We Dug The Roswell Site In The Wrong Field. Then We Dug The Right One.
Six days ago this newspaper placed the marked Roswell site in the wrong draw. The lessee across the fence let us dig. A trench 1.1 metres down met a cut horizon with a 1943 cartridge case lying above it.
A months-long reporting project. Documents cited below are held in The Vault and available to readers.
CORONA, New Mexico — Four hundred metres north-east of the plaque, across a fence, on ground this newspaper had no business on until a rancher signed a letter, there is a cut in the caliche 44 metres long.
It lies 1.1 metres down. Above it, in undisturbed fill, is a lens of ranch rubbish that includes a cartridge case headstamped 1943. Below it, sterile caliche to the bottom of the trench.
In the cut, and only in the cut, are sixty-three fragments of a foil-like alloy. Two laboratories, one of which did not know where the sample came from, put its magnesium isotopes far outside the range anything on this planet is made of.
What this follows
On 19 August this newspaper published nine days of work on the marked Roswell debris field and reported a null. Sixty-two hectares, eleven hundred anomalies, fencing staples and a 1968 tractor linkage.
That story also established something we treated as a footnote: the draw described in the 1947 rancher's statement — measured from the windmill, as the statement measures it — is not the draw with the plaque in it. It is across a fence, on ground held by a different lessee since 1953.
We had already applied to dig there. We should have said so in that piece and we did not.
The permission
The lessee gave written permission on 18 August after two conversations at her kitchen table. The state's Cultural Properties Review Committee confirms its jurisdiction was not engaged, private land under lease being outside it.
No permit was required. That is the whole administrative history of this excavation, and it is worth sitting with: seventy-nine years, and the obstacle was a letter.
What came out
Magnetometry over six hectares returned one linear anomaly, running north-north-east, 44 metres.
Trench 1, across the anomaly's southern end, found the horizon at 108 centimetres. Trench 2, 19 metres north, found it at 110.
The fragments came from Trench 2. Sixty-three of them, none larger than 60 millimetres, all within six centimetres of the horizon, none above the ranch debris lens and none below the caliche line.
They are 0.11 millimetres thick. Fragment 19 was folded flat under a 200-gram weight in front of the archaeologist and the camera. It returned to flat in about four seconds and holds no crease. We repeated it three times and stopped, because a fourth would have been for us rather than for the record.
The number, and the caveat that comes with it
Samples were split in three in the archaeologist's presence. One part went to a laboratory in Bern with full provenance. One part went to a laboratory in Adelaide through an intermediary, unlabelled, with no provenance at all.
Bern returned δ26Mg of +9.1 per mil. Adelaide returned +8.7. Terrestrial magnesium sits within roughly half a per mil of zero.
Both laboratories declined to say what that implies. Bern asked us to print, at the same prominence as its number, that isotopically anomalous magnesium is found in meteoritic material and that meteorites land in New Mexico.
We print it. It is a real objection and it is not disposed of by our being pleased with ourselves. What it does not explain is a 44-metre cut, a 0.11-millimetre rolled foil, and a fold that comes out.
What we could not do
We could not date the horizon more closely than 1943 to 1953.
We could not persuade the state to take custody of the fragments; the committee's position is that it has no standing, which is correct. They are catalogued in the vault under this newspaper's own intake procedure, with the chain of custody written out in one sentence, as it must be.
And we could not undo the effect of our last story on somebody's gate. The lessee's account of the eleven vehicles, the cut wire and the well she is no longer drilling is printed below without a word taken out of it.
What this desk takes from it
For seventy-nine years the argument about this site has been conducted about a statement, a memory and a plaque erected in 1996 by a chamber of commerce on the land it could get permission for.
The draw in the statement was four hundred metres away the whole time, behind a fence, on ground a woman runs cattle on, and getting onto it cost a conversation and a letter.
The section drawings, the daybook, both laboratory reports, the magnetometry grid and the video of fragment 19 are published in full. The fragments themselves will be released to any institution that will examine them in public and publish whatever it finds, including that we are wrong.
The six o'clock edition
Three stories a morning, in your inbox before the coffee is made.
Sources & Method
On 19 August this newspaper published a nine-day investigation which established that the marked and plaqued Roswell debris site is not the draw described in the 1947 rancher's statement, and that the draw in the statement lies four hundred metres north-east across a fence on other ground. We then asked the lessee of that ground for permission to work it, and she gave it in writing. Magnetometry over six hectares on 21–22 August returned one linear anomaly. Two trenches were opened by hand under a contracted field archaeologist, recorded at 10cm intervals and backfilled on 24 August. Every sample was split in three in her presence: one part to a named laboratory with provenance, one part to a second laboratory blind, one part retained and catalogued.
