This brief review on extragalactic globular cluster systems is derived from a lecture given for the award of the Ludwig-Bierman-Preis of the Astronomische Gesellschaft in Göttingen during September 1999. The Oral version aimed at introducing, mostly from an observer's point of view, this field of research and at emphasizing its tight links to galaxy formation and evolution.
The scope of this written follow-up is not to give a complete review on globular cluster systems but to present recent discoveries, including examples, and to set them into the context of galaxy formation and evolution. The choice of examples and the emphasis of certain ideas will necessarily be subjective, and we apologize at this point for any missing references.
Excellent recent reviews can be found in the form of two books: "Globular Cluster Systems" by Ashman & Zepf (1998), as well as "Globular Cluster Systems" by Harris (2000). These include a full description of the globular clusters in the Local group (not discussed here), as well as an extensive list of references, including to older reviews.
The plan of the article is the following. In section 2, we give an introduction and the motivation for studying globular cluster systems with the aim of understanding galaxies. Section 3 presents current and future methods of observations, and the rational behind them. This section reviews recent progress in optical and near-infrared photometry and multi-object, low-resolution spectroscopy. It can be skipped by readers interested in results rather than methods. In Section 4 we present in turn the status of our knowledge on globular cluster systems in ellipticals, spirals and mergers. What are the properties of the systems? How are the galaxy types linked? And do mergers produce real `globular' clusters? In section 5, we discuss sub-populations of globular clusters and their possible origin. The most popular scenarios to explain the presence of globular cluster sub-populations around galaxies are listed. The pros and cons, as well as the expectations of each scenario are discussed. We present, in section 6, some results from the study of globular cluster system kinematics. Finally, in section 7, we revisit the globular cluster luminosity function as a distance indicator. Under which conditions can it be used, and how should it be applied to minimize any systematic errors? It is compared to other distance indicators and shown to do very well. Some conclusions and an outlook are given in section 8.