Recently an international team of astronomers traveled back in time to when our universe was just 1.8 billion years old. They did not go directly, of course, but settled for the next best thing: gathering 17 hours’ worth of starlight from a single small patch of the distant cosmos with the Large Binocular Telescope Observatory atop Mount Graham in southeastern Arizona. Such clock-rewinding virtual voyages are routine in astronomy—light’s finite speed ensures that the deeper into space you see, the further back in time you gaze. And many observatories around the globe can gather faint photons from ancient skies. But this particular cosmic jaunt concerned something special—even disturbing: an abnormally hefty elliptical galaxy dubbed C1-23152. This egg-shaped aggregation of stars is so outsize that it defies conventional models of its origins. Simply put, C1-23152 seems to be too big to fit the early universe.

It is thought that the first galaxies were relatively minuscule, clumping together from smaller building blocks bit by bit and only reaching gargantuan sizes after billions of years of growth. Boasting an estimated 200 billion solar-mass stars, C1-23152 tips and then overturns the scales for this scenario. And it is not alone. Over the past decade, astronomers have discovered several very ancient, very big galactic behemoths. In 2017, for example, a pair of extremely large galaxies—one capable of churning out 2,900 solar masses of stars per year—were found to exist less than 800 million years after the big bang. In 2019 a family of 39 huge galaxies—each a star factory manufacturing perhaps 200 solar-mass stars per year—were found zipping through the universe within two billion years of its birth.

Does this ever increasing number of venerable vast objects threaten to bring down the traditional model of galaxy formation? “The trick here is: How many do you have?” says Marcel Neeleman, an astronomer at the Max Planck Institute for Astronomy in Heidelberg, Germany, who was not involved with the new study. A handful will not matter; the universe is big enough that odd things will crop up every now and then. But if future, increasingly advanced telescopes manage to find far more of them, then perhaps these colossal galaxies from the universe’s childhood may break our understanding of the cosmos.

A Long Time Ago in a Galaxy Far, Far Away

What has become the widely accepted model of galaxy formation is largely gleaned from simulations of cosmic evolution that reproduce our observations of the local universe—the stuff we can see near the Milky Way.

After the big bang, the cosmos expanded and stretched out fairly evenly in all directions. But, Neeleman says, you get “tiny density variations in the fabric of the universe.” These variations are home to clumps of dark matter, a substance that emits little, if any, electromagnetic radiation. As such, dark matter has yet to be directly detected, but observations of galaxies indicate that this invisible mass produces its own gravitational pull. That means that these dark matter clumps attract “ordinary” matter (the stuff we humans can detect and interact with), most of which is gas. The gas tumbles into these gravity wells and squashes together to trigger star formation. More matter continues to tumble into these ever expanding wells—called dark matter “halos” by astronomers—gradually forming bigger and bigger structures over the 13.8-billion-year lifetime of the universe. This process should more or less create the distribution of galaxies we see today, says Paolo Saracco, an astronomer at Italy’s National Institute for Astrophysics and the lead author of a study reporting the recent observations of C1-23152.

That is why ancient massive galaxies are problematic. “For our current understanding of galaxy formation, we sort of built on the galaxies we knew at the time,” says Coral Wheeler, an astronomer at California State Polytechnic University, Pomona, who was not involved with the new study. These galaxies did not include the very old, small or big ones. Looking further back in time with increasingly powerful telescopes began to reveal these apparent outliers. And as the tally of anomalous entities rose, astronomers started wondering if their models needed to expand to make room for them or if those models would buckle and break under the strain.

As reported in the Astrophysical Journal in December 2020, Saracco’s team managed to extract some juicy details out of C1-23152. Light from far-distant cosmic regions is stretched by the expanding universe as it travels to Earth. The more it is stretched, the greater its shift towards the longer-wavelength “redder” section of the electromagnetic spectrum. This “redshift” of C1-23152’s starlight indicates that it appeared 12 billion years ago, way back in the universe's youth. The fact that this galaxy is both ancient and massive alone is problematic enough for traditional slowly-but-surely models of galaxy formation. But it did not just appear fully formed. Saracco and his team’s real breakthrough was to trace C1-23152’s history of star formation from across the universe.