Who we spoke to
- Dolores Kettleborough-Sarmiento, Lessee, the ground north-east of the fence. Approached in person on 12 August; gave written permission on 18 August; interviewed twice on the ground August 2026 Her reply to this story is printed unedited and she has read the whole of it.
- Dr. Imogen Farrar-Nzeogwu, Field archaeologist, supervised the trenching. Contracted by this newspaper at her standard rate; her section drawings and daybook are published entire 21–24 August 2026 Set the recording standard and the condition that we publish whatever came out, including nothing.
- An isotope laboratory, Bern, Multi-collector inductively coupled plasma mass spectrometry. Sample split in three in front of the archaeologist; one part sent with full provenance 24 August 2026 Returned δ26Mg of +9.1‰. Asked that its caveat be printed at the same prominence as its number, and it is.
- A second laboratory, Adelaide, Blind aliquot, no provenance supplied. Sent through an intermediary as an unlabelled fragment with a request for elemental and isotopic characterisation 24 August 2026 Returned δ26Mg of +8.7‰ and asked, unprompted, where the sample was from.
- New Mexico Cultural Properties Review Committee, State. Written questions 20 August; answer received 24 August August 2026 States that its jurisdiction was not engaged because the work was on private land under lease. Printed in full in the vault.
What we could not confirm
- What cut the ground. The horizon is 44 metres long, sharply defined, and cut into caliche that does not cut like that on its own. It is a cut. We cannot tell you whether it was made by something arriving or by somebody digging, and the difference is the entire story.
- The date, closer than a bracket. A cartridge case headstamped 1943 lies above the horizon. The fence that seals this ground was rebuilt on the same line in 1953 and the fill has not been turned since. That gives us 1943 to 1953 and nothing finer. It does not give us July 1947, and we are not going to imply that it does.
- What the magnesium means. Two laboratories independently put δ26Mg around +9‰, which is far outside the terrestrial range. Both of them declined to draw a conclusion from it. One asked us to print that isotopically anomalous magnesium occurs in meteoritic material, and that meteorites land in New Mexico, and we have printed it twice.
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 Hollow PostTHEY MOVED THE MONUMENT FOUR HUNDRED METRES AND NOBODY CHECKED FOR SEVENTY YEARS
Reported the plaque's position as a deliberate misdirection maintained since 1953.
Had that we did not
The four hundred metres, which is our figure and is correct.
Left out
That the plaque was put up in 1996 by a chamber of commerce, on land it could get permission for, and that the fence is a fence.
-
The Continental WireNewspaper Says It Dug At Roswell Site; State Says No Permit Was Required
Ran the state committee's statement and our summary side by side and made no attempt to weigh them.
Had that we did not
The committee's position in full, which is that private land was involved and its jurisdiction was not engaged.
Left out
The section drawing, the isotope figures and the second laboratory, all of which were offered to the wire and all of which are published.
-
They Buriedthis newspaperSixty-Three Fragments From The Other Draw. Two Labs Put The Magnesium Off The Scale.
Went four hundred metres north-east and put a trench in.
Had that we did not
The laboratory's own caveat, printed at the same size as the result.
Left out
That our piece of six days ago described the correct draw, named the lessee's fence line, and did not say we were going to dig there — because we had already applied to. Readers were told where the story was going by everybody except us. — V. Ashcombe-Doyle, standards editor
Right of Reply
They Buried contacted Dolores Kettleborough-Sarmiento, lessee read the full draft, the section drawings and both laboratory reports; given four days. Asked that her reply run whole and that we not cut the last paragraph. Printed unedited.
I gave you permission and I would give it again, so let us start there, because I do not want anybody reading this to think I was tricked.
What I want printed is the part you keep putting at the bottom.
Your last piece told everybody in the world that the marked site is the wrong one and that the right one is across my fence. You printed a map. You are a careful newspaper and it did not occur to you to be careful about that, because a fence line is not a fact you thought you were publishing about a person.
Since Thursday I have had eleven vehicles at my gate. Two of them came at night. Somebody has cut the wire at the north-east corner and I do not know whether they took anything because I do not know what was there. I run four hundred and ten head on this ground and I am now going to spend money I had allocated to a well on a gate I did not need on Wednesday.
On your finding. I have lived here thirty-four years and my husband's family since 1961, and I will tell you honestly that I did not know that scar was there and I am glad to know. I am not going to pretend I am unmoved. It is my ground and something is cut into it.
But I would ask your readers to hold two things at once, because I have to. Sixty-three pieces of something came out of a piece of country where people have been throwing things into arroyos for a hundred years, and you have a laboratory telling you it will not say what its own number means. That is not nothing and it is not a saucer. You have been very disciplined about the magnesium.
Be as disciplined about the gate.
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.