Purpose: The aim of this study was to measure the thickness of bone labial and palatal to maxillary anterior teeth on cone beam computed tomographic (CBCT) images and to compare these measurements with direct clinical measurements to determine the reliability and accuracy of CBCT. Materials and Methods: Eighteen healthy subjects were randomly selected from among candidates for immediate implant placement in the anterior maxilla. After extraction, labial bone thickness was measured at 1, 4, and 8 mm from the bone crest. Palatal bone thickness was also measured at 1 and 4 mm from the bone crest. The same measurements were performed on presurgical CBCT images. The CBCT measurements were compared to the direct measurements, and their accuracy and reliability were assessed by Pearson correlation coefficients and intraclass correlation coefficients, respectively. Results: The mean width of labial bone was 0.50 ± 0.32 mm and 0.76 ± 0.37 mm for direct and CBCT measurements, respectively. Average thickness of the palatal bone was 1.16 ± 0.53 mm and 1.41 ± 0.51 mm for direct and CBCT measurements, respectively. The mean absolute error and mean relative error of CBCT measurements compared to direct measurements were 0.28 ± 0.29 mm and 0.60 ± 0.84 mm, respectively. The Pearson correlation between CBCT and direct measurements was 0.795 (P .001) and the intraclass correlation coefficient between direct and CBCT measurements was 0.840. The correlation between the measurement series increased significantly when the measured bone was more than 1 mm thick. Conclusion: CBCT measurements of labial bone mostly overestimated bone thickness. CBCT has relatively good accuracy and reliability for measurement of labial bone thickness when the alveolar bone is thicker than 1 mm. However, most subjects have labial bone thinner than 1 mm; therefore, CBCT could result in large errors in many patients. Keywords: alveolar bone, cone beam computed tomography, dental implants, esthetic zone, reproducibility of results