The data came from 11 cities with populations of at least 5,000 Hispanic persons in 1980. The research used the data because they were the only reliable source of information from government statistics, while recognizing that these data might not represent the current situation. This data source allowed both disaggregation of Latino killings and the computation of rates for different types of Latino homicide, including murder committed in the curse of another felony, acquaintance homicide, family homicides, and stranger homicides. Results of ordinary least-squares regression analysis provided some support for interpretations of violence as related to economic deprivation and social disorganization. However, the role of immigration varied by the type of homicide and was limited. Findings underscore the need to address the link between immigration, urban conditions, and types of homicide victimization among Latinos. Tables, footnotes, and 24 references (Author abstract modified)