There is a lack of established criteria to identify asthma and bronchial hyperreactivity (BHR) in epidemiological studies, although both conditions appear to bear some relationship to atopy, at least in children. Recent studies indicate a low prevalence of atopy in former Socialist countries in Europe, yet the prevalence of BHR has been reported to be high. We have analysed the relationship between the outcome of various lung function tests, atopy and clinical symptoms of bronchial asthma in an epidemiological survey of Estonian 10-12 year old schoolchildren. Metacholine provocation test (four steps with the cumulative doses 100, 300, 700 and 1100 micrograms), exercise challenge test and PEF-variability over two weeks were done in 806 children in Tallinn (coastal, industrialised city) and 774 children in Tartu (inland, university town). A positive response to the metacholine challenge test was recorded in 19% in Tallinn and in 32% in Tartu (p < 0.001). A similar tendency was observed for a more than 15% decrease of FEV1 in the exercise challenge test, i.e. 6% in Tallinn and 18% in Tartu. There was only a weak relationship between BHR, as defined by either a positive metacholine challenge and/or exercise test, diagnosed asthma and reported wheezing. Thus, 47% of the wheezing children and 30% of the children with asthma had negative test results. Only 17% of the children with a positive metacholine challenge were atopic, as defined by at least one positive skin prick test. In conclusion, none of the methods employed to assess bronchial hyperresponsiveness were very useful for the identification of wheezing and asthmatic children in this epidemiological study. In contrast to the results of studies in Western Europe, most children with bronchial hyperreactivity in Estonia are not atopic.