The data on arrests for murder/manslaughter, robbery, aggravated assault, and burglary were aggregated from raw data made available by the New York City Police Department in the form of arrest counts by crime type, race and age subgroups, and year (1987-1990). In conjunction with population estimates from the census (1987 and 1990), race- age and crime-specific arrest rates were constructed for the years 1987-1990. Additional official arrest-record data came from a pool of 2,615 women. The findings clearly indicate the disparity in the relative rates of offending among black, Hispanic, and white females. Both black and Hispanic females exhibit high rates of violence relative to white females. The criminal violence of black females, however, is substantially higher than that of both Hispanic and white females. The violent offending rates of black females paralleled those of white males. These findings suggest that an examination of gender differences in violent offending, independent of race, will produce incomplete and potentially misleading interpretations. The authors provide a social structural explanation of the disproportionate involvement of black females (and males) in violent crime. 4 tables, 3 notes, and 32 references