The purpose of this study was to use three-dimensional imaging methods to measure the palatal surface of unrepaired cleft patients. The surface area of the palate was defined and measured on three-dimensional computed tomography images of dental plaster models in four different groups of cleft patients at 3 months of age. There were 30 unilateral complete cleft lips and palates (UCLP), 27 bilateral complete cleft lips and palates (BCLP), 23 isolated cleft palates of incomplete form (CP), and 19 unilateral cleft lips without cleft palates (UCL). These patients were nonsyndromic, unoperated, and without other major deformities. The dental casts were scanned, and the computed tomography data were transferred to an imaging laboratory for processing and reconstruction of three-dimensional images. Surface area of the palate was delineated, which was defined as within the alveolar crest and the line connecting both tuberosities. In UCLP and BCLP, the edge of cleft formed the medial boundary of the area for each palatal shelf, and the palatal surface area was the combination of both palatal shelves and the premaxillary area in BCLP group. The surface area was measured. Repeated definition and measurement tasks were performed for calculation of errors. The imaging data management and measurement were performed using the Analyze program (Biomedical Imaging Resource, Mayo Foundation, MN). In addition, linear distances were measured between the canine points on the alveolar crest (line C) and the tuberosity points (line T). The measurements were compared among the different groups. Analysis of variance and multiple comparisons were used for statistical analyses. The results showed that the mean error between repeated area definitions and measurements in this study was 1.86%. The bilateral complete cleft lip and palate (BCLP) and unilateral complete cleft lip and palate (UCLP) groups had significantly smaller palatal surface area than the unilateral cleft lip without cleft palate (UCL) and isolated cleft palate of incomplete form (CP) groups. There was no significant difference between the BCLP and UCLP groups. Line C and line T distances were significantly longer in BCLP and UCLP groups than in UCL and CP groups. The findings suggest that compared with UCL and CP patients, there is an intrinsic tissue deficiency in the palate/maxilla of BCLP and UCLP patients.