State-level data on homicides and traffic fatalities were obtained for 1975, 1980, 1985, 1990, 1995, and 1996. City-level data were collected in 1985 and 1996. Data analysis for the multi-year period revealed a mild to moderate relationship between homicide rates and traffic fatality rates. This relationship was generally stronger in the later years than in the earlier years. In addition, stronger effects were obtained when using better measures of driving (traffic fatalities per million miles driven) than when using a population measure (traffic fatalities per 100,000 population). The significant relationship between homicide rates and traffic fatality rates held for both state-level and city-level data. Study findings are examined in relation to other research studies. The authors contend that law enforcement can play a significant role in determining the level of each type of fatality and that public perceptions of police activity are important. 34 references, 1 note, and 4 tables