Elliptical galaxies are a cornerstone in our understanding of the galaxy formation process in a cosmological frame. According to the currently most popular models, elliptical galaxies are the outcome of (multiple) merging of already formed units, as a natural result of the hierarchical models which successfully account for the large scale structure observed in the local universe. The hierarchical models, however, give strong predictions for the clustering of the dark matter, while the behavior of the luminous matter depends on many parameters, which are less constrained. Of particular importance is to establish if the most massive ellipticals are young or old, so as to derive the epoch of the major episode of star formation. Unfortunately, the vast majority of ellipticals cannot be resolved into single stars, and the age-dating of their stellar content relies on studies of the their integrated light. These studies are severely hampered by the age-metallicity degeneracy, i.e. the same colors and spectral indices are obtained when decreasing the age and increasing the metallicity at the same time. In principle the degeneracy can be broken on index-index plots, where the two indices have different dependences on age and metallicity. In practice, this procedure needs accurate and well calibrated models of the integrated light from stellar populations. In collaboration with C. Maraston (Univ. of Munich) we are testing the sensitivity of the model spectral indices to the input parameters which enter their computations, exploring both physical and numerical uncertainties.
In 1999 we updated the code for the computation of the evolution of
integrated light of single age, and single metallicity populations of
single stars, to predict the evolution of the whole spectral energy
distribution. We applied this instrument to interpretate the
evolution with redshift of color gradients in ellipticals, comparing
the model predictions to data for nearby galaxies and for galaxies at
redshift 
0.4 [HST proprietary data; Saglia et al. (2000)].
We also acquired ESO, 1.5m spectra of globular clusters in the Galactic Bulge. These data will be soon reduced and compared to the models predictions, to achieve a proper calibration in the high metallicity range, which is a relatively poorly explored regime. This work is done in collaboration with A. Renzini (ESO), C. Maraston, R. Saglia and T. Putzia (Univ. of Muenchen).