Times allowed for drinking were 15, 20, or 25 minutes for the different levels of alcohol consumption. Specimens of capillary (fingertip) blood were taken in triplicate at 30-minute intervals for 2 hours and then at 60-minute intervals for a total time of 6 to 8 hours, depending on the alcohol dose. Subjects emptied their bladders before the start of drinking and provided urine specimens at 60-minute intervals for 6 to 8 hours thereafter. Results showed that mean ratios of ethanol concentration (urine alcohol concentration-UAC/blood alcohol concentration-BAC) were mostly less than unity during the absorption phase. Mean UAC/BAC ratios varied between 1.4 and 1.7 when the BAC exceeded 0.50 mg/mL. When the BAC decreased below 0.40 mg/mL, UAC/BAC ratios increased appreciably. In 654 apprehended drinking drivers with a mean BAC of 1.55 mg/mL, the UAC/BAC ratio of ethanol varied widely, with a mean value of 1.49. In 12 subjects (3.2 percent), the ratio was less than or equal to unity. In a second specimen of urine obtained about 60 minutes after the initial void, the mean UAC/BAC ratio was 1.35. The author concludes that the magnitude of the UAC/BAC ratio can help establish whether the BAC curve rises or falls at or near the time of voiding. Further, the status of alcohol absorption needs to be documented if drinking drivers claim ingestion of alcohol after the offense or when back-estimation of BAC from the time of sampling to the time of driving is required by statute. 32 references, 2 tables, and 9 figures (Author abstract modified)