Racial/ethnic residential segregation has been shown to contribute to violence and have harmful consequences for minority groups. However, research examining the segregation-crime relationship has focused almost exclusively on blacks and whites while largely ignoring Latinos and other race/ethnic groups and has rarely considered potential mediators (e.g., concentrated disadvantage) in segregation-violence relationships. This study uses year 2000 arrest data for California and New York census places to extend segregation-crime research by comparing the effects of racial/ethnic residential segregation from whites on black and Latino homicide. Results indicate that (1) racial/ethnic segregation contributes to both Latino and black homicide, and (2) the effects for both groups are mediated by concentrated disadvantage. Implications for segregation-violence relationships, the racial-invariance position, and the Latino paradox are discussed.