We studied 15 normal subjects and 15 patients with cystic fibrosis (CF), to determine the within-subject variability of spirometry, moment analysis, and slope ratios, and to calculate the per cent change for significance. Common sources of variability, such as patient and technician training, equipment errors, lung volume history, posture, and circadian variation, were controlled. Flow-time curves were analyzed with a digital computer. The coefficient of variations of spirometry were smallest for forced vital capacity (FVC), forced expiratory volume in one second (FEV1), and peak flow (PF), and were largest for root moments (M) and slope ratios (SR) in both normal subjects and patients with CF. The coefficient of variations of all tests, except PF, were larger in patients with CF than in normal subjects. Significant change was calculated from the sample variance, mean, and the number of trials using a modified equation for sample size. This calculation estimated the degree of change for significance for each participant that was required in subsequent testing.