects to the undersea network?" I asked.
"It used to. The fibre runs are newer. The conduit belongs to the legacy system."
"Legacy what?"
"The old analogue hydrophone network. It was taken out of service when the range was upgraded, but nobody removed it."
"So it is still physically connected."
I moved closer to the open junction box and shone my light inside. Half the terminals were thick with corrosion. The others looked clean. That made no sense unless someone had maintained part of the system while deliberately neglecting the rest.
I photographed the junction box, then the conduit where it disappeared through the wall. Callum watched but made no attempt to stop me.
"We're documenting this now?" he asked.
"The site is compromised. Someone cut the lock. I'm documenting it."
He did not argue.
A doorway at the back of the room led into another tunnel. This one was narrower, perhaps a little over a metre across, with a lower ceiling. The old conduit followed the left wall and the fibre ran along the right.
Callum ducked through and continued. I followed.
About fifteen metres in, I heard the whistle again.
It was the same clean tone at roughly 60 to 70 Hz. There was no build-up and no fade. It simply appeared, filling the tunnel while seeming to come from everywhere and nowhere.
I stopped. Callum halted a metre ahead but did not turn.
The tone held for four seconds and cut off. The absence left my ears ringing.
"You heard that?" I asked.
"Yes."
"Where did it come from?"
"I don't know."
I turned, sweeping my light back towards the equipment room. There were concrete walls, conduit and shadows, but no loudspeaker and no machinery capable of producing the sound.
The tone returned. Same frequency, same four-second duration.
This time I felt it pressing into my chest like bass at a concert. It was body-felt low-frequency sound, close to the near-infrasonic boundary, but even at that frequency the physical pressure was wrong. It should not have seemed to bear directly against my sternum.
I stepped forwards. The tone did not become louder. I stepped back, and it did not diminish.
"That's impossible," I said.
Callum turned. His face was blank, his jaw set.
"What is?"
"The acoustics. This is a straight concrete tunnel. The signal should reflect. It should form standing waves and give us some sense of direction, but it isn't. The level remains constant wherever I stand."
"What does that mean?"
"It behaves like an omnidirectional source in open water, where there are no close boundaries. You do not get that inside a confined concrete passage."
Callum watched me for two seconds, then turned and continued walking.
The tone stopped again.
I followed because remaining alone felt worse.
The tunnel bent right, then left, before opening into another room. This chamber was larger, perhaps four and a half metres by three and a half. The ceiling was high enough that my head torch did not find it immediately.
Water covered the floor to ankle depth. It was cold and perfectly still. I stepped into it and felt it seep through my boots.
The smell was stronger there: salt, decay and something chemical that burnt the back of my throat.
My light found shapes mounted above the waterline. Racks of dead electronics covered the walls. Old grey boxes with dials and switches stood with cables hanging loose and faceplates streaked by salt and corrosion.
Kelp lay draped across the equipment. It hung from the cables and gathered in the upper corners where the walls met the ceiling. The brown strands were slick and fresh.
It was wrong. Kelp did not grow in sealed underground chambers, and it did not climb four metres up concrete walls.
I moved my light slowly across the nearest rack. Water dripped from the kelp in steady drops, falling into the still pool below.
Callum had crossed to the far side of the chamber, where a metal shelf projected from the wall, fixed with heavy industrial brackets. A lidded plastic crate sat on it.
He pulled the crate to the edge of the shelf and released the catches.
Inside was a ring binder sealed in a clear plastic document bag, the sort used aboard boats to keep paperwork dry. Callum pulled it out, tore open the bag and laid the binder on the shelf. I leant over and aimed my head torch at the pages.
They were maintenance sheets, printed forms filled in by hand, with dates going back three years. Equipment IDs, status codes, technicians' initials. Standard Royal Navy documentation.
Callum flicked through them. Halfway into the binder, the pages changed.
It was a printed memorandum. The header read DEFENCE INTELLIGENCE, NAVAL INTELLIGENCE DIVISION. The date was 14 May 2024. Classification markings had been obscured beneath black bars.
The subject line read: ACOUSTIC MIMIC - PROTOCOL STILL WATER.
