The results of previous studies of the effect of oral contraceptives on visual function have produced discrepancies and have lacked specific restraints on the control and experimental groups. We compared 14 diabetics with normal acuity who were taking oral contraceptives with equivalent diabetic and control groups and used the Pickford-Nicholson anomaloscope and the Farnsworth-Munsell 100-Hue test. Diabetics using oral contraceptives showed a definite trend to increases in tritanomaly and extended red-green matching ranges. When these losses were classified by degree of defect a significantly higher incidence of major defects was found among diabetics on "the pill". Partial correlations for colour variables, duration of diabetes and duration of contraceptive therapy revealed a compounding effect. While subjects using oral contraceptives demonstrated trends toward the extreme scores, these effects were found only in one out of every four users.