Particulate pollution levels were found to decrease, on average, over the 10-day period preceding SIDS incidence in a study of 369 cases over 15 years in Brisbane, Australia. This unexpected association could not be explained by correlations between daily pollution levels and daily precipitation, wind velocity, air temperature or cloud. Of the meteorological variables studied, which might have given rise to the particulate-SIDS associations, only visibility showed associations with SIDS incidence. Average visibility increased prior to SIDS in both summer and winter cases. It is hypothesised that these findings may be due to either direct effects of light, or due to increased outdoor exposure of infants and/or to changes in parental behaviour during clear weather.