The relationship between self-reported frequency of drunkenness at 14-16 years of age and registered alcohol abuse at age 15-25 was studied in a large and representative group of Swedish men. The results showed that the proportion of boys who had experienced drunkenness more than 10 times increased rapidly between ages 14-16. Before the age of 25, 17% of the boys were registered in governmental records (police, social authorities or psychiatric services) because of alcohol abuse. A high frequency of self-reported drunkenness at age 14-16 was found to be significantly related to registered alcohol abuse at age 18-24. At the same time, however, it was found that 70-80% of the adolescent boys with the greatest frequency of drunkenness were not registered for alcohol abuse in early adult years. Using self-reported frequency of drunkenness at age 14-16 as a basis for predicting registered alcohol abuse at age 18-24 allowed correct classification of only 6% more boys than would have been expected by using a random procedure. The present study indicated that initial drinking habits per se are of limited importance in the development of alcohol abuse in early adulthood. However, the results showed that the combination of high self-reported frequency of drunkenness and appearance in government registers as early as age 15-17 constitutes a serious indication of continuing alcohol abuse.