The bioavailability of codeine and extent of its transformation to morphine were stated in 12 smoking and 11 nonsmoking subjects after single doses 60 mg IM codeine and 60 mg codeine sulfate orally, given 1 wk apart. Codeine and morphine plasma concentrations over the 12-hr period after drug were determined by radioimmunoassay (RIA). No differences were found between smokers and nonsmokers with respect to maximum plasma concentration (Cmax) of codeine, time to attain this concentration (tmax), codeine plasma half-life (t1/2), or areas under plasma concentration-time curves (AUC) for codeine or morphine. There was a faster, but clinically unimportant, mean apparent plasma clearance in smokers (52.8 +/- 2.3 (SEM) ml/min/70 kg) than in nonsmokers (45.0 +/- 2.1 ml/min/70 kg) after intramuscular injection only. Mean oral codeine bioavailability in smokers (54.8 +/- 4.9%) and in nonsmokers (50.2 +/- 2.1%) did not offer. Plasma morphine AUC values were higher after oral doses than after intramuscular injections, suggesting a first-pass O-demethylation of codeine. For six of these subjects plasma morphine AUC values were very low after both routes of administration, suggesting less O-demethylation of codeine in these than in the remaining 17 subjects. The observation of higher morphine AUC values after oral codeine, coupled with clinical reports of greater analgesic potency with intramuscular codeine, does not support the hypothesis that the analgesic properties of this drug are mediated entirely by biotransformation to morphine.