Our aim was to examine the contribution of an infant's indoor environment to childhood asthma using prospective data. We conducted a cross-sectional asthma survey in 1995 on 92% (6,378/6,911) of 7-year-olds in Tasmania, Australia. We linked these data with data collected in 1988 as part of the Tasmanian Infant Health Survey, which was designed to investigate sudden infant death. We were able to match 863 records out of the 1,111 in the 1988 survey and the 6,378 in the 1995 survey. The former group was interviewed at home at 1 month of age. In homes where at least one adult smoked in 1988, reported infant exposure to smoking in the same room in 1988 was associated with increased asthma by 1995 (relative risk = 1.52; 95% confidence interval = 1.01-2.29) after adjustment for confounders. The associations between infant exposure to environmental tobacco smoke and asthma were not consistent, however. Gas heater use in 1988 was associated with asthma (relative risk = 1.92; 95% confidence interval = 1.33-2.76). Markers of aeroallergen exposure at 1 month of age were not materially associated with asthma or wheeze. In some settings, air circulation practice with regard to bedroom door closure appeared important. Poor indoor air quality may play an important role in the development of childhood asthma.