Purpose. The Lanthony Desaturated Panel D-15 has been used to measure fine color discrimination for congenital and acquired color vision defects. This study investigated the test-retest reliability of the test using an intertest interval of approximately 1 month. Methods. One hundred twenty-six color vision normals (mean age = 34.5 years) were administered several color vision tests, including the Lanthony Desaturated D-15. Normal color vision status was confirmed using the anomaloscope and HRR color plates. The color vision tests were readministered 3 to 6 weeks after initial testing. The results of the Lanthony test were expressed using the color confusion index of Bowman. The difference in Color Confusion Index (CCI) between the two administrations was calculated and used to determine the intraclass correlation coefficient. Results. The overall mean CCI for the two administrations for these subjects was 1.11 ± 0.136. The mean difference in CCI score between test administrations was -0.02 ± 0.128. There was a strong correlation between the absolute value of the difference in CCI and the mean CCI for each subject (r = 0.51, p < 0.0001). The intraclass correlation coefficient was 0.56 (95% confidence interval, 0.43-0.67). Conclusions. Although the Lanthony Desaturated D-15 test can be used to assess fine color discrimination, there is considerable within-subject variability in test results. The intraclass correlation coefficient is less than that recommended for use in clinical testing or research. Clinicians should consider at least three administrations of the test at each sitting to ensure precision and we recommend taking the mean of those three tests. Copyright © 2005 American Academy of Optometry.