During 1989 and 1990, the Civil Aeromedical Institute received specimens from 975 victims of fatal aircraft accidents. The maximum concentration of ethanol allowed under FAA regulations (0.04%, 40 mg/dL) was exceeded in 79 of these cases (8%). It was determined based on the distribution of ethanol in urine, vitreous humor, blood, and tissue that 21 of the positive cases (27%) were from postmortem alcohol production. Twenty-two of the positive cases (28%) were found to be from the ingestion of ethanol. In 36 cases (45%), no determination could be made regarding the origin of the ethanol. In two cases, postmortem alcohol production exceeded 0.15% (150 mg/dL). The opinion held by some toxicologists that postmortem alcohol production can be inferred from the presence of acetaldehyde, acetone, butanol, and other volatiles was found to be incorrect. Several cases with postmortem ethanol had no other volatiles. Volatile compounds were found in several cases where no ethanol was present. In addition a case was found in which the relative ethanol concentrations in blood, bile, and vitreous humor were solely consistent with the ingestion of ethanol, but acetaldehyde, acetone, and 2-butanol were also found in blood. This clearly indicates that the presence or absence of other volatiles does not establish postmortem ethanol production.