Hubble Traces Milky Way's Oldest Known Merger
Data from the Hubble Space Telescope and Europe's Gaia mission show the Milky Way absorbed the dwarf galaxy LKH about 11.8 billion years ago. The study appears in Nature Astronomy.

New data from the Hubble Space Telescope and Europe’s Gaia mission show that the Milky Way absorbed a dwarf galaxy, now named LKH, roughly 11.8 billion years ago, only about two billion years after the Big Bang. The study, published August 17 in Nature Astronomy, was led by Davide Massari of the Astrophysics and Space Science Observatory in Bologna. The finding pushes the documented history of our galaxy 1.8 billion years further back than anything previously confirmed.
The Milky Way is now a spiral galaxy with several hundred billion stars, but it did not start out that way. It grew in two ways: by forming new stars from its own gas clouds, and by absorbing stars, gas and dark matter from galaxies it collided with. Two major mergers had already been documented, and the new study adds a third, and the oldest yet.
- A dwarf galaxy in the constellation Sagittarius, whose collision with the Milky Way began more than six billion years ago and is still ongoing.
- Gaia-Sausage-Enceladus, a merger about ten billion years ago that reshaped the disk of our galaxy.
- LKH, a merger about 11.8 billion years ago, confirmed for the first time by this study.
Several smaller mergers occurred between the first two. Earlier simulations and observations had hinted at an even older and larger one, but the details were disputed. That is where the new galaxy’s name comes from: LKH stands for Low-energy-Kraken-Heracles, after three earlier papers that had proposed the idea.
“Our home is the Milky Way galaxy, but we don’t know how our home was built. In this study we reveal where the first significant batch of bricks came from: a dwarf galaxy we call LKH,” Massari said.
By the team’s estimate, LKH carried about 500 million solar masses in stars. At the time, that was a substantial share of the young Milky Way’s total mass, since our galaxy was then much smaller and closer in size to the galaxies it was colliding with.
Why do globular clusters preserve a record of the collision?
Because they are older than almost everything else in the galaxy, and because they carry the chemical signature of the environment where they formed. A globular cluster is a dense, roughly spherical group of anywhere from tens of thousands to millions of stars, and in this line of work it functions as an archaeological site.
The team analyzed Hubble observations of 39 globular clusters within the inner 20,000 light years of the Milky Way, a region of the galaxy where traces of the oldest mergers should still survive. For each cluster they measured a precise age and metallicity, meaning the proportion of elements heavier than helium. Gaia data added precise motion measurements, which let the clusters be sorted by origin.
The researchers had expected the sample to split into two groups: clusters born in the Milky Way itself, and those taken from the Gaia-Sausage-Enceladus structure. A third population turned up instead, older than the absorbed clusters and younger than the homegrown ones, regardless of metal content. That separation is the basis for the conclusion about an earlier merger.
“Thanks to the high resolution and depth of Hubble’s images, we were able to measure the ages and metal content of these clusters with a precision that wasn’t possible before. Combined with measurements from the Gaia mission, that let us isolate a population of globular clusters that stands apart from the rest. These are clusters born in LKH, and they tell us when that galaxy was swallowed and how massive it was,” said Chiara Zerbinati of the University of Bologna, a co-author of the study.
What this means for the picture of how the galaxy formed
Some earlier studies held that the earliest stages of our galaxy’s development were shaped solely by stars formed within it. Massari says stars born in other galaxies now have to be factored in as well. The team plans to continue its work on globular clusters, aiming to map out every major merger in the Milky Way’s history.
“Hubble is observing globular clusters that haven’t been studied before, and that will help us reconstruct mergers that happened far back in the Milky Way’s past,” said Fernando Aguado-Agelet of the University of Vigo and the University of La Laguna in Spain, a co-author of the study.
The limits of the finding should be stated plainly. The collision itself was not observed directly, since it happened nearly 12 billion years ago; it was inferred from the age, chemical makeup and motion of the clusters that survived it. The estimate of 500 million solar masses is a projection, not a measured quantity, and it covers only stars, not the gas and dark matter that LKH may also have brought with it. The sample consists of 39 clusters from the inner galaxy, not the full population of Milky Way globular clusters. The study does not claim that LKH is the last unknown galaxy in our past, nor does it rule out other mergers from the same period that left no trace among the clusters in this sample.


