The Repository Abyss Report: Into the Dark

Mid-Atlantic Ridge Holes: The Digger No Camera Has Seen

October 16, 2026 · Runtime 23:19 · Watch on YouTube

Rows of evenly spaced holes have been photographed on the Mid-Atlantic Ridge since the 1960s. In July 2022 a NOAA robot found them twice in five days, sucked one up, and bottled the water beside them. The same week, a ship in the Bering Sea filmed similar rows with an amphipod in the doorway. What digs the Atlantic lines is still undetermined, and here is exactly what it would take to find out.

Transcript

A line of holes 0:00

Two and a half kilometres down, on the Mid-Atlantic Ridge, there is a line of holes in the mud. Small, evenly spaced, perfectly in a row, each with a little pile of sediment beside it. They have been photographed here since the nineteen sixties. Something is digging them. In sixty years, no camera has caught the digger.

So here is the question this episode follows. What makes the holes in a line? In July twenty twenty-two, a NOAA robot found them twice in five days, and the scientists on the live call threw everything they had at the question. They sucked up a hole. They bottled the water beside it. And the same week, on the other side of the planet, another ship filmed something very similar, with an animal standing in the doorway. Every moving picture in this episode is NOAA footage from those dives. Every number comes from the dive record, a published paper, or our own reading of NOAA's public data.

Dive 04, Axial Volcanic Ridge 1:02

The first dive was on the twenty-third of July, twenty twenty-two. The ship was Okeanos Explorer, the robot was Deep Discoverer, and the target was a young volcanic ridge running down the middle of the rift valley, north of the Azores, at forty-three degrees north. The robot touched bottom at twelve thirty-five, universal time, two thousand seven hundred and twelve metres down, and spent four and a half hours driving uphill. Near the top it would find dead hydrothermal chimneys. Before that, it found tracks.

An acorn worm and its trail 1:37

Early on the slope, the camera found a long, soft animal lying on the mud. It is an acorn worm, from a deep-sea family whose best-known genus is named Yoda, for the two big lobes on its head. Its first species was described from this same ridge in twenty twelve. Acorn worms eat their way across the sea floor and leave a trail of their own droppings behind them. On the deep sea floor, that trail is often a perfect spiral.

How a trace gets its author 2:06

For a long time, those spirals were a mystery too. They were seen in photographs, and even in fossil rock, and nobody knew who made them. The answer came from patience, not from a sample. In the Pacific, at a site called Station M, a time-lapse camera sat on the sea floor for months. One sequence showed an acorn worm arriving, spending thirty-nine hours laying down a four-turn spiral, and leaving. That was published in twenty oh five. Keep that idea. In the deep sea, a trace gets its author when somebody catches the author in the act.

Paleodictyon, still unsigned 2:44

Not every trace on this ridge has found its author. In two thousand three, the submersible Alvin dived into the rift valley of the Mid-Atlantic Ridge at twenty-six degrees north, about three and a half kilometres down, looking for a pattern known from fossils. Tiny holes, one millimetre wide, set out in perfect hexagons a few centimetres across. It is called Paleodictyon, and it turns up in rocks hundreds of millions of years old. Alvin brought back intact cores of the living version. Peter Rona and twelve colleagues found the tunnels under the holes. They found no living tissue in them, and they could not say whether it was a burrow or the body of some unknown organism. Published in twenty oh nine. Still unsigned.

The rows appear 3:32

At about half past four in the afternoon, two thousand five hundred and forty metres down, the camera passed over something else. A pale stripe on the mud, and along it a row of small dark holes, one after another. Then another row, and another. This is what the room said.

Here is what is actually on the screen. The holes are not round. They are short slits, and the long side always points along the line. Each line sits on a low ridge of paler sediment, mud that looks dug out from underneath. Some lines run straight, some curve gently, and some cross each other. There are no footprints between the holes, no drag marks, no body. Just the holes, and the dug-out mud.

Photographed since the 1960s 5:57

This is not a new sighting. Bottom photographs from the nineteen sixties, taken around the Azores, already show rows of small craters. In nineteen seventy-one, Bruce Heezen and Charles Hollister put them in their atlas of the deep-sea floor, from depths between fourteen hundred and seventeen hundred metres. They had the pictures. They did not have an explanation.

