Alcohol is second only to smoking in its contribution to preventable health and social costs, with patterns of potentially harmful alcohol use, for example binge drinking, particularly common among young Australians. Harm minimization strategies, such as publishing alcohol consumption guidelines aimed at encouraging alcohol use within levels assessed to hold a low risk of harm, are typically used by Australian public health officials to reduce the costs of alcohol misuse. In attempting to modify such patterns of alcohol use it is important to understand the initiation and establishment of this behavior in adolescence and young adulthood. Longitudinal community studies provide an important method of identifying subgroups that differ across time in their adherence to recommended levels for alcohol use. This study investigated the trajectories into potentially harmful alcohol use, and sought to identify the subgroups that differ across time, from the last year of high school into post-school life, in their adherence to recommended levels for alcohol use. It was hypothesized that three trajectories would be identifiable: 1) students that either did not use alcohol or used alcohol infrequently would continue to show stable patterns of low-level alcohol use into early adulthood; 2) students that showed high-levels of alcohol use in high school would continue to do so through to follow-up; and 3) students who showed escalation in their alcohol use in high school would continue to do so in post-school life. An initial group of 6,176 students in grade 12 completed questionnaires in early 1993. A randomly selected group of 3,300 students were mailed follow-up questionnaires on 3 separate occasions over the next 20 months: at 3 months, 9 months, and 15 months after leaving school. The sample was reduced to 2,591 (males = 1,000; females = 1,591) students that provided data for 2 or more of the 4 waves. Differences were noted for those that continued with the study and those that dropped after wave 1. The retained sample had lower initial rates of alcohol use and had proportionally more females in it. An alcohol consumption index was generated from questionnaire items and participants were classified according to one of the five categories: non-drinkers, infrequent drinkers, low weekly consumption, moderate weekly consumption, and high weekly consumption. Findings show that the greatest increase in alcohol use was between the final year and the 3 months following high school, with the high weekly consumption group showing the largest rise. Subgroups, or trajectories, within the sample were identified according to their general pattern of movement between drinking categories across the four survey waves. A number of similarities were observed for both sexes. A stable trajectory of abstinence was noted for 17 percent of males and 16 percent of females, while 45 percent of females and 46 percent of males maintained infrequent alcohol use across all four waves. Fifteen percent of males showed a trajectory of moderate to high weekly use, a pattern not observed in females. One trajectory unique to females showed 7 percent going from high weekly levels in high school to low weekly levels following high school. Stable high use trajectories were identified for 8 percent of males and 7 percent of females. Findings from this study support previous research demonstrating that patterns of alcohol consumption in high school are important predictors of alcohol use following high school and suggest the value in focusing alcohol health promotion programs on encouraging more students towards a less-than-weekly pattern of alcohol use. 21 references, 3 tables, and 2 figures