Epidemiological data indicated that allergic rhinitis often coexists with and may precede the development of reactive airway disease. In particular, ARP with BHR are more likely to develop asthma. However, the pathogenesis of BHR associated with allergic rhinitis is still remains uncertain. Therefore we designed the study on ARP with/without BHR. The aim of this study were to investigate the presence of an inflammatory process in lower respiratory tract in ARP and to relate these changes to airway responsiveness Eleven ARP with BHR (Group I), eleven ARP without BHR (Group II) and two control patients (Control group) were studied. All of the ARP were judged atopic on the basis of positive skin prick test to common inhalant allergens. Bronchial challenges were performed with increasing concentration of M. All the subjects underwent fiberoptic bronchoscopy, BAL and bronchial biopsies were obtained for pathologic examination. The mean total cell and the mean percentage of macrophages, lymphocytes, neutrophils and eosinophils in BAL fluid were in normal range in all groups without any significant differences between the groups. There weren't any correlation between PC20 to M and the total cell counts and percentage counts of these cells. In bronchial biopsy samples, the absolute numbers of lymphocytes, neutrophils, eosinophils and mast cells in the submucosa showed no differences between the three groups. The epithelial shedding was more extensive in ARP than control subjects (p = 0.05). The thickness of the epithelium was prominent in Group I (p < 0.05) but there was no significant differences in the basement membrane thickening between the three groups. We could only find an inverse correlation between PC20 to M and the mast cell counts in the submucosa (r xy:-0.815 p < 0.05). In conclusion, we couldn't observe any prominent morphological changes which indicate that may cause of BHR in ARP except the increased epithelial shedding in Group I. However, the increased epithelial shedding is not a reliable criterion to comment because of the possibility of mechanical damage of bronchial biopsies caused by the forceps.