The study of globular cluster systems (GCSs) has long been motivated, at least in part, by the idea that these systems can be used as fossil records of the formation history of their host galaxies (e.g. Harris and Racine 1979; Harris 1991). As described in the previous chapter, empirical information concerning GCSs has grown tremendously in both quantity and quality in recent years. This growth has led to more discriminating tests of models of the formation and evolution of galaxies through the properties of their globular cluster systems. Understanding galaxy formation and evolution is one of the primary challenges in extragalactic astronomy and cosmology. In this chapter, we describe models of galaxy formation and the constraints placed on these models by observations of globular cluster systems.

Models of galaxy formation

Galaxy properties

In the search for a physical model of galaxy formation and evolution, one of the primary questions is why galaxies have such a wide variety of morphologies, star formation histories, and stellar kinematics. One specific issue of great interest is why some galaxies are ellipticals which have old stellar populations, are dynamically hot, and follow de Vaucouleurs' surface brightness profiles (see Section 5.2), and others spirals which have been forming stars at roughly a constant rate over a Hubble time in a rotationally supported, exponential disk.

The dynamical differences between ellipticals and spirals are interesting, particularly since these galaxies have roughly similar mass densities.

Like most astronomical objects, globular clusters exhibit a range of properties and characteristics, but certain features are common to the majority of them. Due to their relative proximity, globular clusters within the Milky Way are the best–studied, and most of the ‘typical’ properties described in this chapter are based on observations of these objects. Unless otherwise stated, the data used in this and the following chapter are taken from the McMaster catalog described by Harris (1996; see also Harris and Harris 1997). Differences between the globular clusters of the Milky Way and other galaxies are summarized at the end of this chapter.

Color–magnitude diagrams

The luminosity and temperature of a star are dependent on its mass, age, and chemical composition. Color–magnitude diagrams of globular clusters have long been the subject of intensive study because they reflect these fundamental properties of the constituent stars. Figure 2.1 shows the color–magnitude diagram of M5 and illustrates a number of the basic features of globular cluster color–magnitude diagrams. These include the main sequence, the giant branch, and the horizontal branch, each of which is discussed in the following subsections.

Main sequence

One of the key features of globular clusters is the well-defined main sequence extending from the turn–off to fainter magnitudes and redder colors (see Figure 2.1). Globular cluster stars on the main sequence derive their energy from the conversion of hydrogen to helium in the stellar core. The low–luminosity end of the main sequence shown in Figure 2.1 is determined by the magnitude limit of the observations.

At present, there is no widely accepted theory of globular cluster formation. In this chapter, we describe some of the general ideas that have been proposed in this area, and compare these ideas with the constraints placed on globular cluster formation models by the observations described in earlier chapters.

One piece of evidence that has played an important role in the development of this field is the lack of current globular cluster formation in the Milky Way today. Open clusters and stellar associations form quite happily at the present epoch in the Galactic disk, but globular clusters do not. Until relatively recently, a significant fraction of astronomers would have probably argued that globular cluster formation was something that only occurred in the early universe. Others have claimed for some time that the most massive young star clusters found in the Large Magellanic Cloud and other similar environments are genuine analogs of the old globular clusters of the Milky Way and other galaxies. As discussed in Chapter 5, there is now evidence that globular clusters are currently forming in merging and interacting galaxies. While it is premature to regard this evidence as conclusive, there seems to be a growing acceptance of the idea that globular clusters can, under certain circumstances, form at the present epoch. As we show below, this possibility has significant consequences for models of globular cluster formation.

