Tall fescue has a good summer regrowth but is not always considered as an alternative to timothy because of its possible lack of palatability in dairy rations. We evaluated the effect of (1) replacing timothy with tall fescue, offered as sole forage or in association with alfalfa; and (2) feeding tall fescue as silage (35% DM) or haylage (55% DM). Treatments consisted of modifying the forage portion of the diet (70:30 forage:concentrate) as follows: (1) 100% timothy silage; (2) 100% tall fescue silage; (3) 60:40 timothy:alfalfa silages; (4) 60:40 tall fescue:alfalfa silages; (5) 100% tall fescue haylage. Fifteen Holstein cows were randomly assigned to treatments in a triple 5 × 5 Latin square design. Treatment periods lasted 21 d with the last 3 d used for data and sample collection. Preplanned contrasts tested were timothy vs. tall fescue, sole grass species vs. grass:alfalfa mixtures, interaction between grass species and grass:alfalfa mixture, and tall fescue silage vs. tall fescue haylage. Grass species did not affect DMI (23.5 kg/d; P = 0.36), milk yield (27.5 kg/d; P = 0.62), and milk fat content (4.21%; P = 0.55). Feeding grass:alfalfa mixtures resulted in higher DMI (24.3 vs. 22.6 kg/d; P < 0.01) and milk yield (28.3 vs. 26.7 kg/d; P < 0.01), but in lower milk fat content (4.15 vs. 4.27%; P < 0.01) than feeding sole grass species. Milk protein content was not affected by grass species (3.48%; P = 0,52) when offered in association with alfalfa, but was higher with timothy than tall fescue when offered as sole grass species (3.53 vs. 3.40%; P < 0.01). Dry matter intake was higher (22.6 vs. 20.8 kg/d; P < 0.01), milk protein content tended to be lower (3.40 vs. 3.44%; P = 0.06), but milk yield (27.0 kg/d; P = 0.58) and milk fat content (4.25%; P = 0.53) were similar when tall fescue silage was fed as compared with tall fescue haylage. Results show that tall fescue is a good alternative to timothy for lactating dairy cows, and confirm the beneficial effect of feeding grass species in association with alfalfa. Using tall fescue haylage instead of silage reduced DMI, but did not affect milk yield.