I leant closer. The memorandum ran to two pages. It was written in bullet points, concise and clinical, in the sort of language that assumed you already knew half the context.
Do not respond to unknown acoustic signals originating from network nodes.
Do not follow or attempt to locate return signals.
Do not enter access tunnels alone.
Do not engage in communication with unverified sources via hydrophone or radio.
Report all anomalous activity to station command immediately.
Below that was a paragraph in smaller type.
Acoustic signature exhibits adaptive behaviour. Mimic has demonstrated the capacity to replicate human speech patterns, equipment signals and environmental sounds with increasing accuracy. Learning rate accelerates with exposure. Isolation protocols remain in effect until containment is confirmed.
I read it twice. Callum turned the page.
Clipped to the back of the memorandum was a sheet of lined paper bearing a hurried sentence in blue ink.
Mimic learns faster when you ignore it.
That was all.
I stood in ankle-deep water, staring at the note while my head torch made the wet kelp glisten on the walls. Then I turned to leave. Callum did the same.
We both directed our lights back towards the tunnel we had just crossed.
The corridor was dry.
Not drying. Not draining. Dry.
I stopped. Callum went another three paces before noticing, then turned back.
The water we had waded through less than a minute earlier was gone. The ankle-deep cold that had soaked my boots and slowed every step had simply vanished. The floor was damp concrete, darker in patches where moisture still clung, but there were no pools, no ripples and no standing water at all.
I looked down at my boots. They were still wet. Water dripped from the rubber on to the dry concrete.
"You see this?" I said.
Callum's head torch swept across the floor. He did not answer at once.
Where the water had been, there were footprints instead. Wet, bare footprints, narrow like those on the platform, with five elongated toes, abnormal jointing and a toe-heavy gait. The ball and outer edge of each foot had pressed dark against the damp concrete. They led away from us, deeper into the facility, following the corridor we had not yet taken.
I counted seven before the beam of my head torch could reach no farther.
"We're leaving," I said.
Callum stared at the prints. His jaw worked, but for three seconds nothing came out.
"We need to confirm Owen's kit first."
"No, we don't. That's why we're here."
I turned to face him squarely.
"The water moved, Callum. It was here and now it isn't. The floor rearranged itself while we were standing in it."
"I know."
"Then we leave. We go back up, call this in and get people who know what the hell this is."
Callum shook his head, slowly and deliberately.
"Can't leave without confirming the kit."
"Yes, you can. You just walk."
"The Royal Navy wants confirmation. Owen's equipment, serial numbers, evidence of what happened."
My chest tightened, not with panic but anger.
"You don't care about serial numbers. You care about what took him."
Callum did not deny it. He looked past me, down the corridor where the footprints led.
"The plans put the storage bay about eighteen metres that way," he said. "We confirm it, document it and leave."
"Or we leave now and nobody confirms anything."
"Your contract says-"
"My contract says validation. This isn't validation. This is containment, remember? And I'm not trained for containment."
Callum met my eyes. His head torch made his face look carved from stone, with shadows gathered beneath his brows and jaw.
"You're the only person here who understands the kit," he said. "You're the only one who can tell me whether what's down there is his, or whether it's something else wearing his equipment."
That stopped me. I did not want it to, but it did.
Because he was right. If something could mimic Dr Helen Mercer's voice, if it could copy a radio transmission, layer it wrongly and make it sound wet, it could probably stage a scene. It could assemble a pile of diving equipment that looked genuine but was not. And if Callum went in alone, he would not know the difference.
I looked at the footprints again. They did not waver or overlap. They formed a steady line leading deeper, as though someone had walked that way with purpose.
The place was rearranging what normal meant. I felt it in my stomach, in the way my mind kept trying to account for the dry floor and failing.
This was not a fault. It was active and intentional, and we were about to follow it precisely where it wanted us to go.
"Eighteen metres," I said. "We confirm it, then we get out."
Callum nodded once and started walking.
The footprints remained ahead of us all the way. Whenever my light reached the farthest print, there was another just beyond it, wet and clean as if it had been made seconds before.
