Context: Guidelines in Finland recommend 10 μg of vitamin D3 daily for all infants. Recent observations suggest that this may be insufficient to maintain optimal serum 25-hydroxyvitamin D (S-25-OHD). Objective: The aim of the study was to evaluate effects of various vitamin D doses and determine a dose ensuring S-25-OHD of at least 80 nmol/liter in infants without signs of vitamin D excess. Design: We conducted a randomized double-blind intervention study. Cord blood was obtained at birth for S-25-OHD; 113 infants were randomized to receive vitamin D3 10, 30, or 40 μg/d from age 2 wk to 3 months. Setting: An investigator-initiated study was performed in a single maternity hospital in Helsinki, Finland. Main Outcome Measures: S-25-OHD, calcium homeostasis, and skeletal characteristics were evaluated with peripheral quantitative computed tomography at age 3 months. Results: Baseline S-25-OHD was similar in all three groups (median, 53 nmol/liter). At 3 months, the mean S-25-OHD values were 88, 124, and 153 nmol/liter, and the minimum values were 46, 57, and 86 nmol/liter in the groups receiving 10, 30, and 40 μg (ANOVA; P < 0.001). No hypercalcemia occurred; plasma calcium, serum PTH, and urine calcium excretion was similar between the groups. Peripheral quantitative computed tomography showed a trend toward larger tibial total bone and cortical bone area with higher vitamin D doses. Conclusion: Vitamin D3 supplementation with up to 40 μg/d from age 2 wk to 3 months was safe and caused no hypercalcemia or hypercalciuria. The 40-μg dose maintained S-25-OHD above 80 nmol/liter in all infants. More extensive and longer intervention studies are necessary to assess long-term effects.