Two experiments were conducted with eight rams in a latin-square design of four treatments with different lucerne hay (LH): maize grain (MG) ratios: 100:0, 80:20, 50:50 and 20:80. In Expt 1 the amount of food offered was restricted to 1 kg/day. In Expt 2 the rams were offered 1·5 kg/day, which was about the maximum amount the animals in the 100% LH treatment were able to consume. Animals were kept on each treatment for 4 weeks; during the final 8 days the faeces were oollected and weighed for each animal and analysed for dry matter, organic matter (OM), crude fibre and δC
The values of δC in LH, MG and different faeces samples were used for direct determination of LH and MG digestibilities of mixed diets in the different treatments.
The increase of MG in the ration and the decrease of LH improved the digestibility of total OM. The effect of MG was more pronounced at the low than at the high level of consumption. At the low level of intake, an increase in the amount of MG in the ration did not affect the digestibility of LH, and MG digestibility was close to the extrapolated value. At a food intake level 1·5 times that of Expt 1, the addition of grain to more than 50 % of the ration caused a marked depression in digestibility of LH.
Results of the δC method were discussed in comparison with calculated values of LH and MG digestibilities and with whole ration crude-fibre digestibility.