Their spacing was wrong as well. The stride kept changing. Sometimes the prints were half a metre apart, sometimes more than a metre. There was no rhythm and no consistent gait, only the same elongated toes and unnatural forward loading.
Callum did not look at them. He kept his eyes ahead and one hand against the wall, moving steadily. I stayed a metre behind him and tried not to think about how quiet the tunnel had become.
There was no dripping and no vibration from above. Only our boots against the concrete and the sound of our breathing.
The corridor bent left and opened into a long room, perhaps twelve metres deep and four metres wide. The ceiling was higher there, three or four metres above us. There was no water on the floor, but the smell struck harder: salt, decay and something chemical that burnt the back of my throat.
The whistle began again.
It was the same low-frequency tone, somewhere between 60 and 70 Hz, but this time it did not stop after four seconds. It held, steady and continuous. I felt the near-infrasonic weight of it in my teeth, my ribs and the bones of my skull.
It was not loud. It was simply present, as though the air itself had acquired a frequency.
Callum stopped. I halted behind him.
"You hear that?" I said.
"Yes."
The wall began to vibrate, low and subtle. I pressed my palm flat against the concrete and felt a pulse travel through it at five-second intervals. The same timing as the swells outside.
I snatched my hand away.
Callum's head torch swept across the room. The left wall held conduit and junction boxes, some open and others sealed. Empty equipment racks lined the right, with rust marks where apparatus had once been bolted in place.
At the far end stood a metal door with a watertight seal around its edges. Set at head height was a thick, scratched porthole perhaps twenty centimetres across.
Something moved behind the glass.
I stepped towards it before I could stop myself. Callum caught my shoulder, but I shrugged him off and continued, my head torch fixed on the porthole.
Seawater moved beyond it, swirling slowly as if a current flowed through open ocean. Kelp, or something dark resembling it, tumbled past the aperture every few seconds.
I pressed my ear against the door. There was no pump hum, no flow noise and no sound of water under pressure.
Nothing.
"There's no infrastructure for this," I said.
Callum had come alongside me. He stared through the glass.
"For what?"
"Moving seawater. There's no intake, discharge or pump system. The door is sealed, but that looks like open water on the other side, and I can't hear anything driving it."
"Could be passive. Tide-driven."
"We're nearly four metres underground. The tide doesn't work like that."
The dark material rolled past again. This time I saw that it was too pale and too smooth to be kelp.
I stepped back.
Callum reached for the door handle.
"Don't," I said.
He stopped and looked at me.
"The storage bay is through here."
"You don't know that."
"The plans match. This is it."
"Then we radio it in. We don't open it."
Callum pulled the radio from his belt and pressed the transmit key. Static burst from the speaker, then cut to silence. He tried again. Nothing.
He clipped it back on his belt and turned the handle.
The door opened inwards.
No water came through. The seal broke with a hiss, and damp air rolled over us, colder than the tunnel and carrying the smell of decay more strongly.
Callum's head torch entered first. I followed half a pace behind.
The bay was smaller than I had expected, about three metres square, with a concrete floor slick with moisture. Its walls were bare except for one shelf unit on the left.
The porthole we had viewed from the corridor was now in the far wall. It still showed moving seawater beyond the glass, but there was no tank, conduit or chamber behind it. Nothing that could contain the water at all. It was an impossible aperture, an apparent view straight into open ocean from a buried concrete room.
I avoided looking directly at it.
In the middle of the floor lay a heap of diving kit: black rubber fins, a mask with a cracked lens, a dive knife in a nylon sheath still strapped as though it had been worn against a diver's calf, and a helmet standing upright, filled with water.
Callum crossed the room in four strides and crouched beside the pile. I remained in the doorway, sweeping the corners with my light and checking for anything else.
The helmet was a faded orange Royal Navy surface-supplied model with an integral communications unit. Its faceplate was held in place by a metal retaining ring. Water filled it to the brim and remained perfectly still, without the slightest slosh. The inner surface of the faceplate was fogged, and the condensation looked fresh, as though someone had been breathing inside it moments earlier.
I moved closer.
Callum was examining the knife sheath for identification markings.
"Don't touch the helmet," I said.
