The merging history of the Milky Way

The age distribution, and chemical elemental abundances, of stars in the haloof the Milky Way provide constraints on theories of galaxy formation. As onespecific example, the accretion of satellite galaxies similar to the presentretinue of dwarf spheroidals (dSphs) would provide an observable metal-poor,intermediate-age population. This paper presents a quantitative assessment ofany contribution made by such stars to the stellar halo. The bulk of thestellar populations in the halo show a well-defined turn-off, at B-V ~ 0.4,implying that the vast majority of the stars are old. The fraction of starswhich lie blueward of this well-defined turn-off, with metallicities similar tothat of the present dSphs, is used in this paper to place limits on theimportance of the recent accretion of such systems. Very few (< ~ 10%) starsare found to be bluer (and by implication, younger) than this limit, with thehighest value found for the more metal-rich halo ([Fe/H] >~ -1.5 dex). Directcomparison of this statistic with the colour distribution of the turnoff starsin the Carina dwarf allows us to derive an upper limit on the number of mergersof such satellite galaxies into the halo of the Milky Way. This upper limit is~ 40 Carina-like galaxies. The higher metallicity data constrain satellitegalaxies like the Fornax dwarf; only <~ 5 of these could have been accretedwithin the last < ~ 10 Gyr. We note that the low star-formation rates inferredfor dSphs predict distinctive elemental abundance signatures; future data forfield halo stars, including candidate younger stars, will provide a furtherrobust test of accretion models.