The aim of the study was to assess the seasonal variability of non-specific bronchial responsiveness to methacholine in allergic asthma. One hundred sixty-five patients (83 male and 82 female) entered the study: 86 subjects (group A) with allergy exclusively to mites and 79 (group B) with concomitant allergy to pollens, e.g., "Graminae" and "Parietaria." Inclusion criteria were the absence of sensitization to other allergens, no smoking habit, withdrawal from steroids, bronchodilators, sodium cromoglycate, and antihistamines for at least four weeks before enrollment, FEV1 > 70% of the predicted value, and absence of other respiratory diseases and of upper and lower respiratory tract infections for at least one month before the methacholine challenge. None of the patients had been previously treated with specific immunotherapy. Subjects of each group (A and B) underwent methacholine challenge at first visit and were divided into four subgroups according to the period when the challenge was performed. Subgroups A1 and B1 performed the challenge in December, January, and February; subgroups A2 and B2 in March, April, and May; subgroups A3 and B3 in June, July, and August; subgroups A4 and B4 in September, October, and November. PD20 values were expressed as the natural logs of the cumulative dose of methacholine causing at least a 20% fall in FEV1. Statistical analysis was carried out using multiple group analysis and Student's t-test. Results showed that the highest non-specific bronchial responsiveness was observed in autumn (ln PC20 = 4.54 +/- 1.51) in patients allergic to mites only (group A), and in summer (ln PC20 = 4.72 +/- 2.11) in those of group B. Multiple group analysis showed statistical significant differences between subgroups within each group (group A, p = 0.039; group B, p < 0.001). In patients allergic exclusively to house dust mites (group A), multiple comparisons and Student's t-test showed statistically significant differences between non-specific bronchial responsiveness (NSBR) assessed in autumn and those of other seasons (winter, p = 0.002; spring, p < 0.001; summer, p = 0.082). These results confirm that the level of allergen exposure may influence NSBR. Mite-allergic patients showed an increase of NSBR in autumn, possibly as a consequence of higher indoor mite concentration. However, mite- and grass-allergic patients had wider variations of NSBR, possibly reflecting changes in seasonal pollen concentration.