Although several studies have demonstrated familial aggregation of nonsyndromic cleft palate (CP), the mode of inheritance still remains uncertain. We report the results of complex segregation analysis performed in families of 357 consecutive newborns affected with nonsyndromic CP (i.e., CP not a component feature of malformation syndrome, sequence or association), and registered in the North East Italy and Emilia Romagna congenital malformation registries in the period 1981-1993. This sample, based on a large number of consecutive births, in a well-defined geographical area, with quality control to detect associated anomalies and malformation syndromes, is independent of the number of affected subjects in the family and of CP severity, fitness, and survival. We have analyzed, using the mixed model, the whole sample of nonsyndromic CP, including isolated (i.e., without other anomalies) CP (CPI) and CP associated with at least one other anomaly (CPA), for which a diagnosis of malformation syndrome was not possible. When nonsyndromic CP (including CPA) are considered in the analysis, there is no heterogeneity between CPA and CPI nor between CP including hard palate (CPH) and CP of the soft palate only (CPS). POINTER and COMDS programs cannot discriminate between alternative genetic models; only the hypothesis of non-genetic transmission is rejected. The COMDS analysis two-locus model, which indicates that a modifier locus (or loci) operates in addition to a single major locus (SML), does not show evidence of better fit than SML, polygenic, and multifactorial models. When the severity parameter (defined as CPH and CPS) is added, CPI and CPA show heterogeneity. Eventually, when the analysis is limited to CPI and includes information on severity, a recessive SML, with low penetrance and determining CPH, provides a significant best fit. To have defined a genetic model for CPI and provided evidence for SML inheritance suggests that genetic linkage studies could be implemented. This conclusion is in agreement with previous studies which showed a significant association between alleles of transforming growth factor alpha and CP only in humans, and that single recessive genes may play a crucial role during palatogenesis in mice as well as in Brittany spaniels. Application of the candidate genes to human CPH families could reveal whether these genes are involved.