A population-based incident lung cancer case-control study was conducted among New Jersey women in 1982-1983. Interviews were completed for 994 cases and 995 controls. The association of lung cancer with cigarette smoking had an overall age-, race-, and respondent type-adjusted odds ratio of 8.5, with risks of 11.1, 62.6, and 3.9 for squamous cell, small cell, and adenocarcinoma, respectively. Analyses for squamous cell carcinoma showed significantly higher odds ratios for self-respondents than for subjects with next of kin respondents; this heterogeneity may be partially explained by misclassification in next of kin reports. Analyses for adenocarcinoma showed a significantly lower smoking-related risk among older women, which may be related to the later age at which these women began smoking. Analyses by histologic subtype showed lower risks for poorly differentiated squamous cell carcinoma and higher risks for poorly differentiated adenocarcinoma. This could be due to misclassification of histologic type and suggests that the actual smoking-related risk for squamous cell carcinoma might be higher than observed, and that for adenocarcinoma might be lower. Comparison of the results in this study with those from a similar study of lung cancer among men showed little difference between men and women in the magnitude of the association with smoking, or in the patterns of risk with intensity and duration.