Allergic rhinitic subjects without symptoms of asthma show airway hyperresponsiveness, but to a lesser degree than asthmatics. As airway responsiveness is a determinant of the bronchial response to allergen, rhinitic subjects should also respond to allergen challenge, but to a lesser extent than asthmatics. However, studies have so far failed to show quantitative differences in allergen responses between patients with rhinitis and patients with asthma. We studied 123 allergic subjects classified, on the basis of a scored symptom questionnaire, as follows: pure rhinitics without any symptom of asthma (Group 1, n = 39), true asthmatics with or without rhinitis (Group 2, n = 41), and subjects with borderline symptoms of asthma (Group 3, n = 43). All subjects underwent both methacholine and allergen inhalation challenges, with pollen challenges performed out of season. When the three groups were pooled, the asthma symptom score was directly correlated with the sensitivities both to methacholine and allergen, whilst both the sensitivity to allergen and the severity of late-phase response were correlated with the sensitivity to methacholine. The percentage of subjects with a positive early-phase asthmatic response to allergen was similar in Groups 1 and 2. Group 2 had higher sensitivities both to methacholine and to allergen than Group 1. A late-phase asthmatic response occurred more frequently in Group 2 than in Group 1, and this difference was due to a higher occurrence of late-phase response in subjects allergic to house dust mite in Group 2. This study confirms that the bronchial response to allergen can be predicted, in rhinitic as well as in asthmatic allergic subjects, on the basis of airway responsiveness to methacholine. We conclude that the presence or the absence of asthma symptoms in allergic subjects may be related to a quantitatively different airway responsiveness to allergen.