We have mainly estimated ages and foreground interstellar reddening values for 15 SMC concentrated star clusters, from integrated spectra in the range (3600–6800) Å. Most of the sample are young blue objects (6–60 Myr), while L28, NGC 643 and L 114 are found to be intermediate-age clusters (1–6 Gyr). The present data also constitute a spectral library at the metallicity level of SMC clusters.

We present a set of star cluster integrated spectra based on observations from the ultraviolet to the near-infrared. Mean template spectra were built following a ranking of clusters according to age and metallicity, which were taken from recent studies and put on consistent scales. The complete spectral database including star clusters, galaxies and stars (templates and individuals) is available through the CDS.

Young star clusters in interacting galaxies are often found in groups or clusters of star clusters containing up to 100 single clusters. In our project we study the future fate of these clusters of star clusters. We find that the star clusters merge on time scales of a few dynamical crossing times of the super-cluster. The resulting merger object has similarities with observed dwarf ellipticals (dE). Furthermore, if destructive processes like tidal heating, dynamical friction or interaction with disc or bulge of the parent galaxy are taken into account our merger objects may evolve into objects resembling dwarf spheroidal galaxies (dSph), without the need of a high dark matter content.

We have developed a model of globular cluster self-enrichment, based on the ability of the globular cluster gaseous progenitors to retain the ejecta of a first generation of Type II Supernovae. The key point is that this ability depends on the pressure exerted on the progenitor cloud by the surrounding protogalactic medium and therefore on the location of the cloud in the protogalaxy. The model is able to explain the galactic halo metallicities and the metallicity gradient of the Old Halo which is thought to be the genuine galactic globular cluster system. The possibility that metallicity gradients are a common property of extragalactic globular cluster systems is also presented.

Following Roxburgh's integral constraint, we implemented an upper limit of overshoot in the conventional method of α parameterization in order to remove an overly large overshoot effect for low-mass stars. The erroneously large effect of overshoot due to the failure of α parameterization can be effectively corrected by limiting the amount of overshoot to 15 % of the core radius.

We are developing a method for measuring the detailed chemical composition and evolutionary history of extragalactic star clusters from high resolution spectra of their integrated light as one would from spectra of individual stars. In this paper, we show high signal-to-noise ratio echelle spectra of the integrated light of two Galactic globular clusters and equivalent-quality spectra of individual stars in those clusters in order to briefly illustrate some subtleties of the analysis method.

We discuss the future possibilities for extragalactic star cluster research with the expected new ground-based and space-based telescopes and instrumentation. Significant gains are expected due to improved angular resolution, sensitivity, and area coverage particularly in the infrared and radio, accompanied by progress in evolutionary and dynamical modelling. Improvements in angular resolution are anticipated, especially through new adaptive optics systems (e.g., Keck, Gemini, VLT), and interferometry (e.g., Keck, VLT, LBT, ALMA, SMA, SkA), and space instrumentation (e.g., Chandra, NGST), enabling studies even of deeply embedded, forming extragalactic star clusters. Tidal disruption of Galactic clusters becomes observable through wide-area surveys such as the SDSS, VISTA, PRIME, including proper motion measurements through high-resolution imaging (e.g., HST, LBT, SIM, GAIA). Sensitive new optical and infrared spectrographs (e.g., HET, SALT, GranTeCan, Magellan, Keck, VLT, CELT, OWL, NGST) will push kinematic and abundance studies to new limits, allowing us detailed comparisons with model predictions. One important wavelength range for the study of young, massive star clusters, the far UV, appears to be neglected by future planned instrumentation.

This was a superb conference: scientifically exciting, and wonderfully enjoyable from every viewpoint. The biggest problem was, of course, that trying to absorb all the science has resembled drinking water from a firehose: there has been just too much of it to take in on the spot. So if the purpose of a conference summary talk is to dispense sage advice, then I will say that the best thing all of us can do now is simply to go home and think, and begin digesting this vast array of new material. IAU 207 is sure to be remembered as a launching point for many new directions and syntheses that will emerge from its stimulus over the next several years.