Orientals are at higher risk for cleft lip with or without cleft palate (CL +/- P) than Caucasians or blacks. We collected demographic and family data to study factors contributing to the etiology of CL +/- P in Shanghai. The birth incidence of nonsyndromic CL +/- P (Shanghai 1980-87) was 1.11/1,000, with a male/female ratio of 1.42. Almost 2,000 nonsyndromic CL +/- P probands were ascertained from individuals operated on during the years 1956-83 at surgical hospitals in Shanghai. Detailed family histories and medical examinations were obtained for the probands and all available family members. Genetic analyses of the probands' families were performed under the mixed model with major locus (ML) and multifactorial (MFT) components. The hypotheses of no familial transmission and of MFT alone could be rejected. Of the ML models, the autosomal recessive was significantly most likely and was assumed for testing three complex hypotheses: (1) ML and sporadics; (2) ML and MFT; (3) ML, MFT, and sporadics. None of the complex models were more likely than the ML alone model. In conclusion, the best-fitting, most parsimonious model for CL +/- P in Shanghai was that of an autosomal recessive major locus.