A Gas Cloud That May Be a Galaxy That Never Switched On

Astronomers report that a cloud called AGC 322753 holds about two billion solar masses of hydrogen gas, seen by two separate radio telescopes, in a spot where deep images show no stars to match ^[2].

A solar mass is the mass of our Sun, so the cloud carries the gas equivalent of two billion Suns. If it is what its discoverers suspect, it is a possible dark galaxy, a structure big enough to count as a galaxy that has barely lit up.

The study is by Marco Monaci, Duncan Forbes, Warrick Couch and Jean Brodie of Swinburne University of Technology's Centre for Astrophysics and Supercomputing ^[2][3]. It has been accepted as a Letter in Monthly Notices of the Royal Astronomical Society, a peer-reviewed journal. The authors posted it to the arXiv preprint server on September 30, 2026 ^[2], and Phys.org's October 26 report lists it as published, with the DOI 10.1093/mnras/stag1874 ^[1].

Why a dark patch matters

Galaxies are supposed to be places where gas turns into stars. A clump of dark matter, the invisible material whose gravity holds galaxies together, pulls gas in. The gas cools and collapses, and stars light up.

A halo that gathers gas and fails to make stars would show how that process stalls. The authors say such objects could help researchers test how galaxy formation is suppressed in low-mass halos, and improve the cosmological simulations that try to reproduce the universe ^[3].

The word "dark" needs care. In this paper it means missing or very faint starlight. The authors do not claim the object is made only of dark matter ^[3].

What the telescopes saw

The team went looking in two blind surveys, which scan the sky for neutral hydrogen, the plain unionised gas that radio telescopes can pick up, without aiming at any chosen target. One, ALFALFA, used the Arecibo telescope in Puerto Rico, now decommissioned. The other, FASHI, uses China's FAST telescope ^[1][3].

Cross-matching the two catalogues produced 13 sources that appear in both. The researchers checked each for nearby galaxies, image quality and optical counterparts, then chose AGC 322753 as the best candidate and set the other 12 aside ^[3].

In both surveys the object shows up at the same recessional velocity, the speed at which it recedes from us, with similar spectral features and a hydrogen mass of about two billion solar masses ^[2][3]. Two different observatories seeing the same thing makes radio interference an unlikely explanation ^[1].

It is a small irony of the field that a survey run on a telescope that no longer exists still turned up something its successor could confirm.

Both surveys also show a signal with two peaks. That shape is what a rotating disk would produce, since gas on one side of the disk moves toward us and gas on the other side moves away, splitting the signal. The line is about 85 kilometres per second wide at half its peak ^[2][3].

The telescopes are single dishes, though, so they cannot draw the gas as an image. Nobody has mapped a disk. Where the two telescopes' observing beams overlap, they do narrow down where the source sits compared with either survey alone ^[3].

The empty image

The strange part is what is missing. The team searched the DESI Legacy Imaging Surveys, a set of deep optical images of the sky, and found no counterpart down to about 27.5 magnitudes per square arcsecond, an extremely faint level of surface brightness ^[2][3].

From that depth the authors estimate an upper limit of about 50 million solar masses in stars. On that estimate, the object has at least 40 times more neutral hydrogen than stellar mass ^[2][3].

The limit is a ceiling, not a measurement. The authors say the hydrogen mass is higher than expected for a completely starless halo, and they allow that AGC 322753 may have a faint stellar component below what the images can detect ^[2][3].

Phys.org reports the object is about 310 million light-years away. The nearest known luminous neighbour lies about 950,000 light-years from it in projected separation, meaning as measured across the sky ^[1].

How it could still turn out ordinary

The researchers name an alternative themselves: the gas could have been pulled out of neighbouring galaxies by gravitational interactions. The reported separation and the gas properties weigh against that, but do not rule it out ^[1].

There is a precedent for caution. Another candidate, FAST J0139+4328, was later reclassified as a faint dwarf galaxy after follow-up imaging and velocity measurements found its stellar counterpart ^[3].

So this is one object and a candidate. It is not evidence for a population of dark galaxies, and nothing in the study claims it is.

The next tests are plain enough. Sharper radio observations could show whether the gas really forms a rotating disk, and deeper optical imaging could uncover hidden stars ^[1].

If the disk is real and the stars stay hidden, a dark patch of radio sky becomes a hard case for the simulations. If the stars turn up, it joins the list of candidates that lost their label.