Astronomers have identified a cloud of hydrogen gas roughly 2 billion times the mass of the Sun that appears to have no stars at all, making it one of the strongest candidates yet for a long-sought "dark galaxy." The object, AGC 322753, was picked up independently by two large radio surveys and described in a study in the Monthly Notices of the Royal Astronomical Society by researchers at Swinburne University of Technology in Australia.

Deep optical images show nothing at the cloud's position. The team calculated that any stars present can add up to no more than about 50 million solar masses, which means the gas outweighs the stars by a factor of at least 40. In a typical galaxy like the Milky Way, stars far outweigh the hydrogen gas between them.

The finding matters because dark galaxies are a key prediction of the leading model of how the universe built its structure, yet almost none have been confirmed. A verified example would give scientists a rare look at dark matter halos that never managed to light up with stars.

A Cloud Seen Twice and Spinning

AGC 322753 first appeared in ALFALFA, a survey run on the Arecibo Observatory in Puerto Rico before the telescope collapsed in 2020. It showed up again in FASHI, an all-sky survey by China's 500-meter FAST radio telescope. Both detected the 21-centimeter radio signal that neutral hydrogen gives off.

Two independent detections from different instruments make it much less likely that the signal is radio interference or an equipment glitch, as Daily Galaxy noted in its coverage of the study. The surveys place the object at a distance of about 95 megaparsecs, or roughly 310 million light-years.

The radio signal has two peaks, a pattern typically produced when gas on one side of a disc moves toward us and gas on the other side moves away. That suggests the cloud is rotating like a galaxy rather than drifting as a shapeless blob. From that rotation, the team estimated a total mass of at least 5.5 billion solar masses, more than the visible gas alone, hinting at dark matter.

An earlier catalog had matched the radio source to a visible object, but the Swinburne team showed that object is a foreground star in our own galaxy and lies outside the radio beam.

A Neighbor Too Far Away to Explain It

The usual explanation for starless hydrogen clouds is that they were ripped from a nearby galaxy during a close encounter, leaving a tidal tail of gas. AGC 322753's nearest known galaxies, PGC 069656 and UGC 12179, sit about 290,000 to 330,000 parsecs away in projection, roughly 950,000 light-years or more.

Tidal tails usually stretch about half that distance. The cloud would also rank among the most massive tidal features ever found, and its rotating profile looks like a galaxy rather than torn debris. Those three points led the authors, led by Marco Monaci, to argue against a tidal origin.

They are careful not to close the case, writing "While we cannot exclude a tidal tail scenario in principle," the evidence favors a dark galaxy. The object also is not completely isolated, the paper notes.

Predicted by Theory, Almost Never Confirmed

The standard cold dark matter model predicts that the universe should contain far more small dark matter halos than large ones. In many low-mass halos, theory says star formation should be heavily suppressed, especially after the early universe was reheated by the first stars and galaxies. Some of those halos may hold gas but never form stars.

Proving one exists has been extremely difficult — faint stars can hide below the sensitivity of even deep images, and many candidates have collapsed under follow-up. A FAST candidate called FAST J0139+4328, found in 2023, turned out to host a faint stellar system of about 7 million solar masses. A team from Russia's Special Astrophysical Observatory and Belgrade Observatory concluded that "FAST J0139+4328 cannot be classified as a 'dark galaxy,'"according to the Special Astrophysical Observatory.

The most convincing starless object so far is much smaller — Cloud-9, a gas cloud about 14 million light-years away near the galaxy M94, was confirmed by Hubble to contain no stars and holds only about 1 million solar masses of hydrogen, according to NASA. Its principal investigator, Alejandro Benítez-Llambay, said "In science, we usually learn more from the failures than from the successes."

Faint Stars and Other Open Questions

AGC 322753 holds roughly 1,000 times more hydrogen than theory expects for a truly starless halo. That mismatch is one reason the authors suspect it may contain a very faint population of stars, which would make it an extremely gas-rich but dim galaxy rather than a fully dark one.

The current evidence is limited. Both radio surveys had beams several arcminutes wide, too coarse to map the gas in detail, and the optical limit comes from the DESI Legacy Imaging Surveys rather than a dedicated deep exposure. The team recommends high-resolution hydrogen mapping, for example with the Very Large Array in New Mexico, along with deeper multiband optical imaging.

The same group has cataloged about 70 other dark galaxy candidates in FASHI data within about 160 million light-years, posted as a preprint on arXiv that has not yet been peer reviewed. If several hold up, they could test how well dark matter models predict the number of small, starless halos.

The bottom line: AGC 322753 is a promising but unconfirmed dark galaxy, seen by two telescopes and apparently rotating with no visible stars. Sharper radio maps and deeper images will decide whether it is truly dark, faintly starry or a stray piece of a distant neighbor.

What Readers Want to Know

What is a dark galaxy? A dark matter halo that holds gas but formed few or no stars, making it nearly invisible to optical telescopes.

What did astronomers find? AGC 322753, a hydrogen cloud of about 2 billion solar masses with no detected stars, seen by both the Arecibo ALFALFA and FAST FASHI surveys.

Is it confirmed as a dark galaxy? No. The authors say faint stars or a tidal origin cannot be fully ruled out.

How far away is it? About 95 megaparsecs, or roughly 310 million light-years.

Why are dark galaxies so hard to confirm? Faint stars can hide below image limits, radio interference can mimic signals, and gas torn from nearby galaxies can look similar.

What happens next? The team recommends sharper radio mapping, such as with the Very Large Array, and deeper optical imaging.