Somewhere between 50 and 200 million years ago, in three stellar nurseries not far from our corner of the Milky Way, at least six small red dwarf stars did something violent: they swallowed rocky planets, possibly several at a time, and the chemical evidence is still glowing in their atmospheres today.
That is the conclusion of a study published in Monthly Notices of the Royal Astronomical Society by R. D. Jeffries and colleagues, who found that six early M-dwarf stars in three nearby open clusters carry far more lithium than any standard model of stellar evolution can explain. The most natural interpretation: each star recently ingested between 3 and 10 Earth masses of rocky planetary material, and the lithium from those devoured worlds has not yet been destroyed.
The finding matters because M-dwarfs are not exotic objects. They account for roughly three out of every four stars in the galaxy and host the majority of known rocky, potentially habitable exoplanets. If even a small percentage of these stars consume their inner planets during the first few hundred million years, the implications ripple outward to every estimate of how many Earth-like worlds persist long enough for life to gain a foothold.
Lithium as a forensic tool
Lithium is cosmically fragile. Stars destroy it through nuclear reactions at temperatures found deep inside their convective zones, where churning currents drag surface material downward into regions hot enough to break lithium nuclei apart. In low-mass stars like M-dwarfs, this convective mixing is especially thorough. By the time a young red dwarf settles onto the main sequence and begins stable hydrogen fusion, most of its original lithium should be gone.
That predictable depletion is what makes lithium such a useful forensic marker. When a star suddenly gains a fresh dose by consuming a rocky body, the spike stands out sharply against the depleted background, like finding wet paint on a wall that dried years ago.
Jeffries and his team used high-resolution spectra from the Gaia-ESO Spectroscopic Survey, a major European observing program that cataloged tens of thousands of stars across the southern sky. They measured lithium line strengths in early M-dwarfs within the open clusters NGC 2451a, Blanco 1, and NGC 2516, all young enough that their stars are just reaching the main sequence. For the vast majority of cluster members at similar temperatures, lithium levels matched predictions: low and falling. Six stars broke the pattern, showing lithium abundances several times higher than their neighbors observed with the same instruments under the same conditions.
Reconstructing the meal
To figure out what could produce such enrichment, the team modeled how much rocky material a star would need to swallow to raise its surface lithium to the observed levels. The answer, according to the study and a summary from the University of Exeter, falls between roughly 3 and 10 Earth masses. That range corresponds to super-Earths or small collections of terrestrial planets, exactly the kind of rocky worlds that form readily in the inner regions of protoplanetary disks around M-dwarfs.
The timing fits a known window of planetary chaos. By 50 to 200 million years, the gas and dust disks that birth planets have long since dispersed, but the leftover planetary systems are still dynamically settling. Gravitational interactions between young planets can scatter bodies inward toward the star, especially in the tightly packed architectures common around red dwarfs. Separate modeling work on lithium persistence after engulfment shows that low-mass stars can retain elevated lithium for tens to hundreds of millions of years, particularly if the ingestion occurs after the most vigorous early mixing has subsided. For clusters in this age range, a relatively recent engulfment event would still be visible in spectroscopic surveys conducted in June 2026.
The six outliers represent about 2 to 3 percent of comparable M-dwarfs in the three clusters. That fraction is small but not negligible, and it may undercount the true rate. Some engulfment signatures could have already faded below detection thresholds in older or more massive stars, meaning the real frequency of planetary consumption might be higher than what current data capture.
A pattern across stellar types
The Jeffries result does not stand alone. A 2021 study published in Nature Astronomy by Lorenzo Spina and colleagues examined chemical abundance differences in co-natal pairs of Sun-like stars in wide binaries and found signatures consistent with planetary ingestion in roughly one-quarter of the systems studied. That work suggested planet-eating is not a freak event but a recurring outcome of planetary system evolution, at least for solar-type stars.
Earlier, in the early 2000s, researchers detected the rare isotope lithium-6 in the planet-hosting star HD 82943 and interpreted it as possible evidence of planetary accretion, since lithium-6 is even more easily destroyed than the common lithium-7. That detection, however, was later contested by follow-up analyses that questioned whether the lithium-6 signal was robust, so it remains an open data point rather than a settled case.
Taken together, these studies build a picture in which stars consuming their own rocky planets is a recurring phenomenon across stellar types and ages, not an oddity confined to a single peculiar system.
What the data cannot yet tell us
Six stars across three clusters is a small sample, and the authors themselves frame the result cautiously, posing the lithium enrichment as a question rather than a settled conclusion in their paper’s title. With small-number statistics, a few misclassified objects or unrecognized systematic effects could shift the inferred frequency significantly.
Alternative explanations have not been fully eliminated. During their youth, low-mass stars can experience episodic bursts of accretion from remnant disk material, potentially altering their internal structure and lithium-burning history in ways that mimic the engulfment signature. Unusual magnetic activity or atypical convective histories could also preserve more lithium than standard models predict, though the authors argue these scenarios struggle to explain why only a handful of stars in each cluster show the excess while dozens of comparable neighbors do not.
The modeled mass range of 3 to 10 Earth masses depends on assumptions about how efficiently lithium from accreted material mixes into the stellar envelope. Different prescriptions for convective mixing can shift the inferred mass up or down, and uncertainties in stellar radii and internal temperatures propagate into the final estimates.
None of the six lithium-rich stars are currently known to host surviving planets, and confirming or ruling out remaining planetary companions would strengthen the engulfment interpretation. Instruments like the James Webb Space Telescope or ground-based radial velocity spectrographs could, in principle, search for leftover planets around these stars, testing whether the systems look stripped compared to lithium-normal M-dwarfs of the same age.
What it means for the search for habitable worlds
Most strategies for finding life beyond Earth focus heavily on M-dwarf systems, for practical reasons: red dwarfs are abundant, their small size makes transiting planets easier to detect, and their habitable zones sit close in, producing frequent transits and stronger radial velocity signals. The TRAPPIST-1 system, with seven rocky planets orbiting an ultracool dwarf star, is the poster child of this approach.
But if a measurable fraction of M-dwarfs consume inner rocky planets during their first couple hundred million years, the pool of long-lived habitable worlds around these stars may be smaller than optimistic estimates suggest. A planet that gets scattered into its host star at 100 million years never gets the chance to develop an atmosphere, accumulate surface water, or host biology. The 2 to 3 percent engulfment rate observed in these clusters is a lower bound on how many systems experience destructive instability, since not every scattered planet ends up in the star; some get ejected into interstellar space instead.
This does not mean M-dwarf habitable zones are empty. Plenty of red dwarfs clearly retain rocky planets well past the chaotic early period. But the new lithium data add a concrete, observationally grounded constraint to models that try to predict how many of those planets survive long enough to matter for habitability.
For now, the six lithium-rich M-dwarfs in NGC 2451a, Blanco 1, and NGC 2516 are best understood as strong candidates for recent planetary engulfment rather than conclusive proof. Larger cluster surveys, higher-precision spectroscopy, and refined models of lithium evolution in low-mass stars should clarify whether these red dwarfs really did consume multiple Earths, and how many rocky worlds across the galaxy share that fate.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
