We use the calculations derived in a previous paper (M\'era, Chabrier and Schaeffer, 1997), based on observational constraints arising from star counts, microlensing experiments and kinematic properties, to determine the amount of dark matter under the form of stellar and sub-stellar objects in the different parts of the Galaxy. This yields the derivation of different mass-models for the Galaxy. In the light of all the afore-mentioned constraints, we discuss two models that correspond to different conclusions about the nature and the location of the Galactic dark matter. In the first model there is a small amount of dark matter in the disk, and a large fraction of the dark matter in the halo is still undetected and likely to be non-baryonic. The second, less conventional model is consistent with entirely, or at least predominantly baryonic dark matter, under the form of brown dwarfs in the disk and white dwarfs in the dark halo. We derive observational predictions for these two models which should be verifiable by near future infrared and microlensing observations.

We examine the most recent observational constraints arising from i) small-scale and large-scale Galactic dynamical properties, ii) star counts at faint magnitude and iii) microlensing experiments. From these constraints, we determine the halo and disk stellar mass functions and stellar content down to the bottom of the main sequence, which yields the normalization of the halo/disk total stellar population, and we infer the contributions of sub-stellar objects to the mass budget of the various Galactic regions. The consistent analysis of star counts and of the overall microlensing observations in the Bulge are compatible with a small contribution of brown dwarfs to the Galactic mass budget $\rho_{BD}/\rho_* \leq 0.2 $. However the separate bulge/disk analysis based on the bulge clump giants is compatible with a substantial population of disk brown dwarfs, $\Sigma_{BD}/\Sigma_*\leq 1 $. More statistics of microlensing events towards the Galactic center and a better determination of the velocity dispersions in the bulge should break this degeneracy of solutions. For the halo, we show that a steep mass-function in the dark halo is excluded and that low-mass stars and brown dwarfs represent a negligible fraction of the halo dark matter, and thus of the observed events towards the LMC. The nature of these events remains a puzzle and halo white dwarfs remain the least unlikely candidates.

We use recent low-mass star models, which reproduce accurately the observed sequences of various globular clusters, to convert the observed luminosity functions into bolometric luminosity functions and mass functions down to the bottom of the main sequence. These mass functions are well describedby a slowly rising power-law $dN/dm\propto m^{-\alpha}$, with $0.5\wig < \alpha \wig < 1.5$, down to $\sim 0.1 \msol$, independently of the metallicity, suggesting a rather universal behaviour of the cluster initial mass functions. We predict luminosity functions in the NICMOS filters in the stellar and in the brown dwarf domains for different mass functions and metallicities. We apply these calculations to the determination, slope and normalization, of the mass function of the Galactic halo (spheroid and dark halo). The spheroid mass function is well described by the afore-mentioned power-law function with $\alpha \sim 1.7\pm 0.2$ down to 0.1 $\msol$, although a slowly decreasing mass function below $\sim 0.15 \msol$ can not be excluded with the data presently available. Comparison with the Hubble Deep Field star counts is consistent with such a mass function and excludes a significant stellar population in the dark halo. Consistent analysis with recent microlensing experiments towards the LMC shows that the spheroid and the dark-halo stellar+brown dwarf populations represent at most $\sim$ 1% of the Galactic dark matter density. This clearly excludes brown dwarfs and low-mass stars as significant dark matter candidates.

We examine the recent results of the MACHO collaboration towards the Large Magellanic Cloud (Alcock et al. 1996) in terms of a halo brown dwarf or white dwarf population. The possibility for most of the microlensing events to be due to brown dwarfs is totally excluded by large-scale kinematic properties. The white dwarf scenario is examined in details in the context of the most recent white dwarf cooling theory (Segretain et al. 1994) which includes explicitely the extra source of energy due to carbon-oxygen differentiation at crystallization, and the subsequent Debye cooling. We show that the observational constraints arising from the luminosity function of high-velocity white dwarfs in the solar neighborhood and from the recent HST deep field counts are consistent with a white dwarf contribution to the halo missing mass as large as 50 %, provided i) an IMF strongly peaked around 1.7 Msol and ii) a halo age older than 18 Gyr.