BACKGROUND In addition to their value in assessing pulmonary health and disease, spirometric variables have been shown to be powerful predictors of time until death in aging populations. The sources of variability in these spirometric values are consequently of relevance to basic gerontological research, and also of potential value in clinical application. The objective of this study was to estimate genetic and environmental sources of variance in pulmonary function. METHODS The study involved 230 Swedish twin pairs (mean age = 64.9 years), of which number 37 monozygotic (MZ) pairs and 72 dizygotic (DZ) pairs had been separated and reared apart. Comparing these groups to the 57 MZ and 64 DZ pairs reared together permits stronger interpretation than that of conventional twin studies. Measures of vital capacity (VC) and forced expiratory volume in one second (FEV1) were residualized for height, age, sex, and tobacco consumption in pack-years. RESULTS Maximum likelihood analyses of VC and FEV1 gave heritability estimates of .48 and .67, respectively. Age effects were explored both by dividing the sample into two cohorts, respectively above and below 65 years, and by moving interval analysis. In the two-cohort analysis, heritabilities were somewhat higher for the older cohort than the younger cohort for FEV1. The opposite was true for VC: heritability was lower in the older cohort, and there was evidence for a shared rearing environmental effect for this group. Moving interval analysis suggests these differences are gradual rather than saltatory. There were no gender differences in parameter estimates. CONCLUSIONS Genetic factors account for between one-half and two-thirds of the variability in pulmonary function. There is a suggestion of age differences in the relative importance of genetic and environmental influences.