An important part of the new thinking about asthma is control of preventable risk factors, air pollution among them. The epidemiological evidence that ground-level ozone, in particular, is an important risk factor for provoking asthmatic attacks is strong. however, the evidence that it increases the prevalence of asthma overall, or mortality from asthma at levels frequently encountered in the developed world is lacking. The mechanism of airway response to ozone may be complex, involving modification of the atopic response to an antigen, rather than simple inflammation. Likewise, there may be many complexities in assessing the mechanism of association between exposure to air pollution and airways exposure, including susceptibility states, comorbidity from respiratory disorders, tolerance, and the attack rate for common viral infections. Some jurisdictions have proposed to use the frequency of asthmatic episodes as a means to set air pollution standards. The current state of the art, for many of these same reasons, does not support this. Progress will require mechanistic as well as population based research and broader thinking in the analysis of available data. Two hypotheses that especially require critical testing are age-dependent sensitization as a risk factor for asthma, and gradient-threshold response to irritant exposures.