Globular Cluster Systems: Do They Really Trace Star Formation? (Or Rather: What Mode of Star Formation Do They Trace?)

Abstract

A good reason to believe that globular clusters (GCs) are representative of the star formation in galaxies is that the mean colours of GC systems are correlated with the luminosity of their host galaxies [35][9]. The GC systems of massive galaxies appear bimodal in their colour distributions [38], and it has now been demonstrated that both the metal-rich (red) and metal-poor (blue) GC peak positions are a function of galaxy luminosity [20][32][24]. Although there is increasing evidence for the existence of (an undetermined fraction of) young GCs in elliptical galaxies [28][23](T Bridges, these proceedings), the colour distributions of GC systems are generally thought to reflect underlying metallicity rather than age distributions. Thus, more massive galaxies have, on average, more metal-rich GC systems. However, if GC formation were to mirror the formation of the galaxy field stars at a fixed efficiency (by field stars, we refer to all the stars now in the spheroid), the metallicity distributions of GCs and field stars should be very similar. They do not appear to be [17] (Sec 5). Understanding this difference between the GCs and galaxy stars is essential if we are to use GCs to unravel the detailed formation histories of galaxies. In the following some general ideas about the formation of GCs in the context of galaxy formation are presented. We then discuss ongoing work on the comparison of the GCs and field stars of the nearest elliptical galaxy NGC 5128.