The closing volume test, using helium bolus and nitrogen washout techniques simultaneously, was evaluated in more than 1,900 persons randomly selected as representative of the white population of Tucson, Arizona. Normal values were based on data obtained from those totally asymptomatic persons who had never smoked cigarettes. In the remaining population, the slope of Phase III appeared to be more sensitive than any other measurement in detecting abnormalities, but all measurements taken in conjunction were superior to any single parameter. Closing volumes measured by the bolus method were systematically greater than those obtained by resident nitrogen technique in normal subjects and were slightly more sensitive in detecting abnormalities in other population subgroups. Nevertheless, the single-breath test was of only limited value in distinguishing asymptomatic smokers from nonsmokers. Parameters used in conjunction revealed abnormalities in 36% of symptomatic subjects, many of whom already had physician-confirmed obstructive pulmonary disease. Our data suggest that, when applied to a randomly selected population, abnormalities in the closing volume test occur in only a very small proportion of persons in whom a respiratory disorder would not otherwise be suspected or already diagnosed.