A popular perception is that immigration causes higher crime rates. Yet, historical and contemporary research finds that at the individual level, immigrants are not more inclined to commit crime than the native born. Knowledge of the macro-level relationship between immigration and crime, however, is characterized by important gaps. Most notably, despite the fact that immigration is a macro-level social process that unfolds over time, longitudinal macro-level research on the immigration-crime nexus is virtually nonexistent. Moreover, while several theoretical perspectives posit sound reasons why over-time changes in immigration could result in higher or lower crime rates, we currently know little about the veracity of these arguments. To address these issues, this study investigates the longitudinal relationship between immigration and violent crime across U.S. cities and provides the first empirical assessment of theoretical perspectives that offer explanations of that relationship. Findings support the argument that immigration lowers violent crime rates by bolstering intact (two-parent) family structures.