Prospective studies of carbohydrate metabolism were performed by means of an oral glucose tolerance test in 29 women. The studies were performed before and after 3 months of treatment with an oral contraceptive that contained 0.035 mg of ethinyl estradiol and 0.4 mg of norethindrone. The 24 women with a "normal" control test had no significant alteration in any of their blood glucose and plasma insulin values, except that the fasting glucose level decreased with treatment. Three of these women (12.5%) had a slight deterioration of their glucose curves into a "borderline abnormal" grouping during treatment. There were five women with "borderline abnormal"control glucose curves, and in four of these (80%) the curves returned to "normal" during treatment, whereas one curve stayed "borderline abnormal." These data suggest that this low-estrogen oral contraceptive has minimal short-term adverse effects on the carbohydrate metabolism.