1. Antitussive, antinociceptive and respiratory depressant effects of codeine, morphine and H.Tyr.D-Arg.Gly.Phe(4-NO2) Pro.NH2 (compound BW443C) were investigated in unanaesthetized guinea-pigs. Antagonism of the antitussive and antinociceptive effects was investigated by the use of nalorphine and N-methylnalorphine. Naloxone was used to antagonize respiratory depression. 2. Antitussive ED50s (with 95% confidence limits) for inhibition of cough induced by citric acid vapour were for codeine, morphine and BW443C respectively, 9.1(5.8-15), 1.3(0.7-2.4) and 1.2(0.6-2.6) mg kg-1 s.c. and 8.7(4.2-12), 1.6(1.2-1.9) and 0.67(0.002-3.3) mg kg-1, i.v. The antitussive effects of subcutaneous codeine (25 mg kg-1) morphine (8.1 mg kg-1) and BW443C (2.5 mg kg-1) were significantly antagonized by subcutaneous nalorphine (3.0 mg kg-1) and N-methylnalorphine (3.0 mg kg-1). 3. In the multiple toe-pinch test, the antinociceptive ED50s (with 95% confidence limits) of codeine and morphine were 18(16-22) and 2.3(0.4-4.3) mg kg-1, s.c., respectively. Compound BW443C was ineffective in doses of 2.5 and 10 mg kg-1 s.c., a result consistent with its lacking penetration into the CNS. Subcutaneous nalorphine (3.0 mg kg-1) antagonized the antinociceptive action of codeine (25 mg kg-1) and morphine (8.1 mg kg-1). In contrast, N-methylnalorphine (3.0 mg kg-1) had no significant effect on the antinociceptive action of codeine and morphine, suggesting lack of penetration of the CNS by N-methylnalorphine. 4. At doses near to the i.v. ED50 values for the antitussive activity, morphine (1.5mg kg- ', i.v.) and codeine (10mg kg-', i.v.) caused small but significant depressions of ventilation (7.0 +/- 2.3% and 16.5 +/- 8.4% respectively). Higher doses of morphine (10, 30 and 60mg kg- ', i.v.) caused further doserelated depression of ventilation (9.6 +/- 5.3%, 22.4 +/- 6.2% and 36.2 +/- 9.6% respectively) whereas codeine (30 and 60mg kg-' i.v.) caused stimulation of ventilation which was marked (191.3 +/- 43.9%) at 60 mg kg-'. 5. Compound BW443C in doses of 1 or 10mgkg-',i.v. (approximately equal to, and 10 times the EDo for antitussive activity) did not cause significant depression of ventilation. Only at higher doses of 30 and 60mg kg-', i.v. was there a significant decrease in minute volume (13.1 +/- 6.8% and 15.9 +/- 1.89% respectively). The depression of ventilation caused by either BW443C (60mg kg-', i.v.) or morphine (60mg kg-', i.v.) was prevented by pretreatment with naloxone (3mg kg-', i.v.) administered 15 min before morphine or BW443C. 6. These results in the guinea-pig support the hypothesis that the antitussive action of the opiates codeine and morphine and the opioid pentapeptide BW443C do not require penetration of these drugs into the CNS.