Stereoselectivity has been known to play a role in drug action for 100 years or more. Nevertheless, chiral drugs have been developed and used as racemates, neglecting the fact that they comprise mixtures of two or more compounds which may have quite different pharmacological properties. A very limited access to pure enantiomers in the past has been responsible for this unsatisfactory state of affairs. During the last 20 years, significant achievements have made it possible to perform stereoselective synthesis and analysis. Today, novel chiral drugs are as a rule developed as single enantiomers. Yet, studies of old racaemic drugs are still designed, performed and published without mention of the fact that two or more compounds are involved. In recent years, a number of old racaemic drugs have been re-evaluated and re-introduced into the clinical area as the pure, active enantiomer (the eutomer). While in principle correct, the clinical benefit of this shift from a well established racaemate to a pure enantiomer often seems to be limited and sometimes exaggerated. Racaemic drugs with a deleterious enantiomer that does not contribute to the therapeutic effect (the distomer), may have been sorted out in the safety evaluation process. However, in the future any pharmacological study of racaemic drugs must include the pure enantiomers. This will generate new, valuable information on stereoselectivity in drug action and interaction.The interaction between a drug and its receptor is a threedimensional event. If the drug molecule is chiral, thus forming optically active isomers, this interaction is usually stereoselective. The first definitive example of a difference in pharmacological action between optical isomers, or enantiomers, of a chiral molecule was offered by Arthur Cushny almost hundred years ago. He showed that the natural, levorotatory alkaloid hyoscyamine was twice as potent as its racaemic form, atropine, in antagonizing cholinergic stimuli. Moreover, Cushny was able to show that endogenous adrenaline, which is also levorotatory, is twice as potent as synthetic adrenaline which is racaemic. The dextrorotatory isomer of adrenaline was much less potent. These and other fundamental observations were discussed in a monograph (Cushny 1926). A few years later, Easson & Stedman (1933) postulated a three-point interaction between a drug and its receptor to explain stereoselectivity in drug action.Despite this early understanding of the problem, chiral drugs were developed and used as racemates throughout the twentieth century, often on the assumption that only one enantiomer (the eutomer) is pharmacologically active in relevant doses and that the counterpart (the distomer) is inactive as well as harmless. A major reason for this policy was