A massive galaxy similar to our own Milky Way spotted shockingly early in the universe’s history is challenging astrophysicists’ understanding of galaxy formation. Witnessed just 1.5 billion years after the big bang, when the universe was some 10 percent of its current age, the spinning disk of gas and stars is the earliest of its type ever identified. And it provides strong evidence that some of the first galaxies got off to a cold start.

In standard formation models, galaxies coalesce as gas collects in and around diffuse “halos” of dark matter. All that gas becomes extremely hot as it funnels down into the heart of the newborn galaxy, and it must take time to cool down before it can begin forming stars. In contrast, more recent simulations suggest that gas flowing into young galaxies along long filaments of dark matter can remain relatively cool, allowing star formation to begin sooner. These “cold start” galaxies should form spiral-like disks that resemble the Milky Way.

So far most of the early galaxies observers have managed to identify have been irregular blobs without disks, their shapes distorted, and their gas heated, by repeated collisions with protogalaxies. Astronomers have indeed found a handful of disk galaxies from the first few billion years of the universe’s history. But some researchers argue that these objects are old enough for their gas to have already cooled down, making their origins indefinite.

This particular disk galaxy, however, defies such objections. “We found a galaxy that has a lot of cold gas in it,” says Marcel Neeleman, an astronomer at Max Planck Institute for Astronomy in Heidelberg, Germany, and first author of a study reporting the observations, which was published in the May 21 issue of the journal Nature. “If it had formed through hot-mode accretion, it wouldn’t be there.”

Coral Wheeler, an astronomer who studies galaxy evolution at the University of California, San Diego, agrees. The galaxy provides “very strong evidence of cold-mode accretion,” she says. (Wheeler was not part of the paper.)

Neeleman and his colleagues claim that the new finding means that most of the first generation of galaxies formed through either cold-mode accretion or collisions with other young galaxies.

Hunting Shadows

Researchers have long argued over whether gas pouring into the earliest galaxies was hot or cold. Simulations favor cold gas, but skeptics have raised questions about the validity of those virtual conclusions. And they have done so for good reason: by necessity, those models have simplified many of a galaxy’s most salient environmental effects, such as feedback processes from supernovae and black holes that could heat otherwise cool gas.

“There’s been a controversy about this over the last couple decades now,” says Ryan Trainor, an astrophysicist at Franklin & Marshall College, who was not involved in the Nature study. One of the challenges of hunting for early galaxies is the need for targets that are big and bright enough to be seen across immense cosmic distances. As a result, the most luminous objects are the ones most likely to be observed. To overcome this bias, Neeleman and his colleagues decided to utilize a method pioneered by the late astronomer Arthur Wolfe. Using the Atacama Large Millimeter/Submillimeter Array (ALMA) in Chile, they hunted for galaxies in front of quasars, the brightest known objects in the universe. As light from a quasar passes through a foreground galaxy, the gas from that galaxy absorbs some of the light, creating what Neeleman calls “shadows.”

By studying the shadows, or absorption lines, with ALMA, the astronomers could track the rotating motion of the dim gas of the galaxy DLA0817g, which they discovered in 2017. They nicknamed it the “Wolfe Disk” in honor of the team members’ former adviser and colleague. Follow-up observations with the Hubble Space Telescope revealed some of the galaxy’s brightest stars, which the scientists used to estimate that the Wolfe Disk is churning out an average of 16 sun-sized stars each year. Hubble’s scrutiny also revealed that the gas blocking the quasar came not from the heart of DLA0817g but from the galaxy’s outer edges—a region where gas would be expected to thin out rather than thicken. The researchers suspect what they are seeing is one of the dark matter filaments funneling gas into the Wolfe Disk.

“We can’t prove it’s a filament, but it’s well beyond the star-forming region of the galaxy,” says team member and study co-author J. Xavier Prochaska, an astronomer at the University of California, Santa Cruz.

By using quasars, the team hoped to overcome the observing bias faced by previous studies. To some degree, they were successful. “You probably end up with a fairer sampling of the galaxy population this way,” says Alfred Tiley, an astronomer at the University of Western Australia. Tiley, who was not involved in the research, authored an accompanying commentary about it in Nature.

Not everyone is convinced. Trainor thinks Neeleman and his colleagues’ new method avoids the bias of brightness but may come with its own prejudices. “Their technique is biased toward finding stable rotating disks,” he says. The extended disks created by cool galaxies are more likely to obscure a quasar than a more compact galaxy might. “It’s like throwing darts at a dartboard,” Trainor says. “The larger dartboard is more likely going to get hit.” That analogy does not diminish the technique, which he calls “a really useful and complementary tool.”

While Prochaska agrees that larger galaxies are more likely to block quasars, he argues that the Wolfe Disk’s extended gas in front of the background quasar does not necessarily have a bearing on the galaxy’s structure. The large distribution of gas around a quasar-blocking object could come from a spheroidal shell of gas around the galaxy or from extended filaments funneling gas into it.

Trainor also questions how common galaxies like the Wolfe Disk might be in the early universe. He is not convinced that a single galaxy is enough to demonstrate that cold accretion dominated early galaxy formation. But new galaxies may be uncovered soon. Neeleman’s team plans to continue using ALMA to study quasar-shadowed galaxies in hopes of finding more.

“It’s clear now that you can do this in a subset of cases very early on,” Prochaska says of cold-mode accretion. “We’re all a bit surprised.”