“This is a tale of a failed galaxy,” Alejandro Benitez-Llambay. The object known as Cloud 9 is unusual in a universe in which stars tend to signal the location where the matter has become concentrated, except that it lacks any detectable stars whatsoever.

The system is about 14 million light years away, at the periphery of one of the spiral galaxies Messier 94 (M94). It appears as a small, almost spherical reservoir of neutral hydrogen, sitting inside a significantly larger dark matter halo, a configuration much predicted in theory, but typically not available to observations of structure normally, due to the absence of the usual tracers of structure, starlight and stellar debris.
Cloud 9 is a Reionization-Limited H I Cloud (RELHIC): a dark-matter-dominated halo that has some primordial gas but has never attained the level of star formation. This label is important in that it identifies a particular bottleneck in the assembly of galaxies that is where gravity can accumulate the material, but the environment and internal conditions prevent the same material to cool, fragment and be ignited into stars. The hydrogen itself is detected as neutral H I that can be mapped with the help of its radio emission even when the sky seems empty in the optical sky. In Cloud 9, the core of the neutral hydrogen measures approximately 4,900 light years and the amount of gas in the core is approximately 1 million solar masses. Then, assuming that the halo is held together by gas pressure, it takes approximately 5 billion solar masses of dark matter to hold it together. It is that ratio that is the gas a thin coat over an enormous invisible backbone, which forms the kind of unusual laboratory in which to learn the behavior of those dark matter structures in the absence of stellar feedback.
“This cloud is a window into the dark universe,” said Andrew Fox of AURA/STScI for ESA. “We know from theory that most of the mass in the universe is expected to be dark matter, but it’s difficult to detect this dark material because it doesn’t emit light. Cloud-9 gives us a rare look at a dark-matter-dominated cloud.”
The dark matter is generally defined as the massive portion of the cosmic structure that substantiates the way the normal matter gathers as a result of gravity. But the very thing which renders it fundamental, its want of intercourse with light, renders it hard to interrogate. Cloud 9 has provided a way out: rather than deduce dark matter by the motions of stars, the system enables astronomers to observe a dark halo in the motions and structure of the gas contained within it, and the number of competing processes is lessened to model.
The process of detection also shows the capability of multi-instrument astronomy to distinguish between “there is something” from “there are stars.” The neutral hydrogen signal was originally banners in a radio survey with the Five-hundred-metre Aperture Spherical Telescope (FAST), which was followed up by subsequent radio measurements with U.S. facilities, including the Green Bank Telescope and the Very Large Array. Such measurements put a small H I cloud at M94, though radio measurements alone could not conclusively eliminate such a faint dwarf galaxy lurking below the survey limits.
There Hubble was the discriminator. With the Advanced Camera for Surveys, the team pointed at the area of the greatest radio emission and looked to determine whether there was any detected stellar-population within the cloud.“Before we used Hubble, you could argue that this is a faint dwarf galaxy that we could not see with ground-based telescopes. They just didn’t go deep enough in sensitivity to uncover stars,” said Gagandeep Anand of STScI. “But with Hubble’s Advanced Camera for Surveys, we’re able to nail down that there’s nothing there.”
“nothing there” is a formidable restriction in practice. Smaller galaxies are hard to notice due to their dimness, diffuse nature, and due to overcrowding with foreground stars or due to the background galaxies. The importance of Cloud 9 is based on an observed conclusion instead of a gleaming signature: in case there had been stars on the scales that should be seen as a typical dwarf galaxy within a halo of such great mass, the deep imaging of Hubble would have found them.
Cloud 9 too is not easily mistaken by the extended and discontinuous hydrogen clouds that have long been observed to be around the Milky Way. Those structures can be deformed by the local feedback, tidal interactions and large scale flows. Cloud 9 is relatively small and almost spherical and exists in a regime where theory suggests a boundary between halos which can form stars and dark ones. That boundary is among the most important calibration points in galaxy formation: it determines the abundance of low-mass halos that is supposed to be present in the environment, the prevalence of visible galaxies in those hosts, and the extent to which simulations can relate an unseen dark-matter population with the number that can be counted by telescopes.
Its positioning close to M94 provides another feature of an engineering benefit: a distance anchor which can be made to work. Having that association, Cloud 9 is no longer ambiguous like isolated candidates whose distances, and hence masses, are more difficult to obtain. There are small deviations observed in the finer radio mapping implying that the gas is reacting to its surroundings within the greater galaxy supporting the notion that they are physically related and not merely by coincidence.
“Among our galactic neighbors, there might be a few abandoned houses out there,” said Rachael Beaton of STScI. The name embodies the larger connotation: in case it becomes possible to test by deep optical imaging that a single RELHIC exists, other compact H I clouds in the halo of nearby galaxies should become targets of interest; where the contributions of the dark matter can be studied without the opaque “weather” the generations of stars can create.
That search space is predicted to increase dramatically in the future with wide surveys, such as that offered by the LSST Camera in Rubin Observatory, which will be able to conduct widespread monitoring of the sky. In the case of the important signal, radio hydrogen, more detailed and extensive optical catalogs can be used to remove spurious matches and narrow the boundaries of star-formation. In the case of Cloud 9, the same process of radio discovery and Hubble confirmation characterizes the template: detect the gas, and then the absence of starlight decisively.
Cloud 9 is so far one of those occasional instances when heavenly plan is revealed with minimal embellishment. Having a dark matter halo (in billions of solar masses) and a halo center of hydrogen (in millions), it demonstrates the extent to which a system can approach the formation of a galaxy but never quite forms one.

