1. The ability of a selective CCKA receptor antagonist PD 140548 and a selective CCKB receptor antagonist CI-988 (formerly PD 134308) to modulate the various in vivo properties of morphine was investigated in the rat. 2. PD 140548 dose-dependently (0.001-1.0 mg kg-1, i.p.) antagonised the development of conditioned place preference to morphine (2.0 mg kg-1, s.c.). In contrast, CI-988 (0.01-1.0 mg kg-1, i.p.) did not affect this morphine-induced behaviour. Neither of the CCK receptor antagonists blocked or generalised to the morphine (3.0 mg kg-1, i.p.) discriminative stimulus. 3. CI-988 (0.001-10.0 mg kg-1, s.c.) at doses of 0.05 and 0.1 mg kg-1 (s.c.), potentiated the antinociceptive action of a threshold dose of morphine (5.0 mg kg-1, i.p.) in a radiant heat model of acute nociception, the rat tail flick test. Furthermore, at 0.01 mg kg-1 it potentiated the antinociceptive action of morphine (3.0 mg kg-1) during the acute phase of the rat paw formalin test. And at doses of 0.01 and 0.1 mg kg-1 it also potentiated the antinociceptive action of morphine (1.0 mg kg-1) during the tonic phase of the formalin test. However, in both models, higher doses of CI-988 were ineffective. In contrast, PD 140548 (0.001-10 mg kg-1, s.c.) was only active at a dose of 1.0 mg kg-1 (s.c.) and only in the tonic phase of the formalin test. Neither CI-988 nor PD 140548 possessed any intrinsic antinociceptive action in either of the tests. Chronic treatment with CI-988 (0.01 mg kg-1, s.c.) prevented the development of tolerance to morphine antinociception (4 mg kg-1, s.c.) following a 6 day period of twice daily injections of morphine escalating from 1 to 16 mg kg-1 (i.p.). 4. Morphine dose-dependently (1-10 mg kg-1, s.c.) reduced the distance travelled by a charcoal meal in the rat intestine. Neither PD 140548 (0.01-1.0 mg kg-1, i.p.) nor CI-988 (0.01-1.0 mg kg-1, i.p.) potentiated or suppressed this inhibitory action of morphine. 5. In conclusion, the results of the present study indicate that CCKA and CCKB receptors modulate different properties of morphine. Thus, whilst a selective CCKA receptor antagonist blocked the rewarding properties of morphine, a selective CCKB receptor antagonist potentiated the antinociceptive action. However, neither compound displayed a potential for modulating the influence of morphine on gastro-intestinal motility. It is suggested that these findings may have important implications for development of CCK receptor antagonists as analgesic adjuncts to the therapeutic use of morphine.