AbstractConventional criteria for spirometric impairment (FVC and FEV1⩽ 79% predicted, FEF25‐75⩽ 74% predicted, FEV1/FVC ⩽ 0.69 for most ages) were applied to 507 healthy nonsmoking male subjects in the 1971 Oregon survey. Frequency of air flow impairment was higher than expected for a normal population. Of all subjects, 17.8% had an “abnormal” FEF25‐75%and 19.7% an “abnormal” FEV1/FVC. “Decreased” lung volumes were seen in older subjects; of those over 55 years of age, 8.8% had a reduced FVC and 11.8% a reduced FEV1. These rates in a normal population may be used as a rough baseline for making inferences about a population under study. Similar rates were found in a cross‐sectional survey of nonsmoking males representative of the population of a large industrial state.The major reason for these high rates of “abnormal” spirometry lies in the conventional definitions of abnormality. Tables showing cut‐off points for abnormal spirometric values based on a 95% confidence interval for adult males of all ages and heights permit an alternative method f