(THE 'SUBCULTURE-OF-VIOLENCE' THESIS IS BASED ON THE POSITIVE RELATIONSHIP BETWEEN THE PROPORTION OF BLACK OR NONWHITE, THE PROPORTION OF POPULATION POOR, OR THE PROPORTION OF A POPULATION WHICH IS YOUNG AND MALE WITH THE HOMICIDE RATE IN THE UNITED STATES.) VARIATIONS IN THE HOMICIDE RATE AND AGGRAVATED ASSAULT RATE IN 171 AMERICAN CITIES (WITH A POPULATION OF 50,000 OR MORE IN 1940) ARE EXAMINED FOR THE YEARS 1940, 1950, 1960, AND 1970. ANALYZED WERE THE CORRELATION COEFFICIENTS BETWEEN HOMICIDE AND ASSAULT AND THE EIGHT INDEPENDENT VARIABLES (DENSITY, ROOMS PER DWELLING UNIT, HIGH CROWDING, POPULATION SIZE, THE PERCENTAGE OF NONWHITE POPULATION, THE PERCENTAGE OF YOUNG MALES BETWEEN 15 AND 29 YEARS OF AGE, THE PERCENTAGE POOR, AND THE 'SOUTHERNNESS INDEX'). DATA FOR 1950 AND 1970 INDICATED PARTIAL SUPPORT FOR BOTH THE 'SOCIAL STRUCTURE CAUSES PATHOLOGY' AND THE 'DENSITY CAUSES PATHOLOGY' ARGUMENTS. THE SOCIAL-STRUCTURAL VARIABLES SEEMED TO BE SOMEWHAT MORE STRONGLY RELATED TO THE HOMICIDE AND ASSAULT RATES THAN THE DENSITY MEASURES, ALTHOUGH BOTH APPEARED TO MAKE INDEPENDENT CONTRIBUTIONS TO EXPLAINING VARIATIONS IN THESE RATES ACROSS CITIES. THE DENSITY ARGUMENTS ALSO SUFFERED SOMEWHAT MORE FROM AN EXAMINATION OF TRENDS OVER TIME. APPENDED IS A LIST OF THE CITIES INCLUDED IN THE ANALYSIS. REFERENCES ARE ALSO PROVIDED.