"I need the serial number."
"Don't touch it."
He looked up at me.
"Why?"
I pointed at the faceplate.
"That's fresh condensation. Someone breathed inside that thing within the last ten minutes."
Callum stared at it. Then he leant closer, directing his head torch straight at the fogged glass.
We both watched.
The fog began to clear. It did not evaporate or fade gradually. It vanished in patches, as if something inside the water were wiping the glass clean.
Within five seconds the faceplate was clear. The helmet remained full of water, but the glass was spotless.
I reached towards it, then stopped with my hand fifteen centimetres away.
In the faceplate's reflection I saw my head torch and my outstretched hand. Behind me stood another figure, close enough to be directly over my shoulder.
I spun round.
My light swung wildly across the room. Nothing was there. Only the doorway and the empty tunnel beyond.
I looked back at the helmet.
The reflection showed me, Callum crouching beside me, and the figure still standing behind my shoulder where no one stood.
"Callum."
He did not answer. He was staring at the helmet as well, his face blank.
A sharp metallic crackle came from everywhere and nowhere, like a radio being keyed, though I could see Callum's set still clipped to his belt and the transmit button untouched.
Then a voice spoke. Calm, even and familiar.
Helen's voice.
"Martin, where's the binder?"
It was not a question. It was a statement delivered in her cadence and tone, in precisely the way she asked for documentation when I forgot to file it.
But it was wrong. Two voices had been layered together, fractionally out of time, with that wet quality beneath them, as if they were speaking through water.
I backed towards the door. Callum stood quickly, one hand moving to his radio.
The voice came again with the same words and cadence.
"Martin, where's the binder?"
It was not asking. It was testing. Proving that it could copy her. Proving that it knew my name. Proving that it could wear familiarity as easily as it had worn Owen's breath on the inside of the helmet.
I looked at Callum. His face had gone white.
"We're leaving," I said. "Now."
Callum stepped towards the helmet, reaching for it. He began talking about Owen. He said the Royal Navy needed confirmation. He said Owen's family deserved closure. He said we could not leave without proving whether it was genuine kit or staged equipment.
I was no longer listening.
Beyond the porthole, the water bulged outwards. The glass did not break or crack. It simply bowed into a smooth dome, as if something immensely heavy were leaning against it from the other side.
The pale material that had been tumbling past was gone. There was only dark water and that swelling pressure against the glass.
Then a shape emerged.
Pale. Jointed. Pressing itself flat against the porthole.
I tried to make sense of it. I tried to see an arm, a leg or anything human, but its joints bent in the wrong directions. There were too many of them. The segments were too long and too narrow, and the whole thing tapered in a way that matched no limb I had ever seen.
The whistle stopped.
The clicks began.
They were rapid, sharp and metallic, following the same rhythm as the test sequence I had transmitted from the jetty the previous night.
Three short pulses. Pause. Three more. Pause. Repeat.
My own sequence was being played back to me.
I seized the back of Callum's buoyancy vest and dragged him away. He stumbled and nearly fell, but I did not release him. I hauled him towards the door while he continued trying to argue.
With my free hand, I pulled the portable acoustic test set from my rucksack. It was still powered from our earlier checks. I switched it from receive to transmit, selected the wideband calibration-noise mode I had demonstrated when unpacking the unit, and turned the gain to maximum.
In open water I would have used the hydrophone as the projector. Inside the chamber, I selected the unit's airborne calibration speaker and pressed its contact transducer against the metal doorframe to couple the signal into the structure.
The mode was intended for interference testing. It flooded a broad frequency range with random noise so that a system's response could be measured. I had used it twice before on range work to overwhelm background signals during controlled tests.
I triggered it.
The sound was ugly. White noise tore through the room and the structure across every frequency the unit could produce, loud enough for me to feel it in my chest and jaw. Callum clamped both hands over his ears and shouted something I could not hear.
The whistle cut off. The clicking stopped. The vibration in the walls ceased.
For perhaps two seconds, the entire facility was silent.
Then it answered.
The same wideband noise came back, but cleaner and sharper, without distortion or static, and it was aimed.
