At least one in a dozen stars shows evidence of planetary ingestion.

Stellar chemical compositions can be altered by ingestion of planetary material1,2 and/or planet formation, which removes refractory material from the protostellar disk3,4. These 'planet signatures' appear as correlations between elemental abundance differences and the dust condensation temperature3,5,6. Detecting these planet signatures, however, is challenging owing to unknown occurrence rates, small amplitudes and heterogeneous star samples with large differences in stellar ages7,8. Therefore, stars born together (that is, co-natal) with identical compositions can facilitate the detection of planet signatures. Although previous spectroscopic studies have been limited to a small number of binary stars9-13, the Gaia satellite14 provides opportunities for detecting stellar chemical signatures of planets among co-moving pairs of stars confirmed to be co-natal15,16. Here we report high-precision chemical abundances for a homogeneous sample of ninety-one co-natal pairs of stars with a well defined selection function and identify at least seven instances of planetary ingestion, corresponding to an occurrence rate of eight per cent. An independent Bayesian indicator is deployed, which can effectively disentangle the planet signatures from other factors, such as random abundance variation and atomic diffusion17. Our study provides evidence of planet signatures and facilitates a deeper understanding of the star-planet-chemistry connection by providing observational constraints on the mechanisms of planet engulfment, formation and evolution.

A compositional link between rocky exoplanets and their host stars.

Stars and planets both form by accreting material from a surrounding disk. Because they grow from the same material, theory predicts that there should be a relationship between their compositions. In this study, we search for a compositional link between rocky exoplanets and their host stars. We estimate the iron-mass fraction of rocky exoplanets from their masses and radii and compare it with the compositions of their host stars, which we assume reflect the compositions of the protoplanetary disks. We find a correlation (but not a 1:1 relationship) between these two quantities, with a slope of >4, which we interpret as being attributable to planet formation processes. Super-Earths and super-Mercuries appear to be distinct populations with differing compositions, implying differences in their formation processes.