The globular cluster system of the Milky Way consists of over 150 known members. It is likely that not all the clusters have been detected, primarily because of obscuration by the Galactic bulge. The best estimate for the total population is around 180 objects. The system is centrally concentrated, with roughly half of the globular clusters residing within about 5 kpc of the Galactic center. However, the most remote clusters extend to beyond 100 kpc from the center of the Galaxy. The system is more usefully considered as two or more distinct subsystems. The majority of globular clusters form a roughly spherical, metal-poor, halo distribution. Recent evidence suggests that this halo population may itself be a composite system. A smaller number of globular clusters are relatively metal-rich and have the spatial and kinematic characteristics of a thick disk population. It has also been suggested that these metal-rich clusters can more properly be regarded as belonging to the bulge population of the Milky Way.

In this chapter, we examine the range of globular cluster properties, correlations between these properties, and how the characteristics of globular clusters vary with position within the Milky Way. We describe the evidence that has led to the separation of Milky Way globular clusters into distinct subsystems. The dynamical evolution of the Galactic globular cluster system and constraints on the properties of the initial Milky Way globular cluster system are also discussed. Finally, we consider the important question of what the Milky Way globular cluster system reveals about the formation and evolution of the Galaxy.

In this chapter we discuss the properties of globular cluster systems (GCSs) around galaxies in the Local Group and slightly beyond. The proximity of such globular clusters allows more detailed information to be obtained about some of their properties than is possible for the more distant extragalactic systems discussed in Chapter 5. This information allows for comparisons between these systems and the Galactic globular cluster system. Such comparisons show that some nearby galaxies have globular clusters very much like those of the Milky Way. However, there are also galaxies which contain young, massive star clusters that have no direct counterparts in the Galaxy. Some of these objects may be young analogs of the old globular clusters in the Milky Way and other galaxies.

The M31 globular cluster system

The Andromeda Galaxy (M31) is the nearest bright spiral to the Milky Way and, consequently, its globular cluster system provides an important comparison to the Galactic globular cluster system. Many of the properties of M31 suggest that it is a slightly scaled-up version of the Milky Way, being somewhat more luminous and of earlier Hubble type. One natural question is how similar the M31 globular cluster system is to the Galactic system.

Starting with Hubble's (1932) identification of globular clusters in M31 and continuing through the present, observers have attempted to obtain a reliable globular cluster sample in this galaxy. Although globular clusters in M31 are expected to be fairly bright, surveying the population has been challenging for several reasons.

This book gives a comprehensive account of the current understanding of globular clusters and globular cluster systems. We describe the observed properties of these objects and discuss the theoretical ideas that are believed to account for their properties. A key theme of the book is to link the study of globular cluster systems to other areas of astrophysics, such as galaxy formation and evolution. The text is designed for graduate students working in globular cluster research, and as a resource for research astronomers active in this area. We also hope that the connections we emphasize between globular cluster systems and other astrophysical systems will make the book useful to researchers in related fields.

Many people have been instrumental in bringing this book about. The original suggestion that we embark on the project came from Simon Mitton, who accurately described the potential pitfalls and rewards of writing a book. Our editor, Adam Black, has patiently provided advice and encouragement. We are grateful to our copy–editor, Sheila Shepherd, for her careful reading of the original manuscript and her extremely helpful suggestions and corrections, and many others at Cambridge University Press who assisted in bringing this book to fruition. Our own work in the field would not have been possible without our collaborators: Christina Bird, Terry Bridges, David Carter, Alberto Conti, Jayanne English, Kenneth Freeman, Doug Geisler, David Hanes, Ray Sharpies and Joseph Silk. Many other colleagues have provided considerable help in preparing the book, suggesting revisions to earlier versions and providing us with data and figures.

This appendix contains data on extragalactic globular clusters systems and their host galaxies. These data are used extensively in Chapters 5 and 6 for studying trends and correlations in these properties. Similar information can be found in Harris and Harris (1997) and Kissler-Patig (1997).

An asterisk after the galaxy number indicates that less than 10% of the estimated number of globular clusters in that galaxy have been observed. All galaxy numbers are NGC numbers unless otherwise specified.

