Two experiments, one with 18 and one with 21 cows, were made to compare the feeding value for milk production of concentrates in which the nitrogen normally supplied by oil cake was completely replaced by either biuret or urea. In expt 1 the urea and biuret contributed 52% of the total nitrogen in the concentrate and in expt 2 they contributed 43%. Hay was fed as the sole roughage. The estimated energy intake in expt 1 was 102–104% of the Woodman standard and in expt 2 it was 108–109%.

In both experiments milk production was about 10% less on the urea and biuret treatments than on the control treatment. There was no significant difference in milk yield between the biuret and the urea treatments. The percentages of fat and protein in the milk from the urea and biuret treatments were higher than from the control treatment, but the weights of these milk constituents were slightly lower than from the control treatment. Liveweight changes and water consumption were not affected by the experimental treatments. In expt 2 an intake of nitrogen about 20% lower than is often recommended in the United Kingdom did not appear to depress milk yields. Three cows that had received biuret concentrates in expt 2 continued to do so during a nitrogen balance trial lasting 3 weeks. Each cow was slightly in negative balance, the average value being −4gN per cow per day.