Constraints on the Galactic bulge/bar structure and formation history from stellar kinematics and metallicities mainly come from relatively high-latitude fields (|b|>4) where a complex mix of stellar population is seen. We aim here to constrain the f
ormation history of the Galactic bar by studying the radial velocity and metallicity distributions of stars in-situ (|b|<1). We observed red clump stars in four fields along the bars major axis (l=10,6,-6 and b=0 plus a field at l=0,b=1) with low-resolution spectroscopy from VLT/FLAMES, observing around the CaII triplet. We developed robust methods for extracting radial velocity and metallicity estimates from these low signal-to-noise spectra. We derived distance probability distributions using Bayesian methods rigorously handling the extinction law. We present radial velocities and metallicity distributions, as well as radial velocity trends with distance. We observe an increase in the radial velocity dispersion near the Galactic plane. We detect the streaming motion of the stars induced by the bar in fields at l=+/-6, the highest velocity components of this bar stream being metal-rich ([Fe/H]~0.2 dex). Our data is consistent with a bar inclined at 26+/-3 from the Sun-Galactic centre line. We observe a significant fraction of metal-poor stars, in particular in the field at l=0,b=1. We confirm the flattening of the metallicity gradient along the minor axis when getting closer to the plane, with a hint that it could actually be inverted. Our stellar kinematics corresponds to the expected behaviour of a bar issued from the secular evolution of the Galactic disc. The mix of several populations, seen further away from the plane, is also seen in the bar in-situ since our metallicity distributions highlight a different spatial distribution between metal-poor and metal-rich stars, the more metal-poor stars being more centrally concentrated.
Until recently our knowledge of the Galactic Bulge stellar populations was based on the study of a few low extinction windows. Large photometric and spectroscopic surveys are now underway to map large areas of the bulge. They probe several complex st
ructures which are still to be fully characterized as well as their links with the inner disc, the thick disc and the inner halo. I will review our current, rapidly increasing, knowledge of the bulge stellar populations and the new insight expected towards the Gaia era to disentangle the formation history of the Galactic inner regions.
Two main scenarios for the formation of the Galactic bulge are invoked, the first one through gravitational collapse or hierarchical merging of subclumps, the second through secular evolution of the Galactic disc. We aim to constrain the formation of
the Galactic bulge through studies of the correlation between kinematics and metallicities in Baades Window (l=1, b=-4) and two other fields along the bulge minor axis (l=0, b=-6 and b=-12). We combine the radial velocity and the [Fe/H] measurements obtained with FLAMES/GIRAFFE at the VLT with a spectral resolution of R=20000, plus for the Baades Window field the OGLE-II proper motions, and compare these with published N-body simulations of the Galactic bulge. We confirm the presence of two distinct populations in Baades Window found in Hill et al. 2010: the metal-rich population presents bar-like kinematics while the metal-poor population shows kinematics corresponding to an old spheroid or a thick disc one. In this context the metallicity gradient along the bulge minor axis observed by Zoccali et al. (2008), visible also in the kinematics, can be related to a varying mix of these two populations as one moves away from the Galactic plane, alleviating the apparent contradiction between the kinematic evidence of a bar and the existence of a metallicity gradient. We show evidences that the two main scenarios for the bulge formation co-exist within the Milky Way bulge.
