The galaxy that never lit up: a spinning cloud of hydrogen with no stars

Every galaxy in the catalogues earned its place by shining. This one earned its place by refusing to. The object catalogued as AGC 322753 is a cloud of hydrogen holding about two billion times the mass of the sun, and radio observations show it rotating in the orderly way a galaxy rotates, disc, structure and all. Point a deep optical telescope at the same patch of sky and there is, as far as anyone can tell, nothing there. No starlight. A galaxy's worth of raw material, assembled and spinning, that never made the step every other galaxy made. A team led by researchers at Swinburne University has now published the case that it is exactly what it looks like: a dark galaxy, in the journal Monthly Notices of the Royal Astronomical Society.

The evidence was gathered in two acts, years apart, by two different kinds of telescope. Radio surveys came first: the Arecibo telescope's ALFALFA survey and China's FAST telescope both recorded the hydrogen signal, gas glowing faintly at radio wavelengths whether or not any stars exist to light it. That is the trick of hunting dark galaxies. Hydrogen announces itself in radio even in total darkness, so a radio map can show you the ghost of a galaxy that optical astronomy, for all its power, simply cannot see. What the radio maps of AGC 322753 show is not a shapeless cloud. The rotation curve, the way the gas speed varies from centre to edge, is the signature of a settled disc, the same signature ordinary spiral galaxies carry.

Then came the hard part, proving a negative. The team, Monaci, Forbes, Couch, Brodie and colleagues, examined deep optical images of the position and found no matching starlight where a normal galaxy of that mass would be unmissable. A normal galaxy with two billion suns of gas would have turned a good share of it into stars long ago, and stars are not shy. Their absence is the finding, and it is a robust kind of absence, the sort that survives because the instruments that should have seen the light are the same ones that see far fainter things every night.

Why a galaxy would fail to light up is the question that makes the object matter. Galaxies are thought to form inside haloes of dark matter, gathering gas until it gets dense and hot enough to collapse into stars. A halo that spins too fast, or sits in the wrong environment, or gets its gas heated at the wrong moment, might never cross that threshold, leaving its hydrogen spread thin and stable for billions of years. Astronomers have predicted such failures for decades and have a handful of candidates, each argued over, because a faint galaxy and a dark one can look similar until the images get very deep. AGC 322753 is now among the strongest of those candidates, which is another way of saying the argument about it is about to get properly started.

Our call is that dark galaxies will turn out to be common, and that this is the discovery pattern that will prove it. Radio surveys are getting wide and deep enough to catalogue hydrogen wherever it sits, and every settled, spinning, starless disc they certify is a data point the standard story of galaxy formation has to absorb: forming the disc is not the hard part, lighting it is. If most haloes fail the way this one may have, the bright galaxies we have spent a century studying are the exception in the universe, not the rule, and the typical galaxy is one nobody will ever see. AGC 322753 will not settle that on its own. It will do something almost as useful. It gives the hunters a shape to look for, and they now know the dark is full of them.