The key to that breakthrough was seeing the giant galaxy’s spectrum—a rainbowlike measurement of the various wavelengths, or colors, that an object emits or absorbs. Particular color combinations distinguish specific elements, which means this spectral symphony can be used to determine the composition of a galaxy’s stars. Using that power, Saracco says, “for the first time, we derived, with very good accuracy, the mean age of the stellar population inside [C1-23152] and the time necessary to form those stars.”

The number of elements in C1-23152 that were found to be heavier than hydrogen and helium—which astronomers collectively refer to as “metals”—hinted at its strangeness. Metals are produced by star formation, which jettisons them into a galaxy’s interstellar medium through supernovae—making them available for next-generation stars to use. More metals equal more cycles of star formation, and it took present-day massive galaxies many billions of years to become metal-rich. C1-23152’s spectrum revealed the galaxy to be a veritable metal bonanza back in its early days, which means it made a lot of stars very rapidly not long after it first formed.

How rapidly? The spectral features of stars can answer that question, too, because they reveal which ones have elements typical of younger or older stars. The youngest stars in C1-23152 are roughly 150 million years old. The most ancient are about 600 million years old. That means the galaxy made some 200 billion solar masses in just a half-billion years—a rate of 450 stars per year, more than one per day. The figure is almost 300 times greater than estimates of the Milky Way’s current output. If most galaxies are slow-burning log fires, with new flames popping up every so often, C1-23152 is a gasoline-soaked bonfire.

C1-23152 and its similar cousins present astronomers with a potentially model-breaking conundrum: How can massive galaxies be assembled and set alight so quickly so early on? For now the answer, in short, is that they can’t.

Growing the Universe in a Box

For some time, simulations have struggled to grow these ginormous galaxies. But that does not mean they simply cannot do so. Instead the trouble may lie in how they are programmed.

“When you run a simulation, there’s a trade-off between how big of a volume you want to simulate and how much detail you can simulate because of the computer power you have or don’t have,” says Ben Forrest, an astronomer at the University of California, Riverside, and a co-author of the new study. If these ancient massive galaxies are rare, perhaps we are not using big enough boxes to give one the chance to pop up. “Maybe some of the simulations aren’t really covering enough volume,” he says.

Quickly tweaking them to spawn mega galaxies from the early eras of cosmic time is not easy either. “It takes a long time to rerun them. If you want to change something, you’ve got to be pretty sure that’s right and that’s what you want to do,” Forrest says.

Some of the latest iterations of simulations, with better data and computing power, do predict these massive galaxies to exist in small numbers at early times, he adds. But unlike what is being observed in reality, they tend to still be making stars. Ancient galaxies, including C1-23152, abruptly shut off star formation after a productive peak—either because they run out of hydrogen and helium fuel or because the radiation shooting out from fresh crops of stars and other overzealous astrophysical sources cooks that gas and blasts it out of reach. Clearly, some ingredients are still missing from our virtual recipes, so we cannot rely on them for an explanation yet.

Scientists have found clues elsewhere that may account for these ancient mega galaxies. Anastasia Fialkov, a cosmologist at the University of Cambridge, who was not involved with the latest work, says that, unlike full-blown simulations, analytic physics calculations can “take into account the whole volume of the universe.” And they suggest that a small number of dark matter halos capable of initiating star formation show up just 40 million years after the big bang.

That time is significantly earlier than the majority of dark matter halos that turn up later on in the youthful epochs of the universe—those thought to be responsible for seeding much of the galaxies we see today. Instead the halos that appeared 40 million years after the big bang would have been capable of seeding the beginnings of the ancient massive galaxies that would eventually become detectable via our telescopes. The early universe was also denser, Wheeler notes. That would make scooping up star-making hydrogen and helium around these primordial dark matter halos, and eventually galaxies, fairly effortless.

