There is a paradox at the heart of this subject. Dark matter is called dark because it gives off no light, absorbs none and reflects none. And the way physicists hope to catch it is to descend a mile into the earth, to a place from which the weight of the planet has excluded every last photon of sunlight, and there to sit in the dark, watching for one small flash.To throw light on the dark, you must first put out every light you have.On 1 September 2026, the Lawrence Berkeley National Laboratory announced, on behalf of an international collaboration of some 250 scientists called LUX-ZEPLIN, that its detector had recorded a single particle interaction it cannot explain.
The instrument sits a mile below South Dakota, in the workings of an old gold mine. Ten tonnes of ultrapure liquid xenon fill a chamber watched by 494 light sensors. Between March 2023 and April 2024, the detector logged 220 live days. On 16 June 2023 it registered one event, and months of effort have failed to pin it on radioactivity, on stray neutrons, on neutrinos, or on the machine itself.A caution, since the headlines have been careless. This has not been published. It was announced at a conference in Japan and goes next to a preprint server, and only then to peer review. Rick Gaitskell of Brown University, who speaks for the collaboration, is severe on the point: they are not claiming to have seen dark matter. They have something they could not kill, and they would like the world to try.
Because roughly 85 per cent of the matter in the universe appears to be something we have never identified. It is called dark rather than merely invisible for a reason. Glass is invisible and still bends a beam of light; smoke is dark and still blocks one. This does neither. Nor can it be touched — and that is the same fact rather than a second one, since touch is your electrons refusing the table’s electrons, and light is the messenger of that very force. A thing with no dealings with light is also a thing your hand passes through.
The evidence is entirely gravitational. The famous exhibit is galaxy rotation: in the 1970s Vera Rubin found that stars at the rim of spiral galaxies wheel around far too fast — fast enough to have been flung into space long ago, unless unseen mass were holding them. But that is not the strongest exhibit, and you should know it, because “perhaps we simply misunderstand gravity” is a fair objection to that one alone. It is a much weaker objection to the rest of the file: light bending around galaxy clusters, the pattern of the cosmic microwave background, and the plain existence of galaxies. Ordinary matter feels both gravity and pressure, and so resists being gathered. Something with five or six times its gravitational weight, and no pressure at all, had to do the gathering — otherwise there has not been enough time since the Big Bang to build the universe we see.
What Is a WIMP?
The leading suspect is the WIMP — Weakly Interacting Massive Particle — and every word carries weight. Massive: heavy, comparable to a proton or hundreds of times heavier. Weakly interacting: it ignores light and ignores the force binding atomic nuclei, but may feel the weak nuclear force, the one behind radioactive decay. Particle: a discrete thing, not a correction to the laws of motion.
WIMPs have dominated the field for forty years because of a coincidence physicists call, drily, the WIMP miracle. Calculate how many such particles would have survived the cooling of the early universe, and a particle of roughly the mass associated with the weak force turns out to be left over in almost exactly the quantity astronomers independently measure. Two unrelated sums landing on one number is the sort of thing that persuades people.
If the South Dakota flash was a WIMP, it was a heavy one — over 200 times the mass of a proton — and, more awkwardly, one interacting in a manner more elaborate than the simplest theory allows. It does not merely sit where nobody was looking. It needs machinery nobody predicted.
Xenon is used because it is heavy, dense, purifiable, and does two things at once when struck: it flashes, and it sheds electrons. The ratio between the two reveals whether something hit a nucleus, as a WIMP would, or merely an electron, as ordinary radioactivity does. And the mile of rock above — physicists call it the overburden — is the most effective scientific instrument ever deployed, precisely because nobody designed it or paid for it. Its whole function is to manufacture a darkness deeper than any on the surface of the world, so that one flash, when it comes, has nothing to hide behind.
Reading the 0.5 Per Cent Backwards
Every account attaches a figure: a 0.5 per cent chance the event could be explained by known backgrounds. Almost every account then inverts it.
That figure is not the probability that dark matter has been found. It is the reverse. It says that if nothing but ordinary background were present, an oddity like this would turn up about once in two hundred tries. Physics will not use the word discovery below odds of roughly one in three and a half million, and it insists on that severity because one-in-two-hundred flukes are common in a world running hundreds of analyses at once. Italy’s Gran Sasso laboratory has had a claimed dark matter signal for years that nobody else can reproduce.
Set against that is the most striking line in the announcement, from Aaron Manalaysay of Berkeley: in his whole career, this is the first oddity to survive every test he could devise.
Kolar, 15 August 1965: India Saw It First
Now the part that ought to be taught in Indian schools.
On Independence Day 1965, a team working 2,300 metres down in the Champion Reef mine at the Kolar Gold Fields published the world’s first detection of atmospheric neutrinos. The Indian names were B.V. Sreekantan, M.G.K. Menon, V.S. Narasimham and P.V. Ramanamurthy, sponsored by Homi Bhabha’s Tata Institute, working with partners from Japan and Britain. A rival American group in a South African mine recorded its events slightly earlier; the Indians got into print a fortnight ahead.
The reasoning that took them to Kolar is precisely the reasoning that took LUX-ZEPLIN to South Dakota. Deep mines are where the darkness is. The Kolar laboratory ran nearly four decades and closed in 1993, when the gold mines closed — not by any decision to end Indian underground physics, but because a mine stopped paying and it was nobody’s job to notice that the laboratory inside it was a separate thing. The study of those same neutrinos moved to Japan, and produced a Nobel Prize in 2002.
One footnote the physics literature has only lately made: that shaft was cut over generations by Kolar miners, overwhelmingly Dalit, who appear in no citation. The darkness was free to the physicists. The digging was not.
A Glimmer, Not a Dawn
India has not left the field. On 2 September 2017, a quarter-century after Kolar went dark, the Jaduguda Underground Science Laboratory opened 555 metres below Jharkhand in a working uranium mine a fine irony of address, since uranium ore is itself radioactive, and the rock that pays for the shaft is the very interference the experiment must shut out. India’s dark matter experiment runs there on about seventy grams of superheated fluid. Against ten tonnes of xenon that sounds absurd, but it hunts lighter quarry by a different method, and it keeps alive a tradition that very nearly died at Kolar.
The larger business is unfinished. The India-based Neutrino Observatory, sanctioned by the Cabinet in 2015 at some ₹1,583 crore for a Tamil Nadu site, has not had a metre dug. Clearances, litigation, a state government’s opposition and a public confusion between a neutrino detector and a nuclear plant have between them held it eleven years. China’s equivalent is finished and publishing.
So: can we throw light on dark matter? Not yet. What South Dakota has produced is a glimmer — one flash, at the ragged edge of significance, in a chamber built expressly to contain nothing else. More data will either thicken it into a signal or dissolve it, as many such glimmers have dissolved before.The dark matter, if it is out there, has waited thirteen and a half billion years and can wait a little longer for us. It is our waiting, not the particle’s, that should trouble us.