Reducing the dietary cation-anion difference (DCAD) reduces urine pH and, therefore, has potential to lower NH 3 emissions from manure. We determined the effects of decreased DCAD on dry matter intake, production, nutrient digestibility, manure characteristics, and NH 3 emissions from manure. An in vitro incubation study was conducted to evaluate the degree of reduced urine pH on manure pH and NH 3 emissions from manure. In this study, urine pH was directly decreased from 8.5 to 7.5, 6.5, and 5.5 by adding sulfuric acid, which resulted in decreases in manure pH when manure was reconstituted with the fecal-to-urine ratio of 2:1 (as-is basis). The manures from urine at pH 7.5, 6.5, and 5.5 decreased NH 3 emissions linearly by 19, 33, and 36%, respectively, compared with the manure from unacidified urine. An animal study was conducted with 27 mid-lactation Holstein cows in a randomized complete block design. Cows were blocked by parity and days in milk and assigned to 1 of 3 different DCAD diets:(1) HDCAD, a diet with DCAD of 193 mEq/kg of dry matter (DM); (2) MDCAD, a diet with 101 mEq/kg of DM; and (3) LDCAD, a diet with 1 mEq/kg of DM. A commercial anionic product (predominantly ammonium chloride) partly replaced urea, soybean meal, soyhulls, and corn grain in MDCAD and LDCAD to lower DCAD. The experiment lasted 7 wk (1-wk covariate followed by 6-wk data collection). Spot urine and fecal samples were collected for manure incubation. Data were analyzed using the MIXED procedure of SAS in a randomized block design. Dry matter intake and milk yield were not altered by treatments. No difference in milk fat content was observed among treatments, but fat yield tended to decrease linearly (1.00 to 0.86 kg/d) as DCAD decreased, resulting in a tendency for decreasing energy-corrected milk yield (35.1 to 32.7 kg/d). Milk protein content increased (3.00 to 3.14%) as DCAD decreased, but milk protein yield was not affected. Total-tract digestibility of DM, organic matter, and neutral detergent fiber did not differ among treatments. Digestibility of crude protein tended to decrease as DCAD decreased. There was no difference in fecal and urine N excretion among treatments, but fecal N as proportion of N intake tended to increase as DCAD decreased. Urine pH decreased linearly from 8.42 for HDCAD to 8.11 and 6.41 for MDCAD and LDCAD, respectively, resulting in decreased manure pH (7.57, 7.40, and 6.96 for HDCAD, MDCAD, and LDCAD, respectively). The cumulative NH 3 emissions from manures over 6 d tended to decrease linearly as DCAD decreased (461 to 390 mg/kg of manure), but the decrease was only numerical when calculated on a cow basis (i.e., g/cow). In conclusion, lowering DCAD has potential to reduce NH 3 emission from manure of lactating cows. However, a tendency for decreased milk fat yield and energy-corrected milk yield suggests that DCAD of 1 mEq/kg of DM may be too low, and more studies are needed to examine relatively less reduced DCAD to determine production responses in addition to NH 3 emission from manure.