I have no better way to describe it. Sound does not behave like that, not without a directional array or focused emitter. But the noise struck us head-on like a physical blow, a narrow beam with all its energy concentrated along a single line.
My ears rang. A high-pitched whine drowned out everything else.
I dropped the test set. It clattered on the wet floor and remained there.
Callum collapsed. He landed hard on his knees, pitched forwards on to his hands and convulsed once. Then he vomited, thin bile splashing across the concrete.
I caught his vest again and tried to pull him upright. He was dead weight. His legs would not work properly.
The water returned.
It did not flow from anywhere. It did not seep through cracks or spill from a pipe. It simply appeared, ankle-deep across the entire floor within three seconds.
It was so cold that it burnt through my boots.
I got Callum halfway upright. He tried to help, but his coordination had gone. He kept listing to one side.
Something splashed behind us.
I did not look back. I pushed Callum into the corridor and followed him.
The ringing in my ears was so severe that I could not hear our footsteps, the water or anything beyond the high-pitched whine drilling into my skull. Callum stumbled every few paces, but he kept moving. I stayed behind him with one hand on his shoulder, steering him whenever he drifted towards a wall.
The corridor bent left, then right. I recognised the route from our way in, but everything looked different. The walls were darker and wetter, with condensation running down them in thick streams.
We reached the equipment room, and I directed my head torch towards the ladder. It was still there. The hatch above remained open, with daylight pouring through it.
Callum reached the ladder first and gripped the rungs. He dragged himself up two steps, then stopped and hung there, breathing hard.
I climbed behind him and pushed. He began moving again.
Halfway up, my ears popped. The ringing subsided to a dull buzz and my hearing returned.
I heard Callum's breathing, my own, water dripping somewhere below, and the whistle. It was faint, but rising.
Callum cleared the top of the ladder and crawled forwards on to the rock. I followed quickly.
Three rungs from the top, I looked up and saw daylight, open sky and a silhouette leaning over the hatch.
It had a human shape, broad shoulders and a rescue vest. One arm reached down towards me.
"Martin."
The voice was calm and familiar.
"Take my hand. Helen sent me. She said you'd probably try to do the site check alone again. That's you."
That detail.
Helen had said those exact words after a previous contract, when I had gone out alone to validate a shallow hydrophone array instead of waiting for the boat crew. She had been annoyed but not surprised. She had said it was typical of me.
No one else should have known the story.
I stared at the hand reaching down. The fingers were too long. Not by much, perhaps two or three centimetres, but enough for my mind to catch the discrepancy and refuse to release it.
The skin was pale and waterlogged, like flesh that had been submerged for hours.
I released the ladder.
I dropped two rungs and struck the tunnel floor hard. The impact jarred both ankles and sent me down on one knee.
Callum was three metres away, crawling towards the scrub outside the hatch. The silhouette above shifted and leant farther into the opening.
It flowed.
That is the only word I have for it. It did not climb down or drop. It poured itself through the hatch like water finding a drain.
The whistle surged, so loud that I felt it in my teeth and jawbone.
I grabbed Callum by the vest again and dragged him backwards, away from the ladder and into the equipment room. He was trying to speak, but I could not distinguish the words over the whistle.
I swept the walls with my head torch. Conduit, junction boxes, fibre bundles and a small door recessed into the back corner behind a rusted equipment rack. It was less than a metre high and painted the same grey as the walls, making it almost invisible.
An emergency bulkhead.
It had to be.
I shoved the rack aside. Its feet scraped across the concrete before it toppled with a crash I could barely hear. The door had a wheel handle. I seized it and turned.
At first it was jammed. Then it broke free.
I hauled the door open and shoved Callum through. He went in head first. I followed, pulled the door shut behind us and struggled with the wheel on the inside.
It was stiff. My palms were slippery, and I lost my grip twice before I finally turned it far enough to seat the seal.
The whistle fell to a muffled vibration.
I turned round.
We were in a maintenance crawlspace, perhaps sixty centimetres high, with rust-lined walls and ceiling. Condensation dripped everywhere. The air smelt of iron, salt and decay.
