The case record
What was reported?
Three lighthouse keepers disappeared from Eilean Mòr in December 1900. The contemporary inquiry favoured their being swept into the sea while attending equipment near the west landing.
A relief vessel arriving on 26 December found James Ducat, Thomas Marshall and Donald McArthur absent. Inspection showed major storm damage around the landing. Later poems and popular retellings introduced dramatic details that should not be mistaken for the original lighthouse records.
The range separates the likely loss period from its discovery. The lighthouse marker is representative of the island, not a known point of death.
- Reported date
- December 1900; absence discovered 26 December
- Witnesses
- Historical records, investigators and observers; see case-specific account · Not reliably reducible to a single count
- Investigation
- Historical and research accounts reviewed from: Northern Lighthouse Board; National Records of Scotland
Evidence in the record
The Northern Lighthouse Board publishes the relief report and superintendent Robert Muirhead’s assessment. These documents provide a stronger basis than alleged diary entries about unexplained terror, which are not established by that archive.
Competing explanations
Official and research context
The Northern Lighthouse Board inquiry favoured a large wave sweeping the keepers from the west landing.
Conventional explanations
A large wave striking the landing could explain the missing men and displaced equipment. All three being outside together remains a reconstructive question; absence of recovered bodies is consistent with the exposed sea setting.
Arguments that remain contested
No witness saw the fatal event, so the exact timing and the keepers’ final actions cannot be confirmed.
Locations & timeline
Coordinates are representative. Area and route labels do not imply exact paths or object positions.
- Primary reference location
Eilean Mòr, Flannan Isles
The range separates the likely loss period from its discovery. The lighthouse marker is representative of the island, not a known point of death.
Representative case location · Approximate area radius: 2 km
58.290°, -7.590° · point · Regional reference only; not survey-grade
Record notes
Map coordinates are representative and do not establish an exact incident position.
Case-specific sources researched 8 September 2026. Evidence, alternatives, dates and representative geography reviewed.
Original dataset status: likely-conventional