The purpose of this investigation is to assess the association between racial rioting and postriot crime changes in the United Stated Seven riot cities, each experiencing multiple riots between 1960 and 1991 (Newark, Detroit, Atlanta, Miami, Los Angeles, Baltimore, and Philadelphia), and one nonriot comparison city (Cleveland) were selected. Cleveland served as a nonriot comparison city for Baltimore. Appropriate comparison cities for the remaining six riot cities were not available. Five crime categories (burglary, murder, aggravated assault, robbery, and motor vehicle theft) are analyzed within each city. Uniform Crime Report (UCR) data from 1960 through 1991 were obtained from the Federal Bureau of Investigation. Crime rates per 100,000 were calculated using UCR population estimates. Interrupted time series analyses with multiple replications were used to assess the relationship between rioting and postriot crime changes within each city. However, due to the use of a comparison city, Baltimore crime rates are assessed in relation to Cleveland rates. Autoregressive integrated moving average (ARIMA) models are identified for each series. First, the series was prewhitened. Following prewhitening, the riot dummy variable(s) is/are introduced into the model to assess riot impact(s). Each city analyzed, with the exception of Philadelphia, demonstrates at minimum one significant post-riot increase in one crime category. A total of 16 statistically significant post-riot crime increases were found. Fifteen of these 16 increases are abrupt and permanent in nature while the remaining one is abrupt and temporary. These significant increases are often associated with the most severe riot occurrence within each city. After exploring alternative explanations, it is concluded that the findings of this investigation provide sound methodological and empirical support for an association between severe racial rioting and postriot crime increases. The results of this investigation demonstrate that the consequences of rioting extend beyond the calming of the riot event. The policy and practical implications suggested involve increased law enforcement surveillance and diminished victim availability. Data limitations, theoretical implications, and research implications also are considered. Tables, references