Callum lay on his stomach a metre ahead, breathing hard. His head torch was still on, shining along the crawlspace towards a metal grille about six metres away. Daylight showed through its slats.
I crawled forwards on my knees and palms. Rust flaked from the walls wherever I brushed them and clung to my jacket. Callum started moving too, slowly but steadily.
Three metres from the grille, I heard the whistle again, fainter now and coming from behind the bulkhead door. With less than two metres to go, I could see sand, scrub and open sky beyond the slats.
The grille was secured by four bolts, all rusted solid. I braced my back against the low ceiling and kicked.
The whole frame rattled but held.
I kicked again. Pain shot through my heel and up my leg.
On the third kick, one bolt sheared. The grille shifted.
The fourth kick snapped another bolt. The frame bent outwards.
I kicked twice more and it tore free, crashing down on to the sand outside.
I crawled through, first my hands, then my knees. I stood and turned to help Callum. He was halfway out. I caught his arms and pulled until he cleared the opening, then he collapsed on the sand beside me.
My hands were bleeding. The skin across both palms had torn open, with rust and concrete grit embedded in the wounds. They did not hurt yet. They merely looked bad.
The sand shifted beneath us.
It was not a tremor or a vibration, but a faint displacement, as though something heavy had moved underneath.
I looked around. We were on the eastern side of the outcrop, perhaps twelve metres from where we had tied up the workboat. The sea was visible through the scrub. Its surface was flat. The unnatural swells had stopped.
I hauled Callum upright. He was steadier now and managed to get his legs beneath him. We stumbled through the scrub towards the limestone ridge.
I kept looking back, but nothing emerged from the open grille. There was only darkness and the faint, muffled whistle.
We crossed the ridge and reached the rock shelf where we had landed. The five-metre rigid-hulled workboat remained tied to the cleat exactly as we had left it.
I fumbled with the painter while Callum climbed aboard. My hands were shaking and the rope kept slipping through my bloodied fingers. It took me three attempts to release it.
I pushed the workboat clear and climbed in after him. Callum was already at the tiller. He pulled the starter cord once, and the outboard motor caught without a sputter or cough.
He turned us north and opened the throttle.
I sat in the bow and watched the outcrop diminish behind us. The grille remained visible as a dark square in the limestone. Nothing moved near it.
My ears were still ringing, more faintly now but without interruption.
We ran for about fifteen minutes. Callum said nothing. He kept one hand on the tiller and his gaze fixed ahead.
The main jetty appeared in front of us, followed by the antenna arrays, water tower and prefabricated buildings. A white Land Rover Defender bearing the BATEC crest was travelling along the station access track.
Callum angled towards the jetty and cut the engine six metres out, allowing our momentum to carry us alongside.
The Defender stopped as we secured the painter. The driver's door opened and a stocky older man in overalls climbed out, a radio clipped to his chest.
It was Graham Hales. I had seen him in the station mess the previous day, though we had not spoken.
He looked at us. He looked at Callum leaning against a jetty post, then at the blood dripping from my hands on to the planks.
"You two need the medical centre," he said.
I nodded.
Graham pressed his radio transmit key and called for a Royal Navy medical response. He did not ask what had happened or where we had been. He simply told us to get into the Defender.
Callum climbed into the passenger seat. I sat in the back. Graham drove us to the station medical centre at the southern end of the establishment.
Leading Medical Assistant Priya Nair met us outside. She was in her thirties, with immaculate working dress and her hair pulled back tightly. She already had a first-aid kit in her hands.
She took us into a small examination room containing two chairs, a table and a harsh fluorescent light.
Priya began with Callum. She checked his pupils and pulse, then asked him to follow her finger with his eyes. He answered her questions in one or two words.
"No."
"Yes."
"Don't know."
Then she came to me. She had me sit down and place my hands on the table. She cleaned the cuts with saline, removed the grit and fragments of rust with tweezers, then wrapped both palms in gauze and medical tape.
"You'll need a tetanus jab," she said.
I told her that was fine. She gave me the injection and recorded it on a clipboard. Then she asked what had happened.
I looked at Callum. He was staring at the floor.
