Colour vision defects associated with ocular disease have been reported since the 17th century. Köllner1 in 1912 wrote an acute description of the progressive nature of colour vision loss secondary to ocular disease, dividing defects into “blue-yellow” and “progressive red-green blindness”.2 This classification has become known as Köllner's rule, although it is often imprecisely stated as “patients with retinal disease develop blue-yellow discrimination loss, whereas optic nerve disease causes red-green discrimination loss”. Exceptions to Köllner's rule3 4 include some optic nerve diseases, notably glaucoma, which are primarily associated with blue-yellow defects, and also some retinal disorders such as central cone degeneration which may result in red-green defects. Indeed, in some cases, there might be a non-specific chromatic loss.

Colour vision defects in glaucoma have been described since 18835 and although many early investigations indicated that red-green defects accompanied glaucomatous optic neuropathy,1 later studies suggested that tritan defects predominate.6-9 This change of view largely reflected improved experimental design over time—in early studies no distinction was made between the various types of glaucoma, nor was any allowance usually made for the age distribution of subjects in the sample. A summary of the early research can be found in Dranceet al.7

Although modern studies control for confounding factors such as increasing lens density and decreasing pupil size with age, some controversy still surrounds the nature of colour vision defects in primary open angle glaucoma (POAG). Comparing previous research evaluating chromatic discrimination in POAG is complicated by wide variations in methodology and experimental conditions.

In recent years, computer generated colour tests have provided the means for isolating the processing of chromatic signals from the accompanying achromatic cues.10-15 In these tests, typically, the subject is asked to report the presence of a coloured target such …