The associations between cigarette smoking and alcohol consumption and laryngeal cancer were examined in a case-control study carried out between 1982 and 1985 in Madrid. The analysis was based on 50 histologically confirmed male cases and 103 age- and sex-matched controls (45 hospitalized and 58 from the general population). A dose-response effect was observed for cigarette smoking, with an odds ratio (OR) of 4.33 (95% confidence interval of 1.22 to 15.41) for smokers of 30 or more cigarettes per day, when compared with those smoking less than 10 cigarettes per day. The risk clearly rose in line with the length of the smoking habit. In addition, for smokers of black tobacco, the risk was more than double that for smokers of blond tobacco, irrespective of the depth of inhalation. ORs for alcohol consumption rose significantly in accordance with the average grams intake per week and the overall lifetime consumption, but not with years of drinking. The time trends of risks for duration of alcohol consumption suggest the existence of phenomena related to individual susceptibility. A dose response effect was observed in supraglottal and glottal tumors. The effect of the joint exposure to both tobacco and alcohol fit to a multiplicative model.