2004: G.O. Sars 6:22

Then, on the thirteenth of July, two thousand four, a Norwegian ship, the G O Sars, sent a robot down north of the Azores, about two thousand and eighty metres deep. For more than an hour, it drove across lines of holes. One of the scientists on that cruise was Mike Vecchione, a zoologist with NOAA and the Smithsonian. He and the Norwegian biologist Odd Aksel Bergstad wrote down what they saw. Each hole about six centimetres long and one and a half wide. The gap between holes about the same as a hole. Lines from less than a metre to many metres long.

Their first guess was a large crustacean, a blind deep-sea lobster. They wrote that there was no evidence for it, and dropped it. None of their close-ups showed anything living in a hole. They could not tell whether the holes were connected underground. Their paper came out in January twenty twenty-two, and its title says it plainly. These holes point to knowledge gaps. Six months later, the NOAA robot landed on them.

Dive 09, Cachalote Guyot 7:30

Five days after the first sighting, on the twenty-eighth of July, the robot went down on Cachalote Guyot, a flat-topped seamount further south and west, at thirty-nine degrees north. It reached the bottom at eleven thirty-three, about fifteen hundred metres deep, and within minutes the camera was looking at lines of holes again. This time the team was ready, and so was somebody on the phone.

Vecchione calls the ship 8:45

Mike Vecchione had been following the dive from shore. Eighteen years after his own cruise, he could not stay in the chat room. He called the ship.

Aliens, gas, a crab 9:48

When NOAA posted the first pictures, the public sent in explanations. Aliens. An unknown crab. Gas rising from below the sea floor. On the call, one of the geologists tried a geological version, half joking. A crack in the rock below, fluid flowing up through it, and the holes spaced at the rhythm of the flow. The answer was about evidence. Here is how it went.

The pogo stick 11:15

Another idea that keeps coming back is the pogo stick. A drifting instrument, a float, bouncing along the bottom and punching a hole at every bounce. Bell and colleagues, describing traces on this ridge in twenty thirteen, used that nickname, pogo stick trails. On the call, the scientists took it apart.

Measured with the lasers 12:13

We measured one line ourselves. The robot carries two lasers, ten centimetres apart. In this NOAA photograph from the second dive, the two red dots sit right next to a fresh line. At that distance, ten centimetres is about a hundred and fifty-five pixels. The line has six holes in about twenty-six centimetres. That is six or seven centimetres from hole to hole, on a pale strip about eight centimetres wide. The camera is tilted, so call it plus or minus twenty percent. It matches the spacing written down in two thousand four, on a different seamount, eighteen years earlier. Whatever does this works to the same rhythm.

A fresh end and an old end 12:56

And the lines have a direction. At one end the holes are sharp-edged and the dug-out mud is clean. At the other end, the holes are slumping and half filled in. If that is true, the line is a record of time. Something started at the old end and moved towards the fresh end. The scientists noticed the same thing.

Sucking up a hole 13:53

So they went for the fresh end. At seven minutes past twelve, the robot lowered the nozzle of its suction sampler into a sharp hole, fifteen hundred metres down, and pulled. The mud was so fine that it went up like smoke, and filled a whole canister in one pull. Then they waited for the cloud to clear, hoping something would show itself.

Nothing did. The sample went into the record as EX twenty-two oh five, dive nine, number three G. Its label is two words. Mystery hole. The expedition's own inventory lists what came up with it: a sponge, and a polychaete worm. Neither has ever been linked to holes like these. NOAA's summary was that the sediment revealed no clues. The animal was not at home, or the nozzle missed it, or it was never in the hole at all.

Why holes matter 15:24

Why should anyone care about holes in the mud? Because on a sea floor this quiet, a hole is a lot. Vecchione and Bergstad pointed out that thousands of them change how far oxygen-rich water gets into the sediment, which decides what can live there. Old holes fill with settling plankton, little pockets of food for the animals that graze the surface. Whatever digs these lines is shaping its own patch of ridge. On the call, one voice put it simply.

A cast nobody had 16:18

On land, there is a classic way to see a burrow you cannot see. You pour something in, let it set, and dig up the cast. People have made casts of ant nests and of shrimp burrows on beaches. A cast would answer the first question at once. Are the holes in a line connected by a tunnel? On the call, someone asked whether the robot had anything like that on board.

