Tobacco use has one of the highest rates of addiction and relapse of any abused drug. Paradoxically, however, in animal models of reinforcement nicotine appears weak compared to other abused drugs. We report here that acetaldehyde, a major component of tobacco smoke, enhances nicotine self-administration. Juvenile and adult male rats were implanted with intravenous catheters and tested for self-administration 4 days later at postnatal day 27 or 90, respectively. Animals were tested, without prior response training, in five daily 3-h sessions where each nose-poke delivered an intravenous injection followed by a 60-s timeout. Animals (11-13/group) were offered one of the following solutions: nicotine (30 microg/kg/injection), acetaldehyde (16 microg/kg/inj), nicotine (30 microg/kg/inj)+acetaldehyde (16 microg/kg/inj), or saline. The youngest animals responded significantly more for nic+acet than for saline or for either drug alone. Responding at the reinforced hole was significantly higher than at the nonreinforced hole or at the reinforced hole during noncontingent injections of nic+acet. Tests with receptor antagonists indicated that these drug effects are mediated by central, but not peripheral, nicotinic receptors. There was an age-related decline in self-administration of nic+acet, but not for cocaine. Taken together, these results indicate that acetaldehyde, at the low concentrations found in tobacco smoke, interacts with nicotine to increase responding in a stringent self-administration acquisition test where nicotine alone is only weakly reinforcing, and that adolescent animals are more sensitive to these actions than adults. Animal models of tobacco addiction could be improved by combining acetaldehyde, and possibly other smoke components, with nicotine to more accurately reflect the pharmacological profile of tobacco smoke.