AbstractIn two changeover design experiments, fifteen early‐ and sixteen late‐lactation cows were used to investigate the effects of offering food beet withad libitumgrass silage and concentrates with different CP content on milk yield and quality. In Experiment 1 (early lactation) cows were offered no fodder beet (0) or 4 kg DM d−1(4) in conjunction with one of three concentrates containing 159, 191 or 244g CP kg−1DM (L.M.H.). Treatments were therefore 1L/0, 1L/4, 1M/0, 1M/4, 1H/0 and 1H/4. In Experiment 2 (late lactation) cows were offered the same level of fodder beet in conjunction with two concentrates containing 129 and 229 (L,H) g CP kg−1DM. Treatments were therefore 2L/0, 2L/4, 2H/0 and 2H/4.In both experiments feeding fodder beet reduced silage DM intakes (P<0·001) and increased total DM intake (P<0·05 toP<0·001). The substitution rate (r) ranged from 0·46 to 0·59kg of silage DM (kg−1fodder beet DM).In Experiment 1, fodder beet tended to increase milk yield, composition and yield of constituents, but the effect was statistically significant for milk protein content only (P<0·01). In Experiment 2, milk yields for 2L/0, 2L/4, 2H/0 and 2H/4 were 11·3, 12·1, 11·7 and 12·5 kg d−1respectively (s.e.d. 0·43, non‐significant), fat contents were 44·4, 47·3, 44·3 and 46·8g fat kg−1respectively (s.e.d. 0·73,P<0·001), protein contents were 34·3, 35·6, 35·3 and 36·2 g protein kg−1respectively (s.e.d. 0·28,P<0·001), fat yields were 494, 574, 512 and 579 g fat d−1respectively (s.e.d. 20,P<0·001) and protein yields were 385, 426, 407 and 442 g protein d−1(s.e.d. 13,P<0·01) respectively.Increasing CP in the concentrate significantly increased milk yield in Experiment 1 (23·9, 22·5, 23·5, 23·8, 26·2, 26·5kg d−1for 1L/0, 1L/4, 1M/0, 1M/4, 1H/0 and 1H/4 respectively,P<0·05). Higher CP in concentrate also resulted in significantly increased milk protein yield in early‐lactation (P<0·001) and milk protein content in late‐lactation (P<0·01) cows. There was a significant interaction between fodder beet and concentr