Sixteen first lactation dairy cows were used in a 2 x 2 x 2 x 2 factorial design experiment. Two silages, an extensively fermented inoculant-treated silage or a restricted fermented, mixed aliphatic carboxylic acid-treated silage were offered with two concentrate types (starch- or fibre-based) and supplementary lactic acid (LA, 0.4 kg/ day) or soya oil (SO, 0.41 kg/day). Concentrates containing 180 g crude protein per kg were offered at 6 kg/day. Silage dry-matter intake tended to he higher (P = 0.07) with restricted fermented silage compared with the inoculant silage. Milk fat or protein concentrations were not significantly affected (P > 0.05) by silage fermentation type. LA inclusion had no significant effect (P > 0.05) on animal performance. SO inclusion significantly increased (P < 0.05) milk yield (+ 0.8 kg/day), decreased milk fat concentration and yield (40.2, 32.9 g/kg and 779, 667 g/day for - and + SO; respectively, P < 0.001) and decreased milk protein concentration (- 0.8 g/kg, P < 0.01). A significant concentrate X soya-oil interaction (P < 0.01) suggested a greater reduction in milk fat concentration with the SO, starch compared with the SO, fibre combination. Modified acid-detergent fibre (P < 0.01) and neutral-detergent fibre (P < 0.001) apparent digestibility coefficients were significantly lower with starch- compared with fibre-based concentrates. Results suggest that milk production responses to changes in concentrate composition were not influenced by the extent of restriction of the silage fermentation process.