Three Laboratories Have Dated The 1,412-Metre Band. I Have All Three Numbers.
I obtained them separately, from three laboratories on three continents, each of which believes it is the only one that ever measured this band. Their ages agree within eighty-six years. All three are stratigraphically impossible.
A regular column. The views expressed are the writer's own.
I have three unpublished measurement reports on my desk. They describe the same forty centimetres of ice, they were written by three laboratories on three continents, and each believes it is the only one that measured it.
They reached me separately, over eleven months, from three sources who do not know each other.
I have now told all three that the other two exist. That was the hardest telephone call of my working year and I made it three times.
The three numbers
The band runs from 1,412.06 to 1,412.46 metres in a core drilled in East Antarctica in December 2011. Cut wet, logged at the face as dark and laminated.
Krypton-81 in the trapped gas, at Dr. Anneke Ferreira-Lindqvist's laboratory: 12,847 years before present, plus or minus 110.
Radiogenic argon on the same gas fraction, a different clock, in Prof. Kwabena Ruusuvuori-Adeyemi's group: 12,933, plus or minus 140.
Radiocarbon on the occluded particulate carbon at Dr. Marisol Tanaka-Egbuna's facility — a solid-phase age, not a gas age: 12,881, plus or minus 95.
Eighty-six years separate the highest from the lowest.
All three also report metallic microspherules through the full forty centimetres at 337 to 344 times background, with a platinum-group signature. None knew the others had.
Where the chain breaks
An ice core is dated in links, and you can only see the failure if you know which link is which.
The first is counting. Annual layers are read down the core in visual stratigraphy, in conductivity, in the seasonal swing of the chemistry. It is arithmetic, and in this core it stops at 1,206 metres, where the layers thin past resolving.
The second is tie points. Sulphate spikes from known eruptions anchor the count to fixed dates. Below 1,290 metres there is not another usable one until 1,538. Two hundred and forty-eight metres with nothing to pin.
The third is the flow model, which fills that gap. Accumulation, ice thickness, a thinning function, and an age for any depth you ask about. For 1,412 metres it gives 30,900 years.
That number is not a measurement. It is an output.
The fourth link is the one everybody reaches for first, and rightly. Gas is younger than the ice around it, because air keeps circulating in porous firn long after the layer beneath is sealed. The offset here is modelled at 640 years. The largest ever reconstructed anywhere is 5,700.
To close this gap you would need eighteen thousand.
And it cannot touch the third number at all. A radiocarbon age on solid particulate carbon carries no offset, because the particles were never gas. If the offset were doing the work, that age would sit thousands of years from the other two.
It sits between them.
Why this is not the drill fluid
This newspaper has published the contamination argument and it is a good one: Vostok and the drill fluid showed how far a wetted rind travels into a result.
It does not reach here, and the reason is geometrical. The fluid wets the outside. All three cut back the outer fifteen to eighteen millimetres and worked on the interior, and all three report the spherules spread evenly through it rather than graded towards the rind, which is what contamination always leaves.
We also had a retained sample of the 2011 fluid consignment assayed blind. Platinum, iridium and osmium below detection. This band is interior ice, not the wetted rind.
Clause 9
Here is why none of it is published, from the executed agreement a contracts office released to us:
"The Programme shall have sixty (60) days from receipt to review any manuscript arising from the transferred material and may, at its sole discretion and without statement of reason, require that publication be withheld. This requirement shall not be subject to appeal or arbitration."
It has been exercised on this band three times: August 2022, March 2024, January 2026. The nearest instrument in our vault, PX-1980, advises against granting exactly this power.
What I could not establish
I cannot say the flow model is wrong. Nobody has re-run it with these three ages as a constraint, and the programme has not released the inputs that would let anyone try.
I cannot say the band exists anywhere else. One hole has been cut this deep in that sector; the two nearest stop above 1,300 metres. One band in one borehole, until somebody takes a second core.
And annex B lists a fourth allocation from this band, in 2016, to a recipient redacted in our copy. None of the three knows who it is. Our appeal was filed on 9 July.
What I am doing with them
Printing them.
