Increased constraints on access to alcohol resulted from the closure of the sole hotels in two "experimental" towns. This afforded a natural experiment to study the effects of the change in availability of alcohol on consumption. Dependent measures were derived from public records of liquor sales by all licensed premises, and from computerized police records of drunk-driving, criminal offenses, and serious offenses in both the two experimental and four "control" towns. In addition, interview measures of patterns of consumption, the context for drinking, and leisure activities were compared for the two experimental towns and two of the control towns. When comparisons were made to the period before hotel closures or to the control town, the following results were found: (1) a marked reduction in liquor sales; (2) a significant decline in consumption; (3) a change in the pattern of consumption mediated by a change in the context from drinking in hotels to drinking at home, or in clubs; (4) residents engaged in more sporting and outdoor leisure activities; (5) drunk-driving offenses in experimental towns showed a significant decline; and (6) criminal offenses and serious offenses significantly decreased. (Contains 139 references.) (JBJ)