"Equipment failure," I said. "We had to abandon a site inspection and leave through a maintenance crawlspace."
She wrote it down without pressing for details.
Callum remained with me for perhaps an hour afterwards. We said little. He kept his radio switched off and his mobile in his pocket.
At one point, he said he would file a formal report and include everything: the whistle, the footprints, the helmet and the voice.
I asked whether he wanted my photographs. He said he did.
Using the station's secure incident-report system, I transferred the images to him: the junction box, the conduit, both photographs of the malformed footprints on the platform, and the helmet with its fogged faceplate.
He thanked me. Then he said he had another binder containing documentation on Owen Price: his personnel record, diving logs and last known location. Owen had disappeared on 18 September 2025 while maintaining the South Range array. The four-day air-and-sea search had recovered only his cylinder, fins and surface marker near the disused shore repeater chamber.
Callum said he had been ordered to keep the binder secure and surrender it only if we recovered something.
I asked whether he intended to hand it over.
"No," he said. "I'm keeping it. Evidence."
I did not argue.
My mobile rang at about 17:00. It was Dr Helen Mercer.
I went outside to answer.
"Martin."
Her voice was clear, calm and normal.
"I've just had an alert from BATEC Safety. Are you all right?"
I told her I was fine. I had cuts on my hands, but nothing serious.
"What happened?"
I gave her the short version: a site inspection at a shore access point, equipment failure and an emergency exit through a maintenance tunnel.
She was silent for several seconds.
"I didn't approve any shore work."
I stopped walking.
"What?"
"You were contracted for offshore validation. North Range array, six nodes. I received the schedule from Chief Petty Officer Callum Reed last week. There weren't any shore access points on it."
I asked whether she had authorised confirm-and-contain work.
"What's confirm and contain?"
I explained. Callum had said the Royal Navy wanted us to confirm what was inside the chamber and contain it. He had said it was connected to a missing diver.
Helen was quiet again, this time for longer.
"Martin, I don't have any contacts in Defence Intelligence. This contract came through normal BATEC channels. Equipment validation. That's all."
I felt cold.
"Callum said you approved it."
"I approved offshore work, hydrophone nodes and serial checks. Nothing involving tunnels or containment."
I looked back towards the medical centre. Callum was still inside.
I asked Helen whether she had ever told anyone about my habit of carrying out site checks alone.
"What?"
I repeated the question. She considered it for a moment.
"Yes, it's come up. You have a habit of ignoring procedure."
"Who did you tell?"
"Martin, what's this about?"
I did not answer.
She said she had mentioned it in a couple of ordinary post-contract reviews, perhaps in one or two emails. Standard supervisory notes on the contractor network. Nothing classified.
I thanked her and said I would ring when I was back in Britain.
She told me to have my hands examined properly and to submit an incident report. I said I would.
She ended the call.
I remained outside the medical centre for several minutes. My hands had begun to throb, and blood was already soaking through the gauze in two places.
When I went back inside, Callum was gone.
Priya said he had left ten minutes earlier to file his report. His binder went with him.
I asked if I was free to leave. She said I was, but told me to change the dressings twice a day and return if I saw any sign of infection.
I left.
I flew out of Providenciales two days later, on 13 March 2026, with both hands bandaged and a faint ringing still in my ears.
Helen had withdrawn me from the contract. She said BATEC was dealing with the matter internally and no longer required civilian support. I did not argue.
I continued contracting. Over the following seven months, I completed two North Sea survey contracts and one Atlantic cable-inspection job. All were offshore. All were ordinary validation work.
But I stopped running test tones at normal volume. I kept the gain low. I logged every signal I transmitted and made certain I knew what I was pinging before I sent anything into the water. I used replacement equipment. The portable test set I had dropped remained inside the chamber.
At night, back in my flat in Britain, I kept hearing the whistle.
A thin, low tone between 60 and 70 Hz, coming from the kitchen sink.
I would walk towards it, but it always seemed to remain the same distance away. Constant. Directionless. Unreachable.
I turned the taps on and off. I checked the pipes. I put my ear against the drain.
Nothing changed.
The whistle remained, always present and always just beyond reach.