The determination of the stellar mass of a galaxy is always subject to big uncertainties. However there is much stake to know it : first, knowing the luminosity of the stars, which is observed in various colors, one can derive the mass-to-light ratio (M/L), which depends on the nature of stars, their age, their metallicity, and on their distribution in mass. One can thus go back to the star formation history of the galaxy. In addition, knowing the dynamic mass of the galaxy, derived from the stellar velocity field, one can determine the missing mass, therefore the quantity of dark matter in the galaxy as a function of radius, by subtraction of the stellar mass from the total mass. In the elliptical or lenticular galaxies, which have very little interstellar gas, the dynamic mass is dominated by the stellar mass, the dark matter representing only a few percents. This is thus favorable to measure the stellar mass. However, these early-type galaxies are animated by random motions more than ordered rotation, and it is difficult to model the velocity field, dominated by dispersion. The determination of the dynamic mass then requires a more sophisticated model.
The study just published by the international team, attempts to model in a very detailed and precise way a large sample of 260 galaxies of early-types, some examples of which are displayed in Figure 1. They have obtained the velocity fields on the telescope William Herschel in the Canary islands, and the homogeneous multi-color photometry from the SDSS (Sloan DIGITAL Sky Survey). They varied a great number of parameters in the modeling of the velocity fields, including the quantity and the distribution of dark matter, inclination, velocity anisotropy. The main result is that, not only the M/L ratio is different for each galaxy, but also the IMF function cannot be universal, and must vary from one galaxy to another, and in particular be different between elliptical and spiral galaxies. The spirals would have in proportion less stars of small mass, which have little luminosity, whereas the ellipticals on the contrary would have much more stars of small mass. Many studies had tried to demonstrate the non universality of the IMF, but there was not enough data to conclude in a convincing way.