The nature and the location of the lenses discovered in the microlensing surveys done so far towards the LMC remain unclear.

This contribution is comprised of two distinct parts. In the first part, motivated by these questions, we compute the optical depth for the different intervening populations and the number of expected events for self-lensing, using a recently drawn coherent picture of the geometrical structure and dynamics of the LMC disk. By comparing the theoretical quantities with the values of the observed events it is possible to put some constraints on the location and the nature of the machos. Clearly, given the large uncertainties and the few events at our disposal it is not yet possible to draw sharp conclusions, nevertheless we find that up to 3-4 macho events might be due to lenses in LMC, which are most probably low mass stars, but that hardly all events can be due to self-lensing. A plausible solution is that the events observed so far are due to lenses belonging to different intervening populations: low mass stars in the LMC, in the thick disk, in the spheroid and some true machos in the halo of the Milky Way and the LMC itself. We report also on recent results of microlensing searches in direction of the M31 galaxy, by using the pixel method. The present analysis still does not allow yet to draw sharp conclusions on the macho content of the M31 galaxy.

In the second part (section 5), a preliminary account of the final results from the EROS-2 programme is presented. Based on the analysis of 33 million LMC and SMC stars followed during 6.7 years, strict limits on the macho content of the galactic halo are presented; they cover the range of macho masses between 0.0001 and 100 solar mass. The limits are better than 20% (resp. 5%) of the standard halo for masses between 0.0002 and 10 (resp. 0.001 to 0.1) solar mass. This is presently the data set with the largest sensitivity to halo machos.To search for other articles by the author(s) go to: http://adsabs.harvard.edu/abstract_service.html