SUMMARY1. In unanaesthetized rabbits, haemorrhage was simulated by inflating a cuff placed round the inferior vena cava so that cardiac output fell at a constant rate of 8 % of its resting value per minute. The circulatory responses were measured after injections into the fourth ventricle of saline vehicle, selective opioid antagonists, selective opioid agonists, and agonist-antagonist mixtures. Three sets of experiments were done to determine if a specific subtype of opiate receptor within the central nervous system is responsible for the circulatory decompensation that occurs during simulated haemorrhage.2. In six rabbits the effects of ascending doses of the antagonists naloxone (,uZselective), Mr 2266 (K-and ,t-selective), ICI 174864 (6-selective) and nor-binaltorphimine (K-selective) were tested. In three rabbits the effects of the antagonist naloxone, the agonists HTyr-D-Ala-Gly-MePhe-NH(CHi)2OH (DAGO, ,u-selective), U 50488H (K-selective), and [D-Pen2, D-Pen5]-enkephalin (DPDPE, 6-selective), and combinations of these agonists with naloxone were tested. In four rabbits the doserelated effects of DAGO on respiratory, as well as circulatory, functions were examined.3. After injecting saline vehicle, the circulatory response to simulated haemorrhage had two phases. During the first phase, systemic vascular conductance fell, heart rate rose, and mean arterial pressure fell by only -10 mmHg. A second, decompensatory, phase began when cardiac output had fallen to -50 % of its resting level. At this point, there was an abrupt rise in systemic vascular conductance and a fall in mean arterial pressure to < 40 mmHg.4. The lower range of doses of naloxone (3-30 nmol), Mr 2266 (10-100 nmol), ICI 174864 (10-30 nmol), and all doses of nor-binaltorphimine (1-100 nmol), were without effect on the circulatory response to simulated haemorrhage. Higher doses of naloxone (30-100 nmol), Mr 2266 (100-300 nmol) and ICI 174864 (30-100 nmol) abolished the decompensatory phase. The relative order of antagonist potency was ICI 174864 > naloxone > Mr 2266 > nor-binaltorphimine.5. In the second set of experiments, the critical dose of naloxone necessary to prevent circulatory decompensation during simulated haemorrhage was 30-150 1.S 74: 16 R. G. EVANS, J. LUDBROOK AND A. F. VANLEEUWEN nmol. The d-agonist DPDPE (50 nmol) did not affect the haemodynamic response to simulated haemorrhage, but it did block the effect of naloxone on the response. The u-agonist DAGO (30 nmol) and the K-agonist U 50488H (300 nmol) prevented circulatory decompensation during simulated haemorrhage, whether given alone or in combination with the critical dose of naloxone, but DAGO (30 mmol) also raised arterial pressure, lowered systemic vascular conductance and slowed respiration.6. In the third set of experiments we found that lower doses of DAGO (100-300 pmol) would also prevent circulatory decompensation during simulated haemorrhage. These doses did not affect resting arterial pressure or systemic vascular conductance, and had insignificant effects on res...