Vitamin A deficiency (VAD) is a public health problem affecting millions in developing nations. Supplementation for lactating women, whose needs are high, involves large oral doses of the preformed vitamin. The safety and efficacy of these doses has been inadequately studied. Lactating women typically receive 210 micro mol of retinyl ester during early lactation, but 420 micro mol has also been administered. If larger doses of vitamin A are not significantly more effective in preventing VAD in mothers and infants, then smaller doses would be recommended. We therefore examined the vitamin A concentration of milk from lactating sows (n = 15) that were provided two different doses of vitamin A (i.e., 1050 or 2100 micro mol, n = 6/group) or corn oil (n = 3), corresponding to doses given women on the basis of body weight. Compared with controls, an overall significant treatment effect was found (P = 0.0019), but there was no difference in milk concentration between treatment groups. Theoretically, applying the mean milk vitamin A concentrations of the groups through 12 h and values to 48 h from 4 sows, we estimate that an infant of a supplemented mother could realize an increase of +0.08 or 0.16 micro mol/g liver from the low or high dose, respectively.