The objective of the present experiment was to define the form of the relationship between varying levels of energy intake at constant, high protein intake and the performance of young pigs. By doing so it was expected that we could distinguish between four models that predict the pig's responses to its diet as rates of protein and lipid retention. Forty young pigs were assigned at 12 kg live weight either to an initial slaughter group (n 8) or to one of four allowances of starch intake at a constant intake of a high-protein feed (with 345 g crude protein (nitrogen x 6.25)/kg food). Half the pigs were killed after 4 weeks and half after 8 weeks on the treatments; at each slaughter point on each treatment half the pigs were entire males and half were females. Increasing the intake of starch (energy) resulted in significant increases in the rates of live weight, empty-body, protein and lipid gains of pigs slaughtered at both stages. There was no minimum positive lipid:protein ratio in the gain of the pigs. Male pigs deposited more protein and less lipid than females but this effect of sex on protein and lipid retention was seen only on the two highest allowances of starch intake. The calculated efficiency of protein utilization increased with increasing starch intake up to a maximum of 0.81, when probably the energy:protein in the diet became non-limiting. The results led to the rejection of two of the models that predict the rates of protein and lipid retention as responses to protein and energy intake, but the two remaining models could not be rejected, at least qualitatively.