Theorists predicted it would be one of the most dramatic events in the Universe: two black holes merging in a distant galaxy and then flying out of that galaxy, releasing the energy of tens of millions of Suns.

What we haven't seen: general relativity predicts that colliding black holes should fly free, but astronomers have failed to spot this.Tim Jones/McDonald Observatory

But a survey of thousands of quasars — distant, active galaxies where large black holes reside — has failed to turn up a single such 'kicked' object. The result has theorists scratching their heads and double-checking their calculations. It was presented on 29 May at the American Astronomical Society biennial meeting in Honolulu, Hawaii.

Black holes are objects millions or billions of times more massive than the Sun; their gravity is so powerful that not even light can escape them. Some of the largest black holes are believed to lie at the centre of quasars.

Occasionally, a quasar can have two black holes at its core. When it does, they will slowly spiral inward, until suddenly one lurches towards the other, says Christopher Reynolds, a theoretical astrophysicist at the University of Maryland in College Park.

The colliding holes, Reynolds says, should give off a mighty gravitational shockwave that sends them speeding out of the galaxy at thousands of kilometres per second.

"These events are by far the most powerful things that exist at that point in time in the Universe," says Reynolds. "They release the energy of tens of millions of Suns in about an hour."

Spins in synch

Calculations show that the kicked holes will carry with them a giant cloud of dust, according to Erin Bonning, an astronomer at the Paris Observatory, France. The speeding dust cloud ought to betray its presence as a reddening of the spectra of the elements present in the quasar, she says.

But a survey of 2,600 quasars by the Sloan Digital Sky Survey found no such evidence. Some hydrogen spectral lines appeared to be reddened, Bonning says, but she was unable to rule out other causes for this, such as winds around the black-hole core.

And other lines, notably one for magnesium, didn't change at all. "This is an interesting non-discovery," she says.

It certainly has theorists doing a bit of soul searching. "An easy way out would be to say that numerical relativity is wrong," says Reynolds, referring to the predictions made from Einstein's general theory of relativity. But, he says, different groups, using a variety of techniques, all predict the kick. "I don't think anybody would reasonably believe that there are errors in the simulations."

Reynolds says his group has an alternative theory. Black holes often spin on an axis; if two holes' spins were aligned, they would be kicked out at much lower velocities and would be difficult to detect from Earth.

But Reynolds says he has no convincing reason for why the two black holes should have similar spins. The mystery of the missing kicks, it seems, will persist.

Visit our holesdefy_theory.html">newsblog to read and post comments about this story.