Bronchial hyperresponsiveness (BHR) to inhaled histamine has often been cited as the gold standard in asthma diagnosis, but recently this has been questioned. This report assesses the relationship of BHR to asthma symptoms and asthma diagnosis in a large community-based sample of children. A total of 2,053 children 7 to 10 yr of age were randomly sampled from Auckland primary schools and assessed by a questionnaire and histamine inhalation challenge. In all, 14.3% had had asthma diagnosed, 29.6% reported having had one of the four respiratory symptoms in in the previous 12 months, and 15.9% had BHR (PD20 less than or equal to 7.8 mumol histamine). After a cumulative dose of 3.9 mumol histamine, the percent change in FEV1 from postsaline FEV1 was unimodally distributed, with those in whom asthma had been diagnosed dominating the severe end of the spectrum. However, 53% of those with BHR had no asthma diagnosis, and 41% had no current asthma symptoms. On the other hand, 48% of all subjects with diagnosed asthma and 42% of children with diagnosed asthma and current symptoms did not have BHR. Although severity of BHR tended to increase with wheezing frequency, all grades of severity (including no BHR) were found for any given frequency of wheeze. An existing diagnosis of asthma identified symptomatic children more accurately than did BHR, regardless of the criteria used for BHR or for "symptomatic" and irrespective of ethnic group. In conclusion, BHR is related to, but not identical to, clinical asthma. Bronchial challenge testing is an important tool of respiratory research, but cannot reliably or precisely separate asthmatics from nonasthmatics in the general community.