We determined the quantitative relationships between graded oral dosing with vitamin D3, 25(OH)D3, and 1,25(OH)2D3 for short treatment periods and changes in circulating levels of these substances. The subjects were 116 healthy men (mean age, 28 +/- 4 years, with usual milk consumption of < or = 0.47 l/day and mean serum 25(OH)D of 67 +/- 25 nmol/l). They were distributed among nine open-label treatment groups: vitamin D3 (25, 250 or 1250 micrograms/day for 8 weeks), 25(OH)D3 (10, 20 or 50 micrograms/day for 4 weeks) and 1,25(OH)2D3 (0.5, 1.0 or 1.0 microgram/day for 2 weeks). All treatment occurred between January 3 and April 3. We measured fasting serum, calcium, parathyroid hormone, vitamin D3, 25(OH)D and 1,25(OH)2D immediately before and after treatment. In the three groups treated with vitamin D3, mean values for circulating vitamin D3 increased by 13, 137 and 883 nmol/l and serum 25(OH)D increased by 29, 146 and 643 nmol/l for the three dosage groups, respectively. Treatment with 25(OH)D3 increased circulating 25(OH)D by 40, 76 and 206 nmol/l, respectively. Neither compound changed serum 1,25(OH)2D levels. However, treatment with 1,25(OH)2D3 increased circulating 1,25(OH)2D by 10, 46 and 60 pmol/l, respectively. Slopes calculated from these data allow the following estimates of mean treatment effects for typical dosage units in healthy 70-kg adults: an 8-week course of vitamin D3 at 10 micrograms/day (400 IU/day) would raise serum vitamin D by 9 nmol/l and serum 25(OH)D by 11 nmol/l; a 4-week course of 25(OH)D3 at 20 micrograms/day would raise serum 25(OH)D by 94 nmol/l; and a 2-week course of 1,25(OH)2D3 at 0.5 microgram/day would raise serum 1,25(OH)2D by 17 pmol/l.