The Lighthouse Keeper’s Last Entry

The Flannan Isles mystery and the three men who simply ceased to exist

Here is the story as most people know it.

On the 26th of December 1900, a relief vessel called the Hesperus made its way through heavy seas towards a small cluster of uninhabited rocks twenty miles west of the Outer Hebrides. As it approached the Flannan Isles — known locally as the Seven Hunters, a name that carries its own kind of warning — the crew noticed something wrong. The lighthouse showed no light. The flagpole was bare.

Three men had been stationed on the lighthouse: James Ducat, the principal keeper, Thomas Marshall, his assistant, and Donald McArthur, a replacement keeper covering for a sick colleague. When the Hesperus landed a party ashore, they found the lighthouse empty, the beds unmade, a half-eaten meal still on the table, a chair overturned. One set of oilskins still hung in the hallway, as though its owner had stepped out for a moment and not come back.

No bodies. No sign of struggle. No note of farewell.

A place already strange

The Flannan Isles had a reputation long before the lighthouse was built.

The islands are small and bleak and extraordinarily remote. Local fishermen had traditionally avoided landing on them except in emergencies, believing the islands to be inhabited by spirits. Even those who did land observed a set of strict customs such as removing their hats, speaking in hushed voices, taking nothing from the islands; as though the rocks required a particular kind of careful courtesy.

The lighthouse had been operational for barely a year when the keepers vanished. Within months of its first lighting, it had produced the most enduring maritime mystery in Scottish history.

What the log said

The inquiry, according to the story that circulated for years afterward, relied on Thomas Marshall’s lighthouse log — kept diligently up to the point when the entries simply stopped.

What the log described, in the days before the disappearance, was a storm of unusual and terrifying violence. Marshall wrote of seas he had never seen before, of waves that seemed to come from directions that made no meteorological sense, of all three men being seized by fear. The entries have an almost feverish quality. Ducat, who Marshall described as a steady and experienced keeper not given to displays of emotion, was apparently weeping. McArthur, noted as a man rarely troubled by anything, was praying. The final entry, supposedly written on the 15th of December, read simply: Storm ended, sea calm. God is over all.

Then nothing.

Here was the problem. The storm Marshall described was not recorded anywhere else. Ships in the area during the same period reported rough but not exceptional seas. No other station logged anything approaching the conditions Marshall described. Either the Flannan Isles had experienced a localised meteorological event of extraordinary severity that left no trace in any other record, or something about the entries required a different explanation entirely.

It was a remarkable story. Unsettling, atmospheric, perfectly constructed to resist any rational accounting.

There was a reason for that.

What the record actually shows

The dramatic log entries were fabricated. They were injected into the story several years after Ducat, Marshall and McArthur disappeared, almost certainly to sensationalise what was, in the official record, a rather more straightforward tragedy. No such entries have ever been produced from the actual lighthouse log. The last real entries, covering weather readings for the morning of the 15th of December, were made on a slate ready to be transcribed later. They were entirely routine.

The half-eaten meal was equally fictional. A detail borrowed wholesale from Wilfrid Wilson Gibson’s 1912 ballad Flannan Isle, a poem that had no basis in the official inquiry and introduced several invented details that lodged themselves permanently in the popular version of the story.

The true sequence of events was this. On the night of the 15th of December 1900, the transatlantic steamer Archtor, on a passage from Philadelphia to Leith, passed the Flannan Isles and noted in her log that the lighthouse showed no light. When the Archtor docked in Leith three days later, the sighting was reported to the Northern Lighthouse Board. The relief vessel Hesperus was dispatched immediately but could not reach the island until the 26th due to severe weather — which is why the ship was late, and why the Hesperus crew arrived knowing something was already wrong rather than discovering it fresh.

What the island showed

The physical evidence was more honest than the legend, and in its own way more troubling.

On the western landing (the side most exposed to Atlantic swells) investigators found damage consistent with an exceptionally large wave strike. Iron railings had been bent and twisted. A stone box thirty-three metres above sea level had been broken open and its contents strewn about. Ropes had been torn from their moorings. On the eastern landing, everything was undisturbed.

The superintendent Robert Muirhead, who had personally recruited all three missing men and knew them well, conducted the official investigation. His conclusion was straightforward: two of the keepers had gone to the western landing, almost certainly to secure equipment after a storm, and had been caught by a rogue wave. The third had gone down to help and been taken by a second. They had not expected to be gone long. One of them — probably McArthur — had left without his oilskins. They were just going to check the equipment.

There was a further detail that gave Muirhead’s theory particular weight. Thomas Marshall had previously been fined five shillings (a significant sum) for losing equipment in a previous storm. He would have been strongly motivated to go out and secure anything at risk, whatever the conditions.

It is, as explanations go, entirely plausible. Rogue waves of exceptional size occur regularly in the waters west of the Hebrides. The damage to the western landing is consistent with such an event. The official cause of death, recorded in the register for all three men, was noted simply as probably drowning.

What remains

Strip away the invented log entries, the fictional meal, the borrowed details of the ballad, and what you are left with is this: three experienced men, a dangerous landing stage, and a wave large enough to take all of them before any one of them could raise an alarm.

That is the probable truth of what happened on the Flannan Isles in December 1900. It is not mysterious. It is not inexplicable. It is the kind of thing that happened to lighthouse keepers on exposed Atlantic rocks, and the kind of thing that the Northern Lighthouse Board’s records show happening, in various forms, more than once.

And yet the Flannan Isles had that old reputation for a reason. The local fishermen who refused to land without observing the old courtesies were not credulous. They were careful. Places at the edge of the known world, battered by Atlantic weather, accumulate a kind of gravity over centuries that rational investigation does not entirely dispel.

The myth needed the dramatic log and the overturned chair to work as a supernatural tale. The plain facts of three ordinary, hard working men, a wave, a morning in December — are harder to dismiss precisely because they require nothing supernatural to explain them and nothing at all to make them strange.

They went down to the western landing and did not come back.

That’s enough.