The sample of college freshmen came from a medium-sized, state-supported university in a midwestern community. A computerized random search of freshmen enrollment records who were less than or equal to 20 years of age yielded a sample of 700 students. Freshmen responded to a self-report questionnaire containing 45 items. Paired t-tests were used to determine if a significant change in the use of individual drugs occurred between high school and college. Although the best predictor of drug use at the beginning of college was drug use in high school, for drugs other than alcohol and marijuana, college freshmen used marijuana less frequently than they did in high school and the use of alcoholic beverages increased early in college. While the frequency of alcohol use increased considerably, college freshmen did not increase the number of times they got drunk. Data showed that high school and college drug use rates were similar, and entering freshmen found new friends much like their high school friends to use drugs with and get drunk. The author concludes that alcohol and drug use may be important determining factors in the choice of new college friends and that programs involving positive peer group bonding should be made available to college students. 18 references and 5 tables