Cortisol response to stress appears to differ between lactating and non- lactating animals. Lactating (14 d post partuin) and non-lactating sheep were fitted with probes so that drugs and hormones could be infused directly into the posterior pituitary and paraveritricular nucleus of the hypothalamus. The animals were also fitted with instruments to allow monitoring of heart rate, body temperature and blood cortisol levels. Their reactions to a source of acute stress (a barking dog) were then followed, with or without drug and hormone manipulation. Results in both lactating and non-lactating animals indicated shortcomings in the use of cortisol as a stress indicator. Infusing prolactin and oxytocin into either the posterior pituitary or the para ventricular nucleus of the hypothalamus suppressed cortisol respon- siveness to stress in both lactating and non-lactating animals (the latter to a greater extent). In the absence of drugs, lactating animals had a slightly higher basal level of cortisol and a lower cortisol response to stress than their non-lactating counterparts. Despite suppression of cortisol responses, with or without drugs, other indicators of stress still changed with the presence of a barking dog, suggesting the complexity of control involved in stress responses.