I do not know which link is wrong. I know the 30,900 is model output and the 12,881 is a measurement, and that when the two disagree by eighteen thousand years the honest thing is to say so and let people take it apart.
Three laboratories reached the same conclusion independently and none of them was allowed to say it. That is not a doubt about the ice. It is a sentence in a contract.
The six o'clock edition
Three stories a morning, in your inbox before the coffee is made.
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.
-
Signal & LedgerPublication Vetoes In Polar Sample Agreements: A Rising Compliance Burden
Treated clause 9 as a contracting trend and surveyed nine national programmes for readers who negotiate these instruments.
Had that we did not
The best account anywhere of why host programmes want the veto, and a table of the nine clauses that is better than anything we produced.
Left out
That one of the nine clauses has been used three times on one 40-centimetre band, which is the only fact in this subject that would interest a reader who does not negotiate contracts.
-
The Hollow PostTHE LAYER THEY WILL NOT LET YOU SEE
Reported the veto as suppression of a finding that would rewrite the last glacial maximum.
Had that we did not
The existence of the clause, and one of the three ages, roughly right.
Left out
That the host programme has never disputed a number, that the labs themselves have not resubmitted, and that a 40-centimetre band in one borehole is one band in one borehole until somebody cuts a second core.
-
They Buriedthis newspaperThree Laboratories Have Dated The 1,412-Metre Band. I Have All Three Numbers.
Got the three reports separately, got the agreement that stops them, and printed the clause and the numbers side by side.
Had that we did not
All three ages with their errors, the clause verbatim, and the host programme's defence of it at length.
Left out
This column is written in the first person about documents I hold, which is a shape that flatters the reporter. Three named scientists have their unpublished work printed here by a newspaper and not by themselves, and I did not ask any of them whether that is what they wanted. — V. Ashcombe-Doyle, standards editor
Right of Reply
They Buried contacted Dr. Reinout Baraka-Vlietstra, head of science operations, the host national programme sent the full column, the clause as we intend to print it and the three ages on 4 August 2026, with fourteen days to reply. Replied 18 August 2026. Printed in full and unedited.
I am going to defend clause 9, and I am going to do it without disputing a single number in your column, because we have not disputed them and nobody has asked us to.
Start with what a host programme is. We cut one hole. It cost eleven years and the working lives of a good many people, and it produced a finite object which is consumed by being studied. Every centimetre we send out is gone. When we allocate material we are not lending a book. We are giving away a piece of the only copy, and we do it something like four hundred times a year, and we are answerable for all four hundred to a ministry that could close us in a budget round.
Clause 9 gives us sixty days and a right to withhold. You have printed it accurately, including the part about no statement of reason and no appeal, and I am not going to pretend it is a mild instrument. It is not. It exists because a national programme that cannot control the first publication arising from its own core cannot allocate material to anybody. The alternative to a hard veto is not a soft veto. The alternative is that we stop sending samples out, and then there are no three numbers for you to obtain.
On this band specifically. You say we exercised it three times and you are right about the dates. What you have not printed, because none of your three could tell you, is what our review said, and I am not permitted to tell you either, which I accept is an unsatisfying sentence in a newspaper. I will say this much and it is more than I should. In no instance was the age the subject of the objection.
Now the part I expect you to quote against me. You ask why the three laboratories were not told about one another. The agreement does not require us to tell them. It requires them not to tell anyone else. Those are different obligations and both are ordinary. I did not design a system to keep three scientists ignorant of each other. I inherited a system in which confidentiality runs one way, as it does in every allocation instrument I have ever signed, and it did not occur to me that the effect of running it one way for fifteen years is that nobody can ever replicate anything without a journalist introducing them.
That is a fair criticism. It is yours, not mine, and I have put it in writing to our director this week.
What I object to in your column is the word suppressed, which you have been careful not to use and which every outlet reprinting you will use by tomorrow. Nothing has been destroyed. Three reports exist. Their authors hold them. The material that remains from that band is 6.1 centimetres and it is in our freezer at minus thirty, logged, and any qualified group may apply for it under the same instrument, including one of your three, and none of your three has applied since 2019.
Ask them why. I would like to know as well.
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.