The sources listed in the right-hand column can be found in the references of the main text through the following key:

Ba95, Baum et al (1995); B195, Blakeslee and Tonry (1995); Bo92, Bothun et al. (1992); Br92, Bridges and Hanes (1992); Br94, Bridges and Hanes (1994); Br96, Bridges et al. (1996); Br97, Bridges et al .(1997) Bu92, Butterworth and Harris (1992); Ch3, chapter 3 of this book; Ch4, chapter 4 of this book Ch93, Christian (1993) Di96, Dirsch (1996); Du96a, Durrell et al .(1996a), Du96b, Durrell et al .(1996b); F195, Fleming et al .(1995), Fo96; Forbes et al .(1996); Ge96, Geisler et al .(1996); H91, Harris (1991); Ha92, Harris et al .(1992); Ha95, Harris et al .(1995); Ho95, Hopp et al .(1995); Hi96, Hilker and Kissler-Patig (1996); Ka96, Kaisler et al .(1996); Ki94, Kissler et al .(1994); KP96, Kissler-Patig et al .(1996); KP97, Kissler-Patig et al .(1997); Le94, Lee et al .(1994) McL94, McLaughlin et al .(1994); McL95, McLaughlin et al .(1995); Os93, Ostrov et al (1993); Pe95, Perelmutter and Racine (1995); Ri92, Richtler et al (1992); Se95, Seeker et al (1995); Sh93, Schommer (1993); Sc93, Suntzeff (1993); W97, Whitmore et al (1997) Ze94, Zepf et al (1994), Ze95, Zepf et al (1995).

In this final chapter we consider the promising future directions in globular cluster research. This discussion is driven by those areas where we feel that theoretical and observational progress can have the greatest impact, both in solving currently open questions about globular cluster systems themselves and producing significant advances in other astrophysical fields. To this extent, this chapter is a wish list for the future. The most exciting developments in astrophysics are invariably unexpected, but we feel that offering some speculation on the nature of future advances is useful. We organize this chapter along the lines of the earlier text, starting with globular clusters in the Milky Way and progressing to the distant extragalactic globular cluster systems.

Milky Way globular clusters

In terms of its impact on other areas of astrophysics and cosmology, the age of Milky Way globular clusters is one of the most important topics of current and future research. There are two primary reasons for this interest. First, globular cluster ages provide a lower limit to the age of the universe, so that when combined with other observable cosmological parameters they provide constraints on the geometry and mass density of the universe. Current age estimates already provide important limits on cosmological models, so reducing uncertainties in these estimates is a key goal. Second, the age distribution of Milky Way globular clusters provides critical constraints on the formation history and formation models of the Milky Way.

Understanding the formation and evolution of galaxies is one of the primary goals of extragalactic astronomy and observational cosmology. The galaxies of the Local Group provide a logical starting point for this effort, and the properties of their globular cluster systems (GCSs) have proved to be a valuable tool in this area, as described in Chapters 3 and 4. In order to gain a broader understanding of galaxy formation, it is necessary to extend these studies to galaxies outside the Local Group. Such studies allow an examination of correlations between galaxies and their globular cluster systems with large sample sizes over a wide range of galaxy properties and environments. Of particular importance are giant elliptical galaxies, which are not present in the Local Group and which, by virtue of their old stellar populations and large masses, are a focal point for tests of galaxy formation models.

Beyond the Local Group, little information can be gleaned about the detailed properties of individual globular clusters, but their integrated properties can be observed to roughly 100 Mpc. In contrast, it is a great technical challenge to resolve stars in galaxies at distances greater than a few Mpc (or several times the size of the Local Group). Thus globular clusters provide discrete dynamical and chemical tracers of galaxies at distances far beyond those where individual stars are observable. The great age of many globular cluster systems is also important, since it implies that these objects are direct probes of the physical conditions at the epoch(s) of galaxy formation. Consequently, globular clusters provide one of the few ways to obtain individual snapshots of the formation history of galaxies.