2025MNRAS.537.3730H

Unravelling galaxy formation theory requires understanding galaxies both at high and low redshifts. A possible way to connect both realms is by studying the oldest stars in the Milky Way (i.e. the proto-Galaxy). We use the APOGEE-Gaia surveys to perform a purely chemical dissection of Milky Way (MW) stellar populations, and identify samples of stars likely belonging to proto-Galactic fragments. The metallicity dependence of the distribution of old MW stars in the [Mg/Mn]-[Al/Fe] enables the distinction of at least two populations in terms of their star formation histories: a rapidly evolved population likely associated with the main progenitor system of the proto-MW; and populations characterized by less efficient, slower, star formation. In the Solar neighbourhood less efficient star forming populations are dominated by the Gaia-Enceladus/Sausage accretion debris. In the inner Galaxy, they are largely associated with the Heracles structure. We model the density of chemically defined proto-Galaxy populations, finding that they are well represented by a Plummer model with a scale radius of $a\sim 3.5$ kpc, and an oblate ellipsoid with flattening parameters [$p\sim 0.8$; $q\sim 0.6$]; this finding indicates that the MW plausibly hosts a low-mass, metal-poor, bulge component. We integrate this density for chemically unevolved stars between $-2 \lt \mathrm{[Fe/H]} \lt -0.5$ to obtain a minimum stellar mass for the proto-Galaxy of $M_{*} (r\lt 10~\mathrm{kpc}) = 9.1\pm 0.2\times 10^{8}$ M$_{\odot }$. Our results suggest the proto-Milky Way is at least comprised of two significant fragments: the main in situ progenitor and the Heracles structure.