Adrien Morison’s paper, co-authored by Renaud Deguen, Thierry Alboussière and myself, just got published in Earth and Planetary Sciences Letters. In this paper, we compute the timescale of overturn in a solid mantle cristallising from a magma ocean. We show that the overturn, driven both by the temperature and composition unstable stratification, happens much faster than the time required to freeze the magma ocean. Therefore, solid-state mantle convection must have started before full crystallisation. Moreover, allowing for phase change between the solid mantle and the magma ocean makes the degree one the preferred mode of overturn. This could explain the Moon or Mars dichotomy. This paper is part of the MaCoMaOc Project and can be found on the journal’s webpage.