We Checked 4,142 Frames Against Their Exposure Logs. Seven Should Show A Star.
Every lunar-surface magazine has a logged aperture and shutter speed. We computed the faintest object each frame could have registered. Seven exposures are long enough that a star should be there. None of the seven has one.
Background written by the news staff. Contains no argument or recommendation.
SEABROOK, Texas — Of the 4,142 published photographs taken on the lunar surface, 3,776 were exposed at a two-hundred-and-fiftieth of a second or faster.
At that speed, in that light, the only two objects in the lunar sky bright enough to mark the film are the Sun and the Earth. Both of them turn up, repeatedly, in exactly the frames where the arithmetic says they should.
That is the check on the method, and it is why the rest of this is worth reading.
Three numbers a frame
Each magazine carries a logged film speed. Each frame carries a logged aperture and a logged shutter speed. Given those three numbers, working out the faintest object the emulsion could have registered is a first-year exercise, not a specialist one.
We ran it on all 4,142, using two independently keyed transcriptions of the logs so that a typing error would have to happen twice in the same place to survive.
Only seven frames come out long enough for the question to be live: a quarter of a second or more.
Four of the seven were taken in the shadow of the vehicle to document a piece of equipment out of the sunlight. Two are shaded frames of a deployed experiment, taken with a gloved hand held over the lens. One was exposed while a magazine was being changed, with the camera pointing at nothing in particular, and it is the best of the seven.
What stood above the horizon
Dr. Solveig Mbeki-Ravensworth computed, for each of the seven, what was actually up there inside the field — from the site coordinates, the time, and the camera azimuth.
Six of the seven have nothing in frame brighter than magnitude plus 0.1. Rigel in one, Procyon in another, Achernar, Regulus, Spica. Ordinary bright stars, and every one of them fainter than the threshold those exposures reach once the film is treated properly.
The seventh has Sirius in it.
Page four
The paper threshold for the seven averages magnitude plus 0.6. On that number alone, five of them ought to show something.
They do not, and the reason is reciprocity failure: at long exposures an emulsion stops behaving as though halving the light and doubling the time were the same trade. It gets slower the longer you leave it open.
This is not obscure. It is in the manufacturer's data sheet of March 1968, on page four, in a table of five rows, in the same envelope the film shipped in. For the exposures in question the table costs between 2.4 and 3.1 stops, which moves the average threshold from plus 0.6 to minus 1.6.
Nothing in six of those seven fields is anywhere near minus 1.6.
The card taped to the camera
Wendell Achterberg, who is eighty-seven and lives in Seabrook, Texas, wrote the exposure card fixed to the body of the surface camera. Four lines, laminated.
He told us the crews worked from the card rather than metering, because a suited man at four pounds of pressure cannot turn an aperture ring reliably, and that this was a design decision made on the ground on purpose. He objects, at length and below, to our having called it a cheat sheet.
He also volunteered the thing we had not thought to ask: there should have been a fifth line, for shadow work, and there is not — which is precisely the case four of our seven frames fall into.
The frame we cannot close
Sirius stands at minus 1.46. Our corrected threshold for the frame it appears in is minus 1.61.
That is a margin of 0.15 magnitudes against a table whose own tolerance is a third of a stop, or 0.25. The arithmetic does not decide it. Mbeki-Ravensworth would not initial a draft that said otherwise, and this desk is glad she would not.
What this desk takes from it
The competing explanation in our vault, PX-1992, costs a fortnight of particle work with 2026 tools and stalls entirely with 1969 ones. This one costs four lines on page four of a document that came free with the film.
Neither is hidden. One of them is simply duller, and dullness is not a hiding place — it is only a thing people decline to read.
The six o'clock edition
Three stories a morning, in your inbox before the coffee is made.
Sources & Method
We wrote down in advance what a positive result would be: a frame whose logged settings put its threshold fainter than the brightest object standing in its field. Then we matched all 4,142 published surface frames to their logged aperture and shutter speed, had an astronomer compute the sky for the seven that qualified, and only afterwards went looking for the manufacturer's reciprocity data — which was in the same envelope as the film in 1969.
Who we spoke to
- Magazine logs for every lunar-surface film magazine, Logged film speed per magazine, logged aperture and shutter speed per frame. Taken from the published transcription and checked line by line against a second transcription keyed independently for this story January to March 2026 The two keyings disagree on 63 frames out of 4,142. All 63 are listed with the story and none of them is one of the seven.
- Manufacturer's data sheet for the flown colour reversal stock, March 1968, Technical publication issued with the film; reciprocity table on page four. Copied from a public technical collection, then found again in the flight-preparation folder that travelled with the magazines February 2026 Page four is a table of five rows and a note about colour balance. It has never been out of print, restricted, or behind a paywall.
- Dr. Solveig Mbeki-Ravensworth, Astronomer; computed the sky positions and the limiting magnitudes. Commissioned to work out, for each of the seven frames, what stood above the horizon inside the field, from the site coordinates, the time and the camera azimuth April to June 2026 Required that the seventh frame be printed as unresolved rather than explained, and would not initial the draft until it was.
- Wendell Achterberg, Retired camera systems engineer; wrote the laminated exposure card fixed to the camera body. Interviewed at his house in Seabrook over two afternoons and twice more by telephone; given the full draft and eighteen days May and June 2026 His reply is at the foot of this page and is the most useful thing in the story, including the parts where he is annoyed with us.
