Perifused equine anterior pituitary cells were used to investigate the relationships between the secretion of ACTH and substances known to either stimulate (corticotrophin-releasing hormone (CRH), and arginine vasopressin (AVP)) or inhibit (cortisol) ACTH secretion. The experiments were designed to mimic the hormone milieu present in vivo in the horse, with cortisol (0 or 100 nmol/l) and CRH (0 or 0·02 nmol/l) perifused continuously, and pulses of AVP (10 nmol/l) applied for 5 min at 30-min intervals. In columns perifused with 0·02 nmol CRH/1 there was no significant overall effect of 100 nmol cortisol/l on the ACTH responses to pulses of AVP, although there was a significant interaction between AVP pulse number and cortisol showing that ACTH total area (pmol ACTH proportional to area under response curve) in response to AVP pulses 1 and 2 was significantly (PP=0·0075) and incremental area (P=0·049) at all AVP pulses compared with the responses in columns receiving 0 nmol cortisol/l. There was a fall off in ACTH responsiveness with time during the experiment which, in the presence of 0·02 nmol CRH/1, was significantly (Pin vivo, 'background' CRH could allow an increase in ACTH in response to AVP released by a new stress, despite the presence of elevated cortisol. Journal of Endocrinology (1996) 148, 475–483