Galaxies became chemically mature faster than we believed possible.
In the earliest chapters of cosmic time, a galaxy called Gz9p3 — observed as it existed just 510 million years after the Big Bang — has emerged from the James Webb Space Telescope's gaze as something the universe, by our current understanding, should not yet have been capable of producing. Containing several billion stars and bearing the twin-core signature of an ongoing galactic collision, it stands ten times more massive than any comparable galaxy from its era, suggesting that the infant cosmos assembled itself through merger-driven violence at a pace our models have not accounted for. The light arriving from this ancient object, having traveled thirteen billion years to reach us, is quietly rewriting what we thought we knew about the origins of structure in the universe.
- A galaxy that should not yet exist — ten times more massive than its peers from the same cosmic era — has been confirmed by the James Webb Space Telescope, sending a jolt through the field of cosmology.
- Its double-core structure reveals a collision still in progress, a galactic merger frozen mid-impact in thirteen-billion-year-old light, suggesting violent assembly was far more common in the early universe than models predicted.
- Spectroscopic analysis uncovered an unexpectedly large population of ancient stars already enriched with heavy elements, meaning chemical maturity arrived in the cosmos billions of years ahead of schedule.
- Researchers are now confronting the uncomfortable gap between observation and theory — current models of star formation efficiency and galaxy growth simply cannot account for what Gz9p3 represents.
- The findings, published in Nature Astronomy, do not shatter cosmology but demand a significant revision: the early universe was a faster, more efficient, and more turbulent builder than anyone had modeled.
A faint point of light in old Hubble imagery has become one of the most consequential objects in modern astrophysics. The galaxy Gz9p3, seen as it existed just 510 million years after the Big Bang, was recently examined in detail by the James Webb Space Telescope — and what the international Glass Collaboration found there has unsettled long-held assumptions about how the early universe grew.
The galaxy already contained several billion stars at a time when the cosmos was barely getting started. More striking still, it is roughly ten times more massive than any other galaxy astronomers have observed from the same period. Kit Boyett of the University of Melbourne, a member of the research team, described the shift in perspective: what was once a single indistinct point of light is now a detailed object demanding explanation.
The explanation, it seems, is collision. Gz9p3 bears two distinct bright cores — a telltale sign of an ongoing merger between two early galaxies. When galaxies collide, the gravitational disruption floods the system with fresh gas, igniting furious episodes of star formation known as starbursts. This process, familiar from the Milky Way's own history of consuming smaller neighbors, appears to have operated with extraordinary efficiency in the universe's first half-billion years.
Using spectroscopic analysis, the team was able to distinguish younger stars born in the merger from an older stellar population already present — and that older population was far larger than expected. Those ancient stars had already lived, fused lighter elements into heavier ones, and died in supernovae, seeding the early cosmos with metals long before theory suggested such chemical enrichment was possible.
Published in Nature Astronomy in March 2024, the findings do not overturn cosmology — but they insist it be revised. The early universe, it now appears, was a more violent and more efficient architect than we had imagined, building massive, chemically complex galaxies through mergers at rates our models have yet to fully explain.
A speck of light caught by the Hubble Space Telescope has turned out to be something far stranger than anyone expected: one of the most massive galaxies ever found in the infant universe, and evidence that the cosmos assembled itself much faster than our current models allow.
The galaxy, catalogued as Gz9p3, appears to us as it was just 510 million years after the Big Bang—a time when the universe was still in its infancy, before most of what we see around us today had even begun to form. When the James Webb Space Telescope turned its instruments on this object, researchers with the international Glass Collaboration discovered it already contained several billion stars. That alone was startling. But the real shock was the comparison: Gz9p3 is roughly ten times more massive than other galaxies astronomers have observed from the same era. For a universe that was barely half a billion years old, this thing should not exist.
Kit Boyett, a scientist at the University of Melbourne and a member of the research team, described the transformation in understanding. Just a couple of years ago, Gz9p3 was nothing but a single point of light through Hubble's lens. The newer telescope revealed it in detail—and in doing so, forced a reckoning with how we think galaxies grow. The question that has haunted cosmologists for years suddenly became more urgent: how did galaxies become so massive so quickly?
