The 1961 birth cohort of Tasmanians (n = 8410) was initially surveyed in 1968 and was followed up in 1974 for the history and presence of respiratory symptoms and signs. The study was designed to describe and compare the natural histories of wheeze and productive cough. In 1981 a 10% stratified random sample of the original cohort was again followed up and this sample's responses were compared across the three surveys. Wheezing was found to be more persistent than productive cough. Wheezers were twice as likely as coughers to persist with their symptoms to the age of 20. These persistent wheezers represented 3.6% of the cohort available for follow-up in 1974 (n = 7132). Wheezing was also invariably associated with decreased spirometric performance, particularly FEV at 0.5 s and FEF 25-75. The new wheezers first detected in 1974 had had significantly decreased spirometry measurements in 1968 compared to those of their coughing or symptom-free peers. An early history of cough was associated with triple the incidence of wheeze in previously asymptomatic children. Hayfever and eczema were associated with persistence of wheeze. A child with either of these diagnoses was four times as likely to persist in wheezing to the age of twenty than a wheezer without atopic complaint. No significant associations could be demonstrated between wheezing and smoking.