Constraints on the Merger Models of Elliptical Galaxies from their Globular Cluster Systems

The discovery of proto-globular cluster candidates in many current-day mergers allows us to better understand the possible effects of a merger event on the globular cluster system of a galaxy, and to foresee the properties of the end-product. By comparing these expectations to the properties of globular cluster systems of todays elliptical galaxies we can constrain merger models. The observational data indicate that i) every gaseous merger induces the formation of new star clusters, ii) the number of new clusters formed in such a merger increases with the gas content of the progenitor galaxies. Low-luminosity (about M_V>-21), disky ellipticals are generally thought to be the result of a gaseous merger. As such, new globular clusters are expected to form but have not been detected to date. We investigate various reasons for the non-detection of sub-populations in low-luminosity ellipticals, i.e. absence of an old population, absence of a new population, destruction of one of the populations, and finally, an age-metallicity conspiracy that allows old and new globular clusters to appear indistinguishable at the present epoch. All of these possibilities lead us to a similar conclusion, namely that low-luminosity ellipticals did not form recently (z<1) in a gas-rich merger, and might not have formed in a major merger of stellar systems at all. High-luminosity ellipticals do reveal globular cluster sub-populations. However, it is difficult to account for the two populations in terms of mergers alone, and in particular, we can rule out scenarios in which the second sub-population is the product of a recent, gas-poor merger.