In group-housed mice (ten per cage), mice removed last from their home cage always have higher rectal temperatures than mice removed first from this cage. Stress-induced hyperthermia is calculated as the difference (delta T) between the basal temperature (mouse number 1) and the end temperature (mouse number 10) when the temperature of the ten mice is sequentially measured using a 1-min interval between rectal measurements. Using this protocol, various drugs, belonging to different pharmacological classes, were tested in order to investigate their putative anxiolytic effect, measured as a decrease in delta T. Benzodiazepines (diazepam, alprazolam), alcohol, and some (flesinoxan, buspirone), but not all (ipsapirone) 5-HT1A receptor agonists had anxiolytic properties with this protocol. Clonidine (alpha 2-adrenoceptor agonist) and prazosine (alpha 1-adrenoceptor antagonist) had, but at high doses, some anxiolytic actions. Antidepressants (desipramine, fluvoxamine, nomifensine, tianeptine, amitriptyline, clomipramine, imipramine), serotonergic ligands (ondansetron, ketanserin, 1-(2,5-dimethoxy-4-iodophenyl)-2-aminopropane (DOI), fenfluramine, metachlorophenylpiperazine (mCPP), eltoprazine) and various other drugs (phenobarbital, pentetrazol, haloperidol, apomorphine, amphetamine, (+)-N-[1-methyