90-day subchronic toxicity studies with 3-methyl-1-butanol (MEB) and 2-methyl-1-propanol (MEP) were performed on rats to evaluate the toxicological profile of the compounds under conditions of drinking water studies, to identify the potential target organs, and to determine no-observable-adverse-effect-levels (NOAELs) respective of the substances. The test substances were administered to groups of 10 male and 10 female Wistar rats in drinking water at concentrations of 0, 1000 p.p.m. (about 80 mg/kg/d), 4000 p.p.m. (about 340 mg/kg/d) and 16,000 p.p.m. (about 1250 and 1450 mg/kg/d of MEB and MEP respectively). 2. 16,000 p.p.m. was found to be the maximal concentration for both alcohols applicable to rats in drinking water. Higher concentrations had an influence on palatability and could thus not be tested in drinking water studies. 3. At 16,000 p.p.m. MEB a marginal increase in the red blood cell count as well as a slight decrease in the mean corpuscular volume and the mean corpuscular hemoglobin content was observed in males only. These changes are considered to be treatment-related, although the toxicological significance of these findings is unclear. No other substance-related effects were found on body weight (b.w.), mortality, various parameters of clinical chemistry, organ weights, gross pathology and histopathology. 4000 p.p.m. MEB did not cause any substance-induced changes. Therefore, the NOAEL of MEB was defined as 4000 p.p.m. for male and 16,000 p.p.m. for female rats under conditions of oral application via drinking water. 4. MEP concentrations up to and including 16,000 p.p.m. did not induce any signs of toxicity and were therefore defined as the NOAEL respective of this substance for rats under conditions of drinking water application.