1. The pharmacological properties and location of H3 receptors modulating acetylcholine release have been investigated in non-superfused slices and synaptosomes of rat entorhinal cortex preloaded with [3H]-choline. 2. (R)alpha-methylhistamine, an H3-receptor agonist, potently inhibited the K(+)-evoked tritium release from slices, an effect antagonized by thioperamide, an H3-receptor antagonist, with nanomolar potency. 3. The K(+)-evoked tritium release from synaptosomes remained unaltered in the presence of the potent and selective H3-receptor agonists, imetit and (R)alpha-methylhistamine, suggesting that H3 receptors modulating acetylcholine release are not presynaptically located on cholinergic nerve terminals. 4. Phenylbutanoylhistamine and phenylpropylhistamine, two H3-receptor antagonists of moderate potency, failed to antagonize the inhibitory effects of (R)alpha-methylhistamine observed in slices. Unexpectedly, both compounds when used alone, inhibited tritium release from slices and synaptosomes with micromolar potency and to the same extent (by approximately 50% when added at a final concentration of 200 microM). This inhibitory effect did not involve H1, H2 or H3 receptors and was not mediated by an unknown histamine receptor site, since histamine used at a high concentration neither reproduced nor antagonized the effect of phenylbutanoylhistamine. It remained unaltered in the presence of scopolamine and was neither mimicked nor antagonized by vasoactive intestinal peptide, previously shown to be colocalized with acetylcholine in some neurones. 5. It is concluded that acetylcholine release in rat entorhinal cortex is modulated by H3 receptors presumably not located on cholinergic axon terminals. It remains to be established whether these H3 receptors belong to a receptor subtype different from those previously described since the potency ofphenylbutanoylhistamine and phenylpropylhistamine as H3-receptor antagonists was probably greatly underestimated by additional properties of both drugs.