The Moon Rang For Three Hours And Fifty-One Minutes
A retired engineer spent nine years cataloguing every long-period event in the public lunar archive. We pulled forty entries at random. Four were wrong, and he corrected them while we watched, and asked us to print that.
SILVER SPRING, Md. — Ellery Vaughan-Kimathi retired after thirty-one years of network engineering. He started the project the following Monday.
He has not really stopped since.
It is a catalogue: every long-period seismic event in the public lunar record from 1969 to 1977. Duration, station, and a note. It runs to 1,140 entries, and every record in it is free to anybody with an internet connection.
What is in it
Long-period events on the moon ring. This is not disputed and has not been since the instruments were installed: an impact or a moonquake produces a signal that persists for far longer than an equivalent event on Earth, sometimes for hours, without the clean separation of arrival phases that terrestrial seismology relies on.
The catalogue's longest entry, 1971-284-A, runs three hours and fifty-one minutes.
The compiler's own annotation on that entry is the reason this newspaper drove to Maryland:
compiler's note: ringing persists far beyond the model for a body of this
composition. This is well documented and has ordinary explanations. I am
not proposing an extraordinary one. I am only listing the events.
The correction we owe him
When this story first ran we wrote that Mr Vaughan-Kimathi "declines to offer an explanation."
He does not decline. He offers four, at length, in his own catalogue's introduction, and he was mildly wounded to find himself described as withholding something.
His preferred explanation, which now sits in this paragraph rather than in the fourteenth, is scattering: a heavily fractured, extremely dry outer layer, in which seismic energy takes an enormous number of paths of slightly different length and arrives spread out rather than sharp. Dry rock does not damp. There is no water to turn vibration into heat.
"It is a bell because it is broken and dry," he said. "That is the whole of it. It is not a bell because it is empty."
The spot-check
This newspaper does not report a compilation. It reports a compilation it has checked.
We pulled forty of the 1,140 entries at random and compared them to the public archive. Thirty-six matched exactly. Four did not: two transcription errors in duration, one wrong station identifier, one event that appears twice under different dates.
We took the four to him at his kitchen table. He looked at each for about ninety seconds, said "yes, that's wrong," three times and "oh, that's embarrassing" once, and corrected them while we watched. The version this newspaper holds is the corrected one.
He then asked us to print that he had got four wrong, which is a sentence very few people volunteer.
Dr. Josefina Marek-Anand, a planetary seismologist who reviewed twelve entries for us, described the catalogue as "more careful than most of what crosses my desk from institutions" and added the qualification she wanted printed: that a careful catalogue of a well-understood phenomenon is a useful object and not an anomaly.
"He has done a real service and he knows exactly what it is," she said. "It's the people who cite him who don't."
The six o'clock edition
Three stories a morning, in your inbox before the coffee is made.
Corrections to this story
- clarification — This story said the compiler of PX-1788 'declines to offer an explanation'. He does not decline; he offers several ordinary ones, and asked us to say so. The sentence has been changed to reflect that, and his preferred explanation now appears in the fourth paragraph rather than the fourteenth.How it happened: A cut made for length removed the compiler's own explanations while leaving in a characterisation that depended on their absence.What we changed: Paragraph 4 restored; the word 'declines' removed.
Sources & Method
A private individual gave us a compilation. Compilations are only as good as their sources, so we did not report the compilation — we spot-checked it. We pulled forty of its 1,140 entries at random from the public archive and compared them line by line. Thirty-six matched. Four did not. We took the four back to the compiler, who agreed, corrected them on the spot, and asked us to say that he had got them wrong. He is saying it here.
Who we spoke to
- Ellery Vaughan-Kimathi, Retired network engineer; compiler of the catalogue. Interviewed in person at his home over two days 9–10 June 2026 Corrected four of his own entries during our visit when we showed him our spot-checks.
- Dr. Josefina Marek-Anand, Planetary seismologist. Interviewed by video call; reviewed twelve catalogue entries at our request 22 June 2026
- Ruthanne Ojukwu-Vance, Archivist, planetary data holdings. Interviewed by telephone; confirmed all underlying records are public and unrestricted 24 June 2026
Documents
- PX-1788 — Lunar seismometer network, long-period event catalogue, unofficial compilation accepted
What we could not confirm
- Whether the four entries we found wrong were the only four wrong entries. We spot-checked forty of 1,140.
- The interior structure that would produce the observed durations. There are several published models and this newspaper is not competent to choose between them.
How was this story?
We publish the result, whatever it is. Reader verdicts appear on the front page and in our newsroom metrics.
Readers' Letters 0
Printed at once under the name you give and read by the desk afterwards; anything unfit is removed, with a note saying so, and nothing else is ever deleted — only corrected. Letters that changed something in the story carry a mark saying so, and there are 5 of those across the archive.