This study was conducted to test the supposition that both smoking and consuming alcohol suppress host resistance to viral infections. The relations between smoking, alcohol consumption, and the incidence of documented clinical colds were prospectively studied among 391 subjects intentionally exposed to one of five respiratory viruses and 26 subjects given saline. Clinical colds were defined as clinical symptoms verified by the isolation of virus or by an increase in virus-specific antibody titer. Analyses included control variables for demographics; body weight; virus; and environmental, immunological and psychological factors. Smokers were at greater risk for developing colds than nonsmokers because smokers were more likely both to develop infections and to develop illness following infection. Greater numbers of alcoholic drinks (up to three or four per day) were associated with decreased risk for developing colds because drinking was associated with decreased illness following infection. However, the benefits of drinking occurred only among nonsmokers. Susceptibility to colds was increased by smoking. Although alcohol consumption did not influence risk of clinical illness for smokers, moderate alcohol consumption was associated with decreased risk for nonsmokers.