To test the hypothesis that propionate can reduce hepatic capacity to detoxify ammonia, effects of the inclusion of propionate in intraruminal infusions of urea on the concentrations of ammonia, other metabolites and insulin in peripheral blood were investigated in two experiments with non-lactating dairy cows. Both experiments were of a 4 × 4 Latin square design with four animals, four treatments and four experimental periods of 7 d; feed was given in two equal meals each day, all intraruminal infusions were given for 1 h at the time of the morning feed, and propionic acid was partly neutralized with NaOH. In Expt 1, the treatments were a basal diet of pelleted lucerne and chopped hay alone or with the following infusions (g/d): urea 80, propionic acid 350, urea 80 plus propionic acid 350. The inclusion of propionate in the urea infusion markedly increased (P < 0·001) the concentration of ammonia in plasma compared with infusion of urea alone. Moreover, the inclusion of urea with the propionate infusion abolished (P < 0·01) the increase in blood insulin level seen with the infusion of propionate alone. In Expt 2, less severe treatments were imposed, the aim being to reproduce metabolic loads of propionate and ammonia that might be expected from a diet of high-protein grass silage rich in lactic acid. The treatments were a basal diet of grass silage alone or with the following infusions (g/d): NaCl 145, NaCl 145 plus urea 50, propionic acid 200, urea 50 plus propionic acid 200. Effects were less pronounced than in Expt 1 but, in the period immediately after infusion, similar effects were seen. It is concluded that propionate–ammonia interactions may have potentially important effects on milk production especially for diets with high proportions of grass silage containing high levels of protein and lactic acid.