1. Two calorimetric experiments were conducted to study the utilization of energy in sheep given diets of grass silage or grass silage and barley. Three silages were investigated. One was made from first-harvest grass in the spring (S) and the others from regrowth cut either early or late in the autumn (E and L respectively). All were of perennial ryegrass (Lolium perenne) and preserved with formic acid. Each silage was given at two levels of feeding, the lower providing approximately a maintenance energy intake. The S and L silages were also given supplemented with barley.2. The digestibilities of organic matter, cellulose and energy in the silages were high. Measured at maintenance, digestible energy (de) contents (MJ/kg dry matter (dm)) were 11.83, 14.67 and 12.90 for S, E and L respectively. Thedecontents of the S and E silages were depressed at the higher level of feeding but the effect was offset by changes in the energy losses as methane and urine. Metabolizable energy (me) contents (MJ/kgdm) for the three silages, S, E and L were respectively 9.88, 12.54 and 10.73 at the low level of feeding and 9.91, 11.99 and 11.08 at the high level of feeding. The meanmecontent of barley calculated by difference was 13.76 MJ/kgdm.3. The mean efficiencies of utilization ofmefor maintenance (km) for the S, E and L silages were 0.69, 0.71 and 0.68 respectively. Corresponding values for fattening (kf) were 0.21, 0.57 and 0.59. Excepting thekffor the S silage which was low, observed efficiencies were in broad agreement with those predicted by the equations of the Agricultural Research Council (1965). Similar agreement was obtained with all diets consisting of silage and barley.