In contrast to the water-poor planets of the inner Solar System, stochasticity during planetary formation^1,2 and order-of-magnitude deviations in exoplanet volatile contents³ suggest that rocky worlds engulfed in thick volatile ice layers^4,5 are the dominant family of terrestrial analogues^6,7 among the extrasolar planet population. However, the distribution of compositionally Earth-like planets remains insufficiently constrained³, and it is not clear whether the Solar System is a statistical outlier or can be explained by more general planetary formation processes. Here we use numerical models of planet formation, evolution and interior structure to show that a planet’s bulk water fraction and radius are anti-correlated with initial ²⁶Al levels in the planetesimal-based accretion framework. The heat generated by this short-lived radionuclide rapidly dehydrates planetesimals⁸ before their accretion onto larger protoplanets and yields a system-wide correlation^9,10 of planetary bulk water abundances, which, for instance, can explain the lack of a clear orbital trend in the water budgets of the TRAPPIST-1 planets¹¹. Qualitatively, our models suggest two main scenarios for the formation of planetary systems: high-²⁶Al systems, like our Solar System, form small, water-depleted planets, whereas those devoid of ²⁶Al predominantly form ocean worlds. For planets of similar mass, the mean planetary transit radii of the ocean planet population can be up to about 10% larger than for planets from the ²⁶Al-rich formation scenario.

Cracking Up: Transform faults perpendicular to mid-ocean ridges are some of the most prominently visual features on the sea floor. Because they form slowly over thousands of years, the lack of observational data means their mechanism of formation has remained controversial. Taking a numerical modeling approach, Gerya (p. 1047 ) suggests that due to asymmetric growth of the plate boundary, sections of the mid-ocean ridge become unstable and eventually rotate 90°. As the ridge continues to grow, the transform faults continue to develop long after their initiation. This mechanism also explains how offsets of these transform faults occur discontinuously as a result of new fractures at the ridge.