The experiment described here was designed to test the proposition that lambs, given access to two feeds, as a choice, one abundant in crude protein (CP), the other deficient, can both select a diet which supports their potential growth rate and which avoids excesses of protein intake.

Twenty four Suffolk x Greyface wether and females, at a mean liveweight of 20.2 kg (s.e. 0.6) were individually housed and their feed intake and liveweight recorded weekly.

The four pelleted feeds used were highly digestible and had estimated energy yields of 10.5 MJ ME/kg freshweight. The basal feed (A) was made from equal parts barley and sugar beet pulp plus a vitamin/mineral mix. It contained 91 g CP/kg FW. Feed D was made by substituting fishmeal for the barley in feed A such that it contained 383g CP/kg. Feed B contained one third D and two thirds A and contained 182g CP/kg. Féed C contained two thirds D and one third A, giving a CP content of 266g CP/kg. Thus each of the four feeds A, B, C and D could be described in terms of either the proportion of feed A or the crude protein which it contained.

Four lambs were offered feed A alone, and four lambs received feed B only. Twelve of the lambs were given ad 1ibitum access to both the basal feed A and one of the three other feeds B, C and D, for ten weeks. Four lambs were offered a choice between feed B and feed D for four weeks.