Introduction

This book spans my entire life as a research worker at Cambridge. The circumstances that created this opportunity were based entirely on luck and this aspect played a vital part during subsequent developments. In the following chapters, I have tried to give details of the most relevant methods used in so-called ‘direct integration’ of the classical N-body problem, a method of attack somewhat analogous to scaling a mountain the hard way. This has been enhanced by an extensive discussion of the main algorithms implemented in the associated computer codes. A comprehensive review of related work in the field over the last 40 years is also presented. Throughout the term N-body simulations is used exclusively for methods based on direct summation, in keeping with tradition.

Although a wide range of problems is covered, the emphasis is on the dynamics of star clusters. This involves many aspects of stellar evolution. It is fortuitous that the University of Cambridge has a long tradition in this field that dates back to Eddington and Jeans. Fred Hoyle continued this school, which eventually gave rise to the application of synthetic stellar evolution. This subject was pioneered entirely at the Institute, mainly by the sequential efforts of Peter Eggleton, Christopher Tout and Jarrod Hurley, whose work has been vital for realistic star cluster simulations.