Alcohol is beginning to be recognized as an important risk factor in other transportation, including in boating fatalities. However, little is known regarding the prevalence of drinking on the water. This paper describes the prevalence of alcohol use among recreational boaters on the Chesapeake Bay. From May to October 1997 and 1998, boats were selected at random from those approaching boat ramps, marinas, or boats that were drifting or at anchor. Interviews and self-administered questionnaires were given to the operator and two random passengers, followed by a breath sample. Breath samples were obtained on 82 per cent of boaters (74 per cent of respondents were male). Drinking on the day of the survey was reported by 33 per cent of respondents, but only 24 per cent had a positive BAC: (10 per cent > 50 mg/dl, 5 per cent > 100 mg). Drinking was higher among: 25-44 year olds (30 per cent BAC positive), those in cabin motorboats (34 per cent), those swimming and cruising (33 per cent and 25 per cent retrospectively), and lowest for fishing (9 per cent). Those with positive BACs increased from 4 per cent at 9 a.m. to 40 per cent after 9 p.m. High BACs were similar in operators and passengers, and between men and women. Crude comparison of BAC in boaters in the surv ey to boating fatalities in Maryland show that elevated BACs are much more prevalent in fatalities. Alcohol use on the water is common. Unlike drivers on the highway, the BACs are more likely to be elevated in boaters, and men and women have similar BACs. Data from this prevalence study will also be used in an ongoing case-control study of alcohol and boating fatality risk. Upon completion of case-control analyses, the prevalence data will also be used to estimate attributable risk.