New radio observations suggest that one of the brightest globular clusters in the sky, Messier 22, hosts two stellar mass black holes.

More than 150 swarming hives of stars known as globular clusters orbit the Milky Way galaxy, many of them beautiful telescope targets. Theorists have long argued whether black holes could hide inside these thriving stellar cities. New observations suggest they probably do — and perhaps in greater numbers than expected.

On the one hand, it would be natural for hundreds of stellar-mass black holes to form early in a globular cluster’s lifetime. The hottest, most massive stars in the cluster would evolve the most quickly, transitioning from blazing star to bloated giant to collapsed black hole in less than a billion years. These black holes would then flock to the cluster’s core — being more massive, they would move more slowly than smaller, lighter stars, and thus slide deeper into the cluster’s gravitational well.

But simulations have shown that gravitational interactions within the dense cluster core would eject all but a handful (and probably all but one) of the black holes shortly after they form. Even in less dense environments, such as the young open star clusters that bejewel the Milky Way’s disk, runaway stellar mergers might lead to one giant black hole in the cluster’s core.

Now Jay Strader (Michigan State University and Harvard-Smithsonian Center for Astrophysics) and his colleagues say they’ve discovered not one, but two black holes in Messier 22, one of the brightest globulars in the sky. Published in Nature today, their paper presents observations of two radio sources deep inside the cluster’s core that the astronomers found using the Karl G. Jansky Very Large Array (VLA) in New Mexico. After considering several possible explanations for the radio emission, the authors conclude that the most likely explanation is black holes shooting out jets as they gobble material from companion stars.

“Like all scientific results, confirmation is needed, but in my view, things look very good right now,” says Jon Miller (University of Michigan), who was not involved in the study. “Their interpretation of two black holes is very reasonable and consistent with the data.”

Black holes themselves emit no light, but they typically make themselves known when they feed on gas sucked from a stellar companion. This gas forms a disk-shaped smorgasbord of infalling material around the black hole, and some of it can shoot out as beams of plasma along the black hole’s poles as well. Before the gas in the disk careens past the point of no return, it heats up to over a million degrees Celsius, emitting X-rays in the process.

Yet oddly, the newly discovered black holes emit no detectable X-rays, making them unique among all other stellar-mass black hole discovered in the Milky Way. Given the nature of the radio emission from M22, Strader and his colleagues think the black holes are probably swallowing gas at very low rates. In that case, the disk won’t be as important a source of light as the jets, which emit both in radio and X-ray but put out far less X-ray emission than the disk. Astronomers don’t understand exactly how the jet’s radio and X-ray emission relate to each other, but they do know that, given a certain amount of radio emission, a certain amount of X-rays should appear, too.

But neither black hole shows up in archival observations from the Chandra X-ray Observatory. The authors suggest that could be because black holes forming in the dense globular cluster environment can weigh as much as 20 Suns, roughly twice that of a typical stellar-mass black hole found elsewhere in the galaxy. If the black holes are more massive, they’ll emit less in X-rays than in radio, thus escaping detection by Chandra.

Two black holes might not sound like a lot in the grand scheme of things, but here’s the kicker. We’re only seeing these black holes because they’re accreting gas, and that’s only because they’re lucky enough to have a stellar companion contributing gas to their diet. While these two black holes eat their meager feast, there might be many more fasting in the cluster around them.

To come up with an estimate, astronomers have to know how many black holes are likely to end up in close orbit with another star, and that’s a problem that’s not yet well understood. So the authors settle for a broad estimate: somewhere between 5 and 100 black holes might lurk in the swarming core of M22. That should give theorists something to think about.