Objectives of the experiment were to determine the effects of supplementing 2 doses of either vitamin D₃ (VitD) or 25-hydroxyvitamin D₃ (25D) on peripartum mineral metabolism and lactation performance. One hundred and 33 parous and 44 nulliparous pregnant Holstein cows were enrolled in the experiment. Cows were blocked by parity and previous lactation milk yield or genetic merit and then assigned randomly to receive a daily dietary top-dressed supplement containing 1 mg VitD (1mgD), 1 mg 25D (1mg25D), 3 mg VitD (3mgD), or 3 mg 25D (3mg25D) from 250 d in gestation until parturition. The prepartum diet was approximately −130 mEq/kg in dietary cation anion difference. Data were analyzed by ANOVA with mixed models using the Mixed procedure of SAS. Cows receiving the 25D treatments had greater (P = 0.01) concentrations of P in serum prepartum and a tendency for greater (P = 0.07) concentrations of P postpartum. Cows receiving the 25D treatments had greater (P = 0.04) concentrations of total Ca in serum postpartum. Cows receiving the 3mg25D treatment produced more milk (P < 0.01) in the first 42 d in milk compared with other treatments (Table 1). Prepartum DM intake, postpartum disease incidence, colostrum production, and concentrations of Ca and Mg in colostrum and milk did not differ among treatments. Cows receiving 25D were in a better mineral status postpartum and performed better compared with cows receiving just VitD. Table 1 (Abstr. 340). Effect of dose and source of vitamin D on mineral status and performance

Item	1mgD	3mgD	1mg25D	3mg25D	SE
Prepartum
Ca, mM	2.34	2.32	2.34	2.35	0.02
Mg, mM	0.85	0.84	0.83	0.83	0.02
P, mM*	1.89	1.87	1.96	2.05	0.03
Postpartum
Ca, mM*	2.13	2.11	2.15	2.18	0.02
Mg, mM	0.88	0.85	0.87	0.84	0.02
P, mM	1.72	1.7	1.75	1.8	0.04
Colostrum yield, kg	4.3	3.9	5.4	4.5	0.55
Milk yield, kg/d¥	35.3	32.4	33.4	36.6	1.3

*Source (P < 0.05), §Level (P < 0.05), ¥Level × Source (P < 0.05).