1. 5-Hydroxytryptamine-1B (5-HT 1B, formerly designated 5-HT 1D beta) and 5-hydroxy-tryptamine-1D (5-HT 1D, formerly designated 5-HT 1D alpha) receptors are distinct molecular entities that mediate serotonergic neurotransmission. Both are G-protein-coupled receptors without introns in their coding region, negatively coupled to adenylate cyclase; their precise function in human beings remains to be defined. In brain, they are highly enriched in the globus pallidus and the substantia nigra. 2. Presynaptic 5-HT 1B/D receptors take part in the control of the release not only of 5-HT itself, but also of other neurotransmitters-for example, acetylcholine, glutamate, dopamine, noradrenaline and gamma-aminobutyric acid. Selective blockade of central 5-HT 1B/D autoreceptors should facilitate 5-HT neurotransmission and may offer a novel approach to antidepressant therapy. Other 5-HT 1B/D receptors are located postsynaptically; those receptors may be supersensitive in the pathophysiology of obsessive-compulsive disorder and may be a potential target for its treatment. 3. Few if any ligands show selectivity for 5-HT 1B or 5-HT 1D receptors or both. Most pharmacological studies have been performed with nonselective antagonists-for example, metergoline, I-naphthylpiperazine, methiothepin, ketanserin and ritanserin. Recently, a novel series of benzanilides have been reported as the first examples of selective 5-HT 1B/D receptor antagonists. GR 127935, a representative compound of this series, displays mixed agonist-antagonist properties both in vitro and in vivo. It induces upon systemic administration in the guinea pig either an opposite (decrease) effect or a small increase (65%, 5 mg/kg) in the concentration of cortical extracellular 5-HT compared with fluoxetine (218%, 10 mg/kg). The importance of blockade of 5-HT 1B/D receptors in the raphé region, their possible interaction with 5-HT 1A receptors, and consequent inhibition of 5-HT release in terminal 5-HT 1B/D receptor-containing regions are discussed. 4. To find out whether the available so-called 5-HT 1B/D receptor antagonists are indeed antagonists and not partial agonists, efficacy was measured at recombinant human 5-HT 1B and 5-HT 1D receptor sites by using a [35S]-GTP gamma S binding assay to membrane preparations of stably transfected rat C6-glial cell lines. Metergoline and the selective 5-HT 1B/D receptor ligands GR 127935 as well as GR 125743 showed significant intrinsic activity (43% to 69%) at the 5-HT 1D receptor subtype, whereas the nonselective ligand 1-naphthylpiperazine yielded less (15% to 19%) intrinsic activity at both receptor subtypes. In contrast, the nonselective ligands methiothepin, ketanserin and ritanserin are inverse agonists because they displayed negative efficacy (-14% to -28%). Differential blockade of 5-HT 1B/D receptors by neutral antagonists and inverse agonists is discussed in relation to the 5-HT tone on 5-HT 1B/D receptors. 5. It can concluded that 5-HT 1B/D receptor ligands modulate 5-HT neurotransmission through a terminal 5-HT 1B/D receptor. Future work should be directed toward the identification of selective 5-HT 1B and 5-HT 1D receptor ligands that display either neutral antagonist or inverse agonist properties to evaluate the therapeutic potential of 5-HT 1B/D receptor blockade.