The answer, it turns out, may lie in violence. When Boyett's team examined Gz9p3 more closely using direct imaging, they found something telling in its shape. The galaxy has two bright nuclei—two dense cores—suggesting it was born from the collision of two separate galaxies. The merger, remarkably, appears to still be underway. This is not a finished process but a collision caught mid-impact, frozen in time by the light that has traveled thirteen billion years to reach us.
Galaxy mergers are cosmic engines of creation. When two galaxies collide, the gravitational violence triggers an inflow of fresh gas that ignites what astronomers call a starburst—a frenzy of star formation that can transform a galaxy in a relatively short span of time. Isolated galaxies form stars slowly, eventually running out of fuel. But merging galaxies can reignite the process, drawing in new material and birthing stars at rates that seem almost reckless. The Milky Way itself has grown this way, consuming smaller satellite galaxies over billions of years. In roughly 4.5 billion years, it will collide with Andromeda, and the resulting merger will trigger another burst of star formation after a long period of relative quiet.
What Gz9p3 suggests is that this merger-driven growth was far more important in the early universe than anyone had modeled. The team used spectroscopic analysis—examining the chemical signatures of starlight—to separate the young stars born in the merger from an older population already present in the colliding galaxies. This older population turned out to be much larger than expected. Those ancient stars had already fused hydrogen into helium, and then fused helium into heavier elements like silicon, carbon, and iron. When they died in supernovae, they seeded the early universe with metals—the building blocks for the next generation of stars. The implication is stark: galaxies became chemically mature far faster than theory predicted.
Boyett and his colleagues published their findings in Nature Astronomy on March 7, and the message is clear: the early universe was a more violent, more efficient place than we thought. Galaxies accumulated mass through mergers at rates higher than current models allow. Star formation efficiencies exceeded expectations. The cosmos, in its infancy, was building itself faster than we believed possible. This doesn't mean cosmology is fundamentally broken, Boyett explained, but it does mean astrophysicists must revise their understanding of how quickly galaxies actually formed. The universe we thought we knew is being rewritten by the light of objects that have been traveling toward us since before Earth existed.
Notable Quotes
Just a couple of years ago, Gz9p3 appeared as a single point of light through the Hubble Space Telescope. But by using the JWST we could observe this object as it was 510 million years after the Big Bang.— Kit Boyett, University of Melbourne, Glass Collaboration
Our cosmology isn't necessarily wrong, but our understanding of how quickly galaxies formed probably is, because they are more massive than we ever believed could be possible.— Kit Boyett
The Hearth Conversation Another angle on the story
So this galaxy Gz9p3—it's just a point of light through Hubble, but Webb sees something completely different. What changed?
The resolution. Hubble was looking at it from too far away, essentially. Webb can see fine detail in objects that are incredibly distant and incredibly old. It's like the difference between squinting at a crowd and actually seeing the faces.
And what Webb found was that this thing is ten times more massive than other galaxies from the same period. Why is that such a problem?
Because our models say galaxies shouldn't be able to grow that fast. We have timescales built into our understanding of how stars form, how gas cools, how gravity works. Gz9p3 broke the schedule.
The merger explains it, though. Two galaxies colliding, triggering a starburst. That's not new physics.
No, but the scale is. We knew mergers happened. What we didn't fully appreciate was how dominant they were in the early universe, how efficiently they could pack stars into a galaxy in such a short time. Gz9p3 shows the process was more violent and more productive than we modeled.
You mentioned the older stars in this galaxy—that they had already made heavy elements. Why does that matter?
Because it compresses the timeline even further. Those old stars had to live their whole lives, fuse their fuel, and die. That takes time. Finding a large population of them in a galaxy that's only 510 million years old means the galaxy was already mature when it was still an infant by cosmic standards.
So what happens now? Do we throw out the models?
No, but we rewrite them. We adjust the parameters. We accept that the early universe was messier and more efficient than we thought. And we keep looking for more objects like Gz9p3 to understand just how common this really was.