Improve vitamin D status in lactating women and their recipient infants, and measure breast milk calcium concentration [Ca] as a function of vitamin D regimen. Fully breastfeeding mothers were randomized at 1 month postpartum to 2000 (n = 12) or 4000 (n = 13) IU/day vitamin D for 3 months to achieve optimal vitamin D status [serum 25(OH)D > or =32 ng/mL (80 nmol/L)]. Breast milk [Ca], maternal and infant serum 25(OH)D and serum Ca, and maternal urinary Ca/Cr ratio were measured monthly. Mothers were similar between groups for age, race, gestation, and birth weight. 25(OH)D increased from 1 to 4 months in both groups (mean +/- SD): +11.5 +/- 2.3 ng/mL for group 2000 (p = 0.002) and +14.4 +/- 3.0 ng/mL for group 4000 (p = 0.0008). The 4000 IU/day regimen was more effective in raising both maternal and infant serum levels and breast milk antirachitic activity than the 2000 IU/day regimen. Breast milk [Ca] fell with continued lactation through 4 months in the 2000 and 4000 IU groups. Decline in breast milk [Ca] was not associated with vitamin D dose (p = 0.73) or maternal 25(OH)D (p = 0.94). No mother or infant experienced vitamin D-related adverse events, and all laboratory parameters remained in the normal range. High-dose vitamin D was effective in increasing 25(OH)D levels in fully breastfeeding mothers to optimal levels without evidence of toxicity. Breast milk [Ca] and its decline in both groups during 1 to 4 months were independent of maternal vitamin D status and regimen. Both the mother and her infant attained improved vitamin D status and maintained normal [Ca].