The relationship between alcohol consumption (AC) has been repeatedly confirmed. However, the respective contribution of the various types of beverages has not been clearly established. The cross-sectional data of the initial examination of the Paris prospective study II of the GREA, concerning 4547 male civil servants were thus analysed. Among subjects without any antihypertensive medication, systolic (SBP) and diastolic (DBP) blood pressure were positively associated with total AC; the differences between the first and the fifth quintile were respectively 6 and 3 mmHg (p less than 0.01 for wine, beer and spirits consumption). Using a linear combination of wine, beer and spirits consumption significantly improved the prediction of BP, as compared to total AC (p less than 0.001). In the multivariate analysis including age and body mass index (BMI), the consumption of 40 ml of alcohol from beer was associated with an increase of 5.7 mmHg for SBP and 2.6 mmHg for DBP (p less than 0.001). The elevation was 2.3 (SBP, p less than 0.001) and 0.8 mmHg (DBP, p less than 0.01) for wine consumption. Spirits consumption was associated with DBP (1.4 mmHg, p less than 0.001), but not with SBP. In conclusion, a positive relationship was observed with each of the three types of alcoholic beverages studied; however, this association was more pronounced for beer than for the two other beverages.