The Davidson and Hemmendinger (DH) color rule was evaluated for color vision screening of normal and congenital color-defective subjects. Ninety-eight normal and 14 color-defective subjects were tested on the color rule under Macbeth illumination of 5,400 K. The color-defective subjects were also tested on the Nagel anomaloscope, the Farnsworth D-15, and the H-R-R pseudoisochromatic plates. The DH color rule performed as accurately as the anomaloscope and was superior to the other two tests in detecting anomalous trichromats and in discriminating protanomalous subjects. The color rule also discriminated dichromats from anomalous trichromats. For severe color-defective subjects (dichromats, achromats), the color rule was more time-consuming than the other tests and discrimination was less certain. Response patterns on the DH color rule and response variability of the different classifications are reported.