OBJECTIVE The purpose of this study was to identify and compare lateral cephalometric measurements in noncleft parents of children with cleft palate (CP) and cleft lip (CL), and cleft lip and palate (CLP). The hypothesis was that discriminant analysis would enable identification of morphometric features that predispose to orofacial clefting and that differ for CP, CL, and CLP and are unevenly distributed within parental pairs. METHODS This was a prospective, parametric analysis. METHODS The study was conducted by the Department of Dental Health, University of Dundee, and the Department of Statistics, University of Glasgow, Scotland. METHODS From a completely ascertained sample of 286 children with cleft lip and/or palate born in the West of Scotland between January 1, 1980, and December 31, 1984, a sample of 83 parents of children with non-syndromic clefts volunteered for lateral cephalometric examination. METHODS Thirty-seven cranial and 99 facial landmarks were identified and 37 linear, angular, and area parameters were used to describe the craniofacial skeleton. Analysis of variance was used for a three-way comparison of CL/CLP/CP, and stepwise discriminant analysis was used to determine which variables discriminate best between cleft lip with or without cleft palate [CL(P)] and isolated cleft palate (CP) parents. RESULTS There were no significant differences whatsoever in the craniofacial morphology between the parents of children with CL and CLP, but differences were found between the CL(P) and CP groups. The most significant of these were in mandibular length, ramus length, mandibular area, and cranial area. Mandibular ramus length alone discriminated between the two groups in 71.4% of CP and in 62.5% of CL(P) cases, while separate analysis of fathers and mothers showed that ramus length and cranial height together reliably distinguish between mothers in 75% of CP and 80% of CL(P) cases. CONCLUSIONS Previous studies suggests that unaffected parents with non-syndromic children with cleft lip and/or palate have differences in their craniofacial morphology when compared to the general population. This study indicates that these morphologic features differ for CP and CL(P).