27 SEP 2026 — Astronomers have found a planet orbiting a red dwarf, the most common kind of star in the galaxy, in the opposite direction to the star's spin. No backwards orbit had been measured around a red dwarf before.
The planet, GJ 3090 b, is slightly more than twice Earth's size and lies about 73 light-years away. The result was published on 21 September, led by Yann Carteret of the University of Geneva.
What was measured
Most planets orbit in roughly the same direction their star rotates, because both formed from the same spinning disk of gas and dust. The paper puts the angle between GJ 3090 b's orbit and its star's spin axis at 136 degrees, with an uncertainty of 24 degrees above and 18 below. Anything beyond 90 degrees is retrograde.
The team rules out a polar orbit, one passing over the star's poles at exactly 90 degrees, at 2.5 sigma. That is meaningful but short of the certainty physicists usually ask for in a discovery, and the paper states the figure rather than rounding it up.
How you measure a planet's direction
The technique relies on the planet crossing in front of its star. As the star rotates, one side moves towards us and the other away, shifting its light very slightly towards blue and red. A planet crossing the face blocks a little of each side in turn, and the order in which it does so shows which way it is travelling.
The team observed six such transits with NIRPS, a near-infrared spectrograph on the European Southern Observatory's 3.6-metre telescope at La Silla in Chile, and four of them also with the HARPS instrument. Red dwarfs are faint in visible light and brighter in the infrared; that is why NIRPS was built.
One instrument carries the result
The final result rests on NIRPS alone. The authors write that they discarded one NIRPS visit and the HARPS data for the final fit. They explain why. The choice is disclosed rather than buried, but the measurement still comes from five nights on one instrument.
Why it is hard to explain
The usual explanation for a tilted or backwards orbit is a heavy neighbour: a giant planet or companion star whose gravity drags the smaller planet's orbit over time. GJ 3090 has no known neighbour like that.
The paper excludes a companion star within about 3 astronomical units, anything heavier than Jupiter within about 15, and anything above 13 Jupiter masses within about 100. Those limits assume any companion orbits in the same plane as the planet, and they weaken if it does not. The system also has at least one other confirmed planet, and a backwards orbit in a multi-planet system without a massive companion is itself a first, according to the authors.
"The absence of a massive companion to explain this singular orbit will lead us to study other hypotheses," Vincent Bourrier of the University of Geneva said in the university's release.
The proposed answer, and its hedging
The authors propose that the star gathered a second disk of material late in its life, tilted away from its own spin, and that the planets formed or moved within it. With a late migration through the disk, they call this the most likely mechanism. That is a proposal, and the paper says further observation is needed to test it.
Queen Mary University of London's release says the finding challenges ideas about how planets form. The paper itself is more careful: it describes its results as disfavouring one explanation and pointing towards another.
The release also calls Andrew Winter of Queen Mary a lead author. He is the fourth of 59 authors. The study was led by Carteret and the Geneva group.
View author profile → · Editorial policy