The dive moved on, across corals and sponges, for four more hours. Then, a little before five in the afternoon, at fourteen hundred and forty metres, the holes came back. This was the fourth sighting of the day, and the clearest. Long lines, some crossing, a fresh one with the lasers right beside it. The robot stopped, and the team tried the last thing they had.

A bottle of water 17:43

At four fifty-six and fifty-six seconds, the robot closed a water bottle just above the holes. One point seven litres, at four point nine degrees. The idea is environmental DNA. Every animal sheds skin, mucus and waste, and the water around it carries traces of its genes. Filter the water, sequence what is caught, and you get a list of who has been nearby, without ever seeing them. The bottle was labelled, in NOAA's own words, eDNA, mystery holes, near bottom.

What the DNA says 18:16

In twenty twenty-two, the scientists said the results would take weeks. We could not find them published anywhere as an answer. But NOAA and the Smithsonian have since released the raw results for the whole expedition, sample by sample, as open data. So we added them up. In the main marker, that bottle held about a hundred and sixty thousand DNA reads. Seventy-seven and a half percent were dinoflagellates, single-celled plankton. Nineteen percent were tiny copepods. Polychaete worms, eighty reads. Amphipods, zero.

That zero sounds like an answer. It is not. Across all forty-nine water samples from this expedition, amphipod DNA turned up in only two, and neither of them was near the holes. In this survey, the method almost never sees amphipods at all. And the bottle sampled the water above the mud, which is mostly drifting plankton. Whatever digs these holes lives in the mud. The water test answered a different question. Here is why amphipods are the name to check.

The same week, the Bering Sea 19:25

That same month, July twenty twenty-two, the German research ship Sonne was nine thousand kilometres away, in the Bering Sea, towing a camera over abyssal mud about three and a half kilometres deep. The team was counting animals. Then, in the words of one of them, they suddenly saw a lot of holes in the sea floor, in rows, one after the other. In the end they counted one hundred and ninety-six, in about five thousand six hundred square metres.

And on the thirtieth of July, three thousand five hundred and ninety metres down, the camera caught one row with something at the end of it. An amphipod, a shrimp-like crustacean with big digging claws, standing in the newest opening, apparently carrying mud. A similar amphipod was caught by sledge nearby, probably a species nobody has described. One of the authors remembered an experiment from the nineteen eighties. An Antarctic relative, kept in a cold tank, used its big claws not to hunt but to shovel sediment. Angelika Brandt and colleagues published the Bering Sea burrows in twenty twenty-three, with a plausible maker.

Not the same holes 20:37

So is that the answer for the Atlantic? Their own paper says no. The Bering Sea rows are short, around thirty-four centimetres, always with a clear entrance and a fresh pile at the working end, and the openings are joined by one tunnel underneath. The Atlantic lines can run for metres, nobody has seen a tunnel joining them, and nothing has been seen inside. The authors call the Pacific structures a different thing, probably with a different maker. Their words about the Atlantic ones are three: remains undetermined.

So, what would it take? On the call, the scientists already knew part of the answer. It is time.

What would close it 21:59

Here is the list, and none of it is exotic. First, a camera. A lander left at the fresh end of one line, taking a picture every hour, the method that found the author of the spirals. Second, a cast or a core. A cylinder of mud pulled straight up through a fresh hole, kept intact, and sieved on deck, to see if there is a tunnel, and who is in it. Third, the DNA of the mud itself, not of the water above it. Each one fits on an ordinary dive. On the twenty-eighth of July, twenty twenty-two, there was no latex on board, and the only sample they took came up without an answer. Sixty years after the first photographs, that is still exactly what is missing.

Next: the Upsweep 22:45

Next time, a different kind of mystery. Since August nineteen ninety-one, NOAA's underwater microphones in the Pacific have been picking up a sound. A train of tones, each one rising in pitch for a few seconds, loud enough to cross an ocean, coming from somewhere far south, near fifty-four degrees south. It peaks in spring and autumn, and it has been getting quieter for thirty years. Its source has never been confirmed. That is the next Abyss Report. The Upsweep.

