Two experiments were conducted to evaluate the response of sows to high levels of dietary riboflavin. Responses evaluated were erythrocyte glutathione reductase activity coefficient (EGRAC, a biochemical indicator of riboflavin status) and reproductive performance. In Exp. 1, eight sows were fed diets either slightly deficient or superadequate in riboflavin throughout a reproductive cycle, beginning in midgestation. The EGRAC values confirmed the slight deficiency (P < .01) but did not clearly show a time when the sow is most likely to be deficient in riboflavin. In Exp. 2, 118 sows at two farms were fed diets providing 10 (control, 1.4 times the NRC  requirement estimate), 60, 110, or 160 mg of riboflavin/d for 21 d beginning at breeding, and the control level through the remainder of the reproductive cycle. The percentage of sows farrowing tended to be lower (P < .10) for the control sows than for sows fed the other treatments (66.7, 85.7, 93.3, and 86.7% for 10, 60, 110, and 160 mg/d, respectively). This response is similar to other results reported previously. No other measure of reproductive performance, nor EGRAC measured at any of five stages of the reproductive cycle, was affected by treatment. These results suggest that massive riboflavin supplementation of the sows' diet during early pregnancy may increase the percentage of farrowing but does not increase litter size.