A total of ten experimental diets with protein concentrations ranging from 154 to 400 g/kg and two lipid levels (46 and 85 g/kg) with identical energy densities were offered to 240 male Ross 308 broilers from 7 to 28 d post-hatch. Growth performance was monitored and nutrient utilisation (apparent metabolisable energy (AME), N-corrected AME (AMEn), AME daily intake, AME:gross energy ratios, N retention) was determined. The weight gain response of broiler chickens to dietary protein concentrations in diets containing high and low lipid levels was diverse, with the relevant quadratic regressions being significantly different (P<0·05). With low lipid levels, the predicted maximum weight gain of 1809 g/bird equated to 342 g/kg dietary protein, whereas, for high lipid levels the predicted maximum weight gain of 1694 g/bird equated to 281 g/kg dietary protein. AME was linearly correlated with dietary protein concentration but regressions in diets with different lipid content were not significantly different (P>0·05). AMEn was also linearly (P<0·0001) increased with dietary protein concentrations but regressions in diets with low and high lipid content were significantly different (P<0·03). Carcass protein content increased linearly with dietary protein content in diets containing high lipid concentrations (r 0·933, P<0·0001); by contrast, this relationship was quadratic (R
2=0·93, P<0·0001) in diets with low lipid levels. In conclusion, predictably, the effects of dietary protein concentrations on broiler performance were profound; however, the impact of dietary protein on performance in broiler chickens was modified by dietary lipid concentrations.