The radius valley — the gap between super-Earths and sub-Neptunes at about 1.5 to 2 Earth radii — is one of the strongest demographic patterns in the exoplanet population around Sun-like and early M dwarf stars. It appears in Kepler data, persists in TESS data, and has spawned competing theoretical explanations: photoevaporation stripping atmospheres from small rocky cores, core-powered mass loss driving gas escape, or compositional differences between rocky and water-rich planets.
Note (2026-11-14; read against the paper's full text): the essay says photoevaporation and core-powered mass loss both predict that the valley persists around all stellar types, and that its disappearance means neither is operating at the needed scale. The paper says the two mechanisms make consistent predictions for Sun-like stars but that the valley locations they predict diverge toward mid-to-late M dwarfs, and that finding no valley means its survey cannot directly disentangle their relative effects. It describes the dearth of sub-Neptunes as consistent with water-rich pebble accretion models and says the results support the emerging idea that the sub-Neptune population around M dwarfs is composed of water-rich worlds; it does not state that the valley was never a feature of planets. The numbers (8,134 stars, 77 candidates, a peak at 1.25 Earth radii, 5.5 to 1, 1.10 planets per star, hot Jupiter limit 0.012) are the paper's.
Gillis, Cloutier, and Pass (arXiv 2602.23364, February 2026) find that around mid-to-late M dwarfs — the smallest, coolest, most common stars — the radius valley disappears entirely. In a survey of 8,134 such stars observed by TESS, yielding 77 transiting planet candidates, the radius distribution is unimodal. One peak at 1.25 Earth radii. No gap.
The numbers are stark. Super-Earths outnumber sub-Neptunes by 5.5 to one. The cumulative rate is 1.10 planets per star with radii above 1 Earth radius in orbits shorter than 30 days — comparable to rates around early M dwarfs. These are the most prolific small-planet hosts in the galaxy. They just don't produce the bimodal distribution.
The disappearance of the valley constrains the formation mechanism. Photoevaporation models predict the valley should persist around all stellar types, shifting to smaller radii around cooler stars but remaining present. Core-powered mass loss makes similar predictions. Both mechanisms remove gas from rocky cores; the valley marks the transition between cores that kept their envelopes and cores that lost them. If the valley vanishes, neither mechanism is operating at the scale needed to sculpt the distribution.
Water-rich pebble accretion offers a different prediction. Around the lowest-mass stars, the ice line is closer to the star, and the solids available for planet formation are dominated by icy pebbles. Planets form water-rich from the start, without ever acquiring thick hydrogen-helium envelopes that could later be stripped. The resulting planets are intermediate in density — denser than gas-enveloped sub-Neptunes, less dense than bare rock — and cluster at a single characteristic radius set by their water-mass fraction rather than splitting into two populations divided by atmospheric loss.
No hot Jupiters were found — an upper limit of 0.012 per star within 10 days. The smallest stars make small planets, preferentially.
The demographic pattern that defined the exoplanet population around Sun-like stars is not universal. It is conditional on the stellar mass — and therefore on the materials, timescales, and temperatures available for planet formation. The valley was never a feature of planets. It was a feature of certain kinds of stars.