The results of previous studies suggested that the sleep-time differential between long-sleep (LS) and short-sleep (SS) mice decreases for a series of sedative-hypnotic drugs as lipid solubility increases. Using the LS x SS recombinant inbred (RI) strains, we have tested whether this relationship arises because of common genetic influences on the sleep-time responses or was simply a fortuitous difference between LS and SS mice. Mice were sleep-time tested after intraperitoneal injections of a variety of sedative-hypnotic drugs that vary in lipid solubility including alcohols, barbiturates, chloral hydrate, and urethane. Sleep-time values from each of these drugs were compared with ethanol-induced sleep times in the LS x SS RI strains. Significant genetic correlations were observed between ethanol-induced sleep time and those responses elicited by butanol, propanol, and chloral hydrate, but not by pentobarbital or secobarbital. When the partition coefficient (log P) values were compared with the genetic correlations, a significant relationship (r = -0.85) was observed. These data suggest that common genes mediate sedative-hypnotic reactions to ethanol and some drugs resembling it in log P value, and that anesthetic agents with low lipid solubility may be working through different mechanisms than drugs with greater lipid solubilities.