The increased risk of periparturient hypocalcaemia through offering high-K feeds in the precalving period has been reported previously. Two experiments (experiment 1 and 2) investigated the effect of KCl fertilizer on pasture mineral concentration, the dietary cation-anion difference (DCAD), and the subsequent effect of this on periparturient plasma and urine mineral status. Experiment 2 examined the effect of precalving Mg source (MgO, MgSO4, and MgCl2) and postcalving Ca supplementation on the concentration of Ca and Mg in plasma and urine. Unexpectedly, pasture DCAD increased (P = 0.06) from 434 to 535 mEq/kg DM in experiment 1 as pasture K concentration decreased from 4.2 to 3.5%, primarily because of a corresponding and greater decrease in pasture Cl concentration (from 1.9 to 1.3%). Plasma Ca or Mg concentrations were not affected by pasture K concentration. A linear decline (P < 0.10) in urine Mg suggested a decline in Mg absorption as pasture K increased. In experiment 2, pasture DCAD decreased (P < 0.05) linearly from 403 to 350 mEq/kg DM as pasture K concentration decreased from 3.8 to 3.3%. However, precalving urine pH was not affected by the declining DCAD. Postcalving plasma Ca concentration was affected by precalving Mg source with MgSO4 > MgCl2 > MgO. Differences in acid-base balance do not explain the difference between Mg salts. These results indicate that precalving dietary S and Cl concentration plays an important role in Ca homeostasis, in addition to its role in acid-base balance. Supplementation with Ca postcalving increased plasma Ca concentration for 2 d postcalving. Milk production was not affected.