In a world where a deep-space signal turned out to be a snack, astronomers at one of the planet's great radio telescopes spent years chasing a cosmic mystery that was, in the end, their own kitchen microwave reaching for an early lunch. The story of the perytons is the best cautionary tale in modern radio astronomy precisely because it sits tangled up with one of the field's most important real discoveries — and for a while, nobody could tell the snack from the cosmos.

The real discovery came first. In 2007, Duncan Lorimer and colleagues, sifting through archival data from the 64-metre Parkes radio telescope in New South Wales, found a single, spectacularly bright burst of radio waves lasting only milliseconds. Its key feature was dispersion: lower-frequency components arrived measurably later than higher-frequency ones, exactly as a pulse smeared by passage through vast amounts of intergalactic plasma should be. The dispersion measure implied the burst had traveled from far outside our galaxy — hundreds of megaparsecs. The 'Lorimer burst,' as it became known, was the founding event of what we now call fast radio bursts, FRBs, one of the hottest mysteries in astrophysics.

Then the trouble started. Parkes kept finding other millisecond bursts that looked tantalizingly similar — dispersed-ish, brief, dramatic — but that were obviously wrong in a telling way: they appeared in all the telescope's beams at once. A genuine source on the sky lands in one direction; something detected in every beam simultaneously is local, near the dish, terrestrial. These impostors were named 'perytons,' after a mythical beast that casts a human shadow. For years they cast doubt on the whole enterprise. If Parkes was generating its own fake bursts, maybe the celebrated Lorimer burst was just the cleanest peryton of all — an instrument artifact, not a message from eight billion trillion kilometres away. An entire emerging field hung under that suspicion.

The break came in 2015, and it is a masterpiece of unglamorous detective work. Emily Petroff and a team that included the original peryton investigators noticed a damning pattern in the timing: perytons clustered around midday, on weekdays, at the telescope site. They also installed a radio-interference monitor that listened across a wider band. The monitor caught the perytons red-handed at 2.4 GHz — the operating frequency of a microwave oven. The mechanism turned out to be exquisitely specific. A microwave's magnetron does not shut off cleanly the instant you open the door; if an impatient staff member yanked the door open before the timer finished, the magnetron's dying gasp leaked a burst of radio energy as it powered down. When the telescope happened to be pointed toward the staff kitchen, that leak swept across the beams and painted a 'peryton' — a fake fast radio burst, served fresh at lunchtime, for the better part of seventeen years of accumulated data.

The skeptical reading here is the satisfying one, because skepticism actually solved the case rather than just casting shade on it. The perytons were a textbook reminder that an unexplained signal is not an exotic signal — that the prior probability of 'someone reheated their coffee' enormously exceeds 'aliens,' or even 'new astrophysics,' until you've eliminated the building you're standing in. Petroff's team did the eliminating with timing statistics, a frequency monitor, and, gloriously, a deliberate experiment: they reproduced perytons on demand by opening the site microwave mid-cycle while the dish was aimed the right way. Causation nailed, not merely correlated.

But the inversion that makes this story matter is what the microwave did not explain. The same 2015 work that convicted the kitchen appliance also formally cleared the cosmos. The team showed that the Lorimer burst, FRB 010724, could not have been a microwave: it showed up in only some beams in the pattern an external sky source produces, its dispersion was far cleaner and deeper than any peryton's, and its properties were flatly inconsistent with magnetron leakage. Killing the impostor sharpened the real signal. Far from debunking fast radio bursts, the microwave discovery rescued them — by amputating the terrestrial confound that had been making people doubt the genuine article.

And that is where the case stays open, beautifully. The fakes are solved; the real thing is not. Fast radio bursts are now confirmed extragalactic events — thousands have been catalogued, a few have been traced to specific galaxies, at least one to a magnetar in our own — yet what actually produces these millisecond detonations of energy remains genuinely unexplained. The microwave saga is a parable with two edges: it shows how easily a kitchen can masquerade as the cosmos for seventeen years, and it shows that ruthlessly ruling out the mundane is exactly how you earn the right to keep calling the remainder a mystery. The perytons were lunch. The fast radio bursts, so far, are not.