The usefulness of the pulmonary function variable, area under the expiratory flow-volume curve (AEFV), in methacholine provocation (MP) studies in normal subjects was evaluated. The baseline coefficients of variation (CV), maximal fall from post-saline values (MAX) and dose-response slope [DRS = maximal percentage fall in pulmonary function/maximal noncumulative methacholine dose (mumol)] were calculated for AEFV, and were compared to those of forced expiratory volume in one second (FEV1), maximum expiratory flow at 50% (MEF50%) and at 25% (MEF25%). Also the repeatability after 2 and 8 weeks was assessed. The rank order of CVs was FEV1 less than AEFV less than MEF50% approximately MEF25% all differences, except MEF50% vs. MEF25%, being significant (P less than 0.05). The order of sensitivities (estimated with MAX and DRS) was FEV1 less than AEFV less than MEF50% less than MEF25%. Again, all differences were significant (AEFV vs. MEF50% P less than 0.01, others P less than 0.001). After two weeks the correlation coefficients for MAX- and DRS-values of AEFV were 0.84 and 0.94 (P less than 0.001), respectively. After an eight week period the correlations were still high, 0.86 and 0.92 (P less than 0.001), respectively, but the actual MAX- and DRS-values tended to be smaller than eight weeks before. The repeatability of MAX of AEFV was not quite as high as that of FEV1. On the other hand, the correlations of DRS-value derived from AEFV were markedly better than those derived from FEV1. In conclusion, MP in normal subjects is well repeatable at least within two weeks, but after an eight week period disturbing factors may intervene reducing repeatability. AEFV can be considered at least as good a pulmonary function variable as FEV1 in demonstrating bronchoconstriction during MPs in normal subjects, and it might even be a better variable to obtain DRS-values than other variables tested.