Segregation analysis was performed on a subset of a large body of French data comprising 298 nuclear families. Two models were used in this analysis: the transmission probability model [Elston and Stewart, 1971; Elston, 1981] and the mixed model [Morton and MacLean, 1974]. Both models are consistent with familial aggregation of neural tube defects, in this sample, being due to either the segregation of a recessive major gene or a sibling environmental effect, or both factors. In each case, other environmental factors are also involved. These results were compared to the findings of other studies and discussed in respect to the diversity of the epidemiological features displayed by different populations. Some observations of vertical transmission in a British study and the proportion of affected first cousins, in both France and Great Britain, lead us to reject a possible absence of transmission. We propose a monogenic component with a large influence of environmental factors, some of which may be common to sibs, to explain the occurrence of neural tube defects in this sample.