Decreasing the dietary cation-anion difference (DCAD) by using anion sources before calving reduces hypocalcemia in cows at calving. Reduced DCAD from CaCl2-fertilized timothy hay achieves similar results, but the effects of feeding low-DCAD forage as silage have not been determined. The objective of this study was to evaluate the effect of low-DCAD timothy silage on dry cows. Six nonlactating and nonpregnant Holstein cows were used in a replicated 3 x 3 Latin square. Treatments were 1) control diet (DCAD = 232 mEq/kg of dry matter, DM); 2) low-DCAD diet using a low-DCAD timothy silage (LDTS; DCAD = -21 mEq/kg of DM); and 3) low-DCAD diet using a fermentation by-product (LDBP; DCAD = -32 mEq/kg of DM). Differences between dietary treatments were considered statistically significant at P < or = 0.05 and tendencies were noted when 0.05 < P < 0.10. Compared with the control, feeding LDTS tended to decrease DM intake (10.6 vs. 12.5 kg/d) and decreased urinary pH (6.15 vs. 8.18) as well as apparent digestibility of DM (67 vs. 69%). Blood pH (7.37 vs. 7.42), HCO3- (25.3 vs. 27.5 mM), and base excess (0.4 vs. 3.1 mM) were decreased, and blood Cl- (29.6 vs. 29.1 mg/dL) was increased. Apparently absorbed Na and Cl were higher and apparently absorbed K, P, and digested ADF were lower for LDTS compared with the control. Both LDTS and LDBP resulted in similar DM intake. Urinary pH tended to be higher (6.15 vs. 5.98) and percentage of digested DM was lower (67 vs. 70%) with LDTS compared with LDBP. Blood ionized Ca (5.3 vs. 5.4 mg/dL) tended to be lower and blood Cl- (29.6 vs. 30.1 mg/dL) was lower, whereas blood pH (7.37 vs. 7.33), HCO3- (25.3 vs. 21.5 mM), and base excess (0.4 vs. -3.8 mM) were higher with LDTS compared with LDBP. Apparent absorption of Na, Cl, S, and P, as well as apparent digestion of acid detergent fiber, neutral detergent fiber, and N were lower, and K, Cl, S, P, Mg, and N were less retained with LDTS compared with LDBP. Results confirm that low-DCAD timothy silage can be used to produce a compensated metabolic acidosis by decreasing the DCAD of rations served to nonlactating dairy cows.