Documents
- PX-1992 — Attempted reproduction of 90 seconds of lunar surface footage — working file and two costings accepted
What we could not confirm
- The seventh frame. Sirius stands at magnitude minus 1.46 inside that field, and our corrected threshold for that exposure is minus 1.61 — a margin of 0.15 magnitudes against a table whose own stated tolerance is a third of a stop, which is 0.25. The arithmetic does not settle that frame in either direction and we are not going to write a sentence pretending it does.
- The 63 frames where the two keyings of the log disagree. We placed each by the sequence it sits in and none of the 63 lands in the long-exposure group, but a keying error in the original log rather than in either transcription would not show up in that check. If one of the 63 is really a half-second exposure, our seven is an eight and this story is short one frame.
- Whether the published set is the whole set. We measured 4,142 published surface frames. Frames exposed and never published — because they were blank, fogged, or duplicates — are not in our denominator, and a frame that came back black is exactly the frame this exercise would most want to look at.
How Others Covered This
The same events, as reported elsewhere on the same day. We list what each outlet had that we did not, as well as what we had that they did not — including where we come off worse. Why we print this.
-
The Meridian TelegraphSeven Photographs From The Moon That Nobody Has Ever Explained
Took our count of seven, credited us for it, and stopped one paragraph before the reason.
Had that we did not
The seven, correctly identified by magazine and frame number, from our own published list.
Left out
Page four of a data sheet that has never been out of print, and that we sent them, unasked, four days before they ran.
-
Signal & LedgerReciprocity Failure: Why Your Long Exposure Is Slower Than The Meter Says
Read the whole thing as a technical refresher for working photographers, and is better on the chemistry than we are.
Had that we did not
The 1968 table reproduced entire, in a legible scan, which we did not do and should have.
Left out
That the argument it is quietly answering has been running for fifty-odd years, and that the engineer who wrote the exposure card is eighty-seven and lives in Texas and will take a telephone call.
-
They Buriedthis newspaperWe Checked 4,142 Frames Against Their Exposure Logs. Seven Should Show A Star.
Set every frame against its own logged settings, decided in advance what a positive result would look like, and then read the manufacturer's table.
Had that we did not
All 4,142 frames with their settings, the computed threshold for each, the astronomer's working, and the one frame we cannot close.
Left out
We spent eleven weeks and a five-figure sum on a question a page-four table settles in four lines, and we put the table in paragraph twenty-one because it made a better shape. An eighty-seven-year-old man had to explain his own work to a newspaper. — V. Ashcombe-Doyle, standards editor
Right of Reply
They Buried contacted Wendell Achterberg, retired camera systems engineer on 14 May 2026 in person at his home, again on 3 June, and sent the full draft on 21 June with eighteen days to answer. Replied by e-mail on 8 July 2026 in five parts over two days, which he asked be run as one piece. Printed unedited.
You have done the arithmetic correctly and I want to complain about almost everything else, so let me start with the correct part and get it out of the way. The thresholds are right. The reciprocity numbers are right, and you have used the 1968 table rather than the 1971 revision, which is the right table because it is the one that was in the folder. Most people who write to me about this have not read either.
Now. You call it a cheat sheet twice, and once a laminated cheat-sheet, and I would ask you to consider what that word does.
A cheat sheet is what you make because you did not learn the material. That card is not that. That card is a control. We had a man in a pressure suit, in a vacuum, in light he had never seen before, with a hard time limit, wearing gloves at four pounds of differential pressure. You cannot feel an aperture ring through that glove. You can barely turn it. Metering was not a thing we took away from anybody, it was a thing that a suited crewman physically cannot do reliably, and the answer to a task a human cannot do reliably is not to tell him to try harder. It is to do the work beforehand, on the ground, where there is time, and hand him the answer.
So yes, they read the card. They were meant to read the card. Every hour that card saved on the surface was an hour of something else. I have never once been embarrassed about that and I do not know where your correspondent got the idea, unless it was from the way he asked the question.
What I am embarrassed about is smaller and you did not ask about it. There are four lines on that card and there should have been five. There is no line for the shadow. If you were photographing something in the shadow of the vehicle you were off the card and on your own judgement, and the frames that give you the most trouble in your own piece — the shadow frames, the shaded ones — are the ones I did not write a line for. I have thought about that considerably more than I have thought about the stars.
On your central point about general knowledge. I do not accept it. I do not think people stopped knowing an f-stop from a shutter speed and I do not think that is why the argument runs. Plenty of people who argue about these photographs are photographers. The reason the argument runs is that nobody reads a data sheet, and nobody has ever read a data sheet, including in 1969. I sent that page to three television producers between 1974 and 1981. One wrote back to say it was too technical for the audience. The audience he had in mind had been buying film with that exact table folded inside the box for a decade.
My last point is about your headline, and I say this without much hope. You have written that seven frames should show a star. They should not. They should show a star if you ignore a table that was printed and shipped with the film, and the whole of your piece is about not ignoring it. So your headline says the thing you spent eleven weeks disproving, in larger type than the disproof, and the seven people out of ten who read only the headline will come away with the opposite of what you found.
I have written thirty years of documentation and the one document anybody has ever argued about is a piece of card four inches by two with four lines on it, laminated by a woman in the photographic laboratory whose name was Rosalie and who used the machine for people's children's school badges. That is the sentence I would like printed, if you are going to print any of this.
Published unedited under our right-of-reply guarantee.
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.