The effects of prepartum dietary concentrations of K, Na, and Ca on the incidence of periparturient hypocalcemia or milk fever was determined for older (> or = 4th lactation) Jersey cows. Cows were fed one of six diets differing in K and Ca contents. In addition, the effect of dietary Na (tested only at the high concentration of dietary Ca, and low concentration of dietary K) was examined. Treatments were arranged in an incomplete 2 x 4 factorial design; dietary Ca (0.5 or 1.5%) and dietary strong cations (1.1, 2.1, and 3.1% K or 1.3% Na) were the main effects. Dietary Ca did not significantly affect the incidence of milk fever of the degree of hypocalcemia experienced by the cows. Milk fever occurred in 2 of 20 cows that were fed the prepartum diet containing 1.1% K and 0.12% Na. Increasing dietary K to 2.1 or 3.1% increased the incidence of milk fever to 10 of 20 cows and 11 of 23 cows, respectively. Increasing dietary Na to 1.3% in the diet containing 1.5% Ca induced milk fever in 5 of 8 cows. Addition of strong cations to the prepartum diet increased blood and urine pH and reduced plasma hydroxyproline concentrations, suggesting that bone resorption of Ca is inhibited in cows fed high K or high Na diets as a result of metabolic alkalosis. These data demonstrated that dietary Ca concentration is not a major risk factor for milk fever and that dietary strong cations, especially K, induce metabolic alkalosis in the prepartum dairy cow, which reduces the ability of the cow to maintain Ca homeostasis.