Racial threat theory posits that as the relative size of the Black population increases in a geographical location, social control is intensified to curtail the rising threat perceived by Whites to their political, economic, and social ascendancy. Increased competition between Whites and Blacks for employment and other economic resources has also been posited to lead to the increased use of social control on the Black population. The current study engaged in a multilevel test of racial threat theory by merging data from the National Incident-Based Reporting System (NIBRS) with Census data to examine whether factors related to racial composition influence the relationship between an offender’s race and likelihood of arrest. Micro-level data on violent crime incidents (N=145,255) from 182 United States cities and 18 States for the year 2000 were obtained from the NIBRS. The authors hypothesized, in accordance with racial threat theory, that in cities where racial threat is heightened, police officers will be more likely to arrest Black citizens suspected of committing crimes, controlling for other relevant factors. The results fail to offer empirical support for racial threat theory; the relative size of the Black population actually has a negative effect on the probability of arrest. Additionally, racial segregation was found to condition the relationship between an offender’s race and the probability of arrest; in cities with a high degree of racial segregation, crimes involving Black offenders were less likely to result in an arrest. This finding suggests that racial segregation in itself serves as an instrument of State control whereby “problem” populations are managed by physically marginalizing them as a group. Tables, references