The homicide epidemic of the late 1980s and early 1990s continues to receive attention from students of crime. However, while a few studies have begun the task of uncovering the causes and characteristics of the epidemic, a clear understanding of the epidemic remains elusive. This book can be viewed in the context of an evolving examination of the epidemic. The findings take another step toward disentangling the causes of the homicide epidemic. The findings include: 1) hypotheses based on the research on general homicide rates are not confirmed when applied to the parameters of the epidemic; 2) no significant interaction effects were found between the cocaine measures and the social variable included in the best-fitting statistical model for each of the parameters; 3) while both the presence of the epidemic and the duration of the epidemic were most strongly influenced by population density, the magnitude measures were not; 4) the effect of population density changed signs, from a positive effect on the presence, to a negative effect on the duration; 5) the number of cocaine hubs was negatively associated with the absolute magnitude epidemic; and 6) the per capita number of police was not a significant predictor in any of the preferred statistical models. One of the most important theoretical contributions of this book is the adoption of the epidemiological approach to understanding the rapid homicide rate increases of the late 1980s and early 1990s. The general approach of this book represents a significant extension of previous research on the homicide epidemic because it relies on a fusion of concepts and ideas from the field of epidemiology with subject mater that has been most thoroughly explored by criminology. Tables, figures, appendixes A-E, references, and index