Description and sources

EX2205 dive 04 found sets of holes at about 2,540 m on Axial Volcanic Ridge 1 (23 July 2022); dive 09 found them four times at 1,440-1,500 m on Cachalote Guyot (28 July). Mike Vecchione, who described them from a 2004 cruise with Odd Aksel Bergstad, called the ship live. Sample 03G (12:07 UTC, 1,499 m) brought up a sponge and a polychaete, no answer. Our sum of NOAA's public eDNA results for bottle 09W, taken beside the holes: 77.5% dinoflagellates, 18.9% copepods, zero amphipod reads, in a survey where amphipod DNA appeared in 2 of 49 samples. Against the robot's 10 cm lasers, one fresh line has six holes in about 26 cm. In the Bering Sea, R/V Sonne proposed maerid amphipods for similar burrows (Brandt et al. 2023); the Atlantic maker remains undetermined.

PRINT-READY, FROM THIS CHANNEL

Read This Dive, Vol. 01 — the method, and five dives read with it

https://therepository.gumroad.com/l/read-this-dive

PRIMARY SOURCES

NOAA Ocean Exploration, The Case of the Mysterious Holes on the Seafloor (27 July 2022): https://oceanexplorer.noaa.gov/news/oer-updates/2022/mysterious-holes-seafloor/mysterious-holes-seafloor.html
NOAA Ocean Exploration, EX-22-05 expedition report, dive table and sample inventory (03G 'Mystery hole suction'; 09W eDNA): https://repository.library.noaa.gov/view/noaa/48487/noaa_48487_DS1.pdf
NOAA Ocean Exploration, EX2205 dive video, track and ancillary data: https://www.ncei.noaa.gov/waf/okeanos-rov-cruises/ex2205/
NOAA Ocean Exploration, seawater eDNA metabarcoding (EX2205), OBIS; sample EX2205_D09_09W, NCBI SAMN54143495
Vecchione, M. & Bergstad, O.A. (2022). Numerous sublinear sets of holes in sediment on the northern Mid-Atlantic Ridge point to knowledge gaps in understanding mid-ocean ridge ecosystems. Frontiers in Marine Science 9: 812915. doi:10.3389/fmars.2022.812915
Brandt, A., Chen, C., Tandberg, A.H.S., Miguez-Salas, O. & Sigwart, J.D. (2023). Complex sublinear burrows in the deep sea may be constructed by amphipods. Ecology and Evolution 13: e9867. doi:10.1002/ece3.9867
Smith, K.L., Holland, N.D. & Ruhl, H.A. (2005). Enteropneust production of spiral fecal trails on the deep-sea floor observed with time-lapse photography. Deep-Sea Research I 52: 1228-1240. doi:10.1016/j.dsr.2005.02.004
Priede, I.G. et al. (2012). Torquaratorid acorn worms from the North Atlantic, with a new genus and three new species. Invertebrate Biology 131(3).
Rona, P.A., Seilacher, A., de Vargas, C. et al. (2009). Paleodictyon nodosum: a living fossil on the deep-sea floor. Deep-Sea Research II 56: 1700-1712. doi:10.1016/j.dsr2.2009.05.015
Bell, J.B., Jones, D.O.B. & Alt, C.H.S. (2013). Lebensspuren of the bathyal Mid-Atlantic Ridge. Deep-Sea Research II 98: 341-351.
Heezen, B.C. & Hollister, C.D. (1971). The Face of the Deep. Oxford University Press.
Next episode: NOAA PMEL Acoustics Monitoring Program, Upsweep: https://www.pmel.noaa.gov/acoustics/sounds/upsweep.html

Abyss Report reads the deep ocean through the record that resolved it -- instrument, date, position and the primary sources above.

About the material. Every moving frame is NOAA Ocean Exploration video, public domain, credited on screen with dive, depth and position: EX2205 dives 04 and 09, Mid-Atlantic Ridge, 23 and 28 July 2022. Depths come from the vehicle's 1 Hz track. Several cuts drop the narration and play the original control room audio; each window was re-transcribed in isolation twice and used only where both passes matched. The eDNA percentages are our own sum of NOAA Ocean Exploration's public metabarcoding results (OBIS dataset 'NOAA Ocean Exploration seawater eDNA metabarcoding (EX2205)', sample EX2205_D09_09W). The hole spacing is our own measurement against the robot's 10 cm lasers in a NOAA still. Every diagram is drawn in code from the sources listed.

#DeepSea #NOAA #OceanExploration #DeepSeaMystery #MidAtlanticRidge