Our objective was to determine the effects of varying dietary cation-anion differences (DCAD: meq[(Na + K) - (Cl + S)]/100 g of dry matter) in prepartum diets on Ca, energy, and endocrine status prepartum and postpartum. Holstein cows (n = 21) and heifers (n = 34) were fed diets with varying amounts of CaCl2, CaSO4, and MgSO4 to achieve a DCAD of +15 (control), 0, or -15 meq/100 g of dry matter for the last 24 d before expected calving. Dietary Ca concentration was increased (by CaCO3 supplementation) with decreasing DCAD. Plasma ionized Ca concentrations prepartum and at calving in both cows and heifers increased with reduced DCAD in the diet. At calving, plasma ionized Ca concentration was 3.67, 3.85, and 4.35 for cows and 4.44, 4.57, and 4.62 mg/dl for heifers fed diets containing +15, 0, and -15 DCAD, respectively. All heifers had normal concentrations of plasma ionized Ca (>4 mg/dl) at calving. Also at calving, plasma concentrations ofparathyroid hormone and calcitriol were less in cows and heifers fed diets containing reduced DCAD, but the plasma concentration of hydroxyproline was not affected by diet. Prepartum dry matter intake, energy balance, and body weight gains were lower and concentration of liver triglyceride was higher for heifers but not cows fed the -15 DCAD diet. Also, nonesterified fatty acids the last week prepartum were positively correlated with liver triglyceride for heifers but not cows. Feeding of anionic salts plus CaCO3 to reduce DCAD to -15 and increase Ca in prepartum diets prevents hypocalcemia at calving in cows, but decreases prepartum dry matter intake and increases the concentration of liver triglyceride in heifers. That heifers maintained calcium homeostasis at calving regardless of diet but ate less when fed the -15 DCAD diet suggests that they should not be fed anionic salts before calving.