Analysis of Dutch accident statistics shows a major drop in involvement of drivers in alcohol-related injury accidents following the 1974 legislation. However, since mid-1975, a sharp rise in accident involvement suggests that the positive effects of the November 1974 law were temporary. Next, an examination of Rotterdam DWI reports for 1965, 1975, and 1976 shows that the number of accidents involving alcohol was considerably greater in 1975-76 than in 1965, and that the police detected more DWI drivers by roadside surveys in 1976 than in 1975. There were significantly more lower BAC's in 1975-76 than in 1965, but the number of drivers were using drugs in combination with alcohol. Changes in the 10-year period in DWI offender characteristics show that males still constituted the majority of cases, although the number of females was increasing rapidly. Drunken drivers apprehended in 1975 were substantially younger than the 1965 offenders. No significant differences were detected by social class, marital status, or place of residence. A comparative analysis of DWI cases detected by driving behavior, cases detected by accidents occurring, and cases detected by roadside surveys revealed significant differences in BAC levels and use of drugs between the accident and roadside survey groups. However, the absence of significant differences on eight other variables makes it likely that the drinking drivers detected by roadside surveys probably typify the driving behavior and accident offenders of the future. However, roadside surveys were found to be an effective deterrent to driving while under the influence of alcohol. One figure, 14 tables, and 13 references are provided.