Plasma cortisol levels measured one hour after injection of ACTH reflects the hypothalamic-pituitary-adrenocortical (HPA) axis activity. The test was performed on piglets at 6 weeks of age. From a base population of Large White pigs, two divergent lines were selected for 3 generations, for high and low cortisol levels respectively. At the 3rd generation of selection, the divergence between the two lines was about 5 genetic standard deviations. At 158 days of age, thirty-two entire males were slaughtered with or without mixing before slaughtering. A blood sample was collected at exsanguination. Urine was collected at carcass evisceration directly from the bladder. A fat sample was taken at the neck level 24 hours post mortem. Skin lesions were counted on the fore, middle and rear parts of the carcass. At slaughter, the high cortisol line showed a significantly higher level of plasma cortisol (81.7 vs 31.8 ng/ml, P<0.001) and urinary deconjugated cortisol (106.3 vs 72.3 ng/ml, P=0.009), cortisone and dopamine, as well as a higher level of estradiol in plasma (3.85 vs 3.39 on a logscale, P=0.038) and androstenone in fat (0.189 vs -0.366 on a logscale, P=0.019), with a tendency for higher levels of testosterone in plasma and skatole in fat. There was no difference between lines for the number of skin lesions and urinary levels of noradrenaline and adrenaline. Stress had a significant effect on cortisol which increased in urine from mixed animals but not on fat androstenone and skatole. Mixing also increased lesions on the fore and middle parts of the body, as an index for an increased number of fights. This experiment confirmed the effect of mixing on cortisol levels and skin lesion scores. It also provided evidence that selection for cortisol level influences steroidogenesis in entire male pigs.