Another option, Neeleman says, is that a combination of things could have occurred. Rare hyperdense pockets of the universe would permit multigalaxy mergers very early on, while streams funneling gas into the hearts of galaxies could supercharge star formation.

In any event, the emergence of huge ancient galaxies is more easily explained if dark matter is cold. Here, “cold” means the dark matter moves relatively slowly. “Hot” dark matter would move at velocities approaching the speed of light. Generally speaking, the colder the dark matter, the easier it can condense into galaxy-seeding halos. This assumption may not necessarily be correct, but “cold dark matter is the simplest dark matter scenario that works,” Fialkov says.

It is unclear which amalgam of these events, if any, best explains C1-23152’s origins and evolution, let alone its colossal cousins. “This is not a special corner of the universe” we are looking at, Saracco says. But, importantly, nothing here threatens to overthrow the traditional slowly-but-surely model of galaxy formation, he says. These ancient, massive galaxies just represent another pathway for galaxies to take.

Back to the Future

The traditional model survives for now but only, in part, because few of these massive galaxies have been found. “We’re dealing with small-number statistics,” Forrest says. Scientists do not have a good grasp of the true amount of the behemoths, however. Until that changes, understanding what impact they have on our cosmic comprehension and how galaxies evolve in different ways will remain ambiguous.

Perhaps we have already seen many more of these old mega galaxies than we yet realize. For detailed studies, our telescopes are often drawn to the brightest massive but burnt out galaxies before their nature is revealed. Astronomers have spotted fainter objects with otherwise similar characteristics hanging about in the early universe, however, says Stijn Wuyts, an astronomer at the University of Bath in England, who was not involved with the recent work. They could turn out to be merely less massive galaxies or yet more ancient massive ones observed long after their star-forming prime. Are these objects dimmer candles closer to home or vast pyres farther afield?

As ever, more data are required—. And several upcoming telescopes will aid us in this time-traveling galactic census.

First, suspicious bright splotches in the distant past need to be spotted. “If you want to get a bunch of candidates, then a wide field of view is great,” Forrest says. The Nancy Grace Roman Space Telescope, formerly known as WFIRST and currently targeted for a 2025 launch, will have a field of view equivalent to 100 Hubble Space Telescopes: its wide, sensitive eyes will see plenty of possible ancient massive galaxies.

Those candidates will then need to be forensically examined by looking at their various spectra to determine their properties and confirm they are indeed such galaxies and not imposters. “Ideally, you want a really big telescope,” Forrest says. “That gives you more collecting area—it’s a bigger bucket for photons to go into from an object.” Hawaii’s Thirty Meter Telescope could be suitable if it is built, and the Extremely Large Telescope could fit the bill as well. The James Webb Space Telescope—which is finally launching this October after an abundance of delays—should work well, too. “It’s not as big,” Forrest says. “The bucket for the photons is a little bit smaller, but then you don’t have to look through the atmosphere,” so there is less interference to deal with.

Saracco is particularly excited for these upcoming next-generation magnifying glasses because they will do more than merely finding extremely distant objects. “We will be able to observe inside [a] galaxy, at single star-forming regions,” he says. In other words, instead of a blurry picture of a galaxy’s bulk characteristics, astronomers will get a more granular view—the difference between a rough sketch and a detailed painting—opening up a new chapter in our understanding of how galaxies form.

Until this help arrives, this scientific field will remain in its infancy. “There’s so much uncertainty that goes into galaxy formation,” Wheeler says.

It can be unnerving to chase monsters in the dark. They threaten the dogmas of the era, forcing us to expand our earlier models to fit them. And if those models stretch to the point of breaking, that’s okay. “We want to challenge, in some way, the model,” Wheeler says. “When things don’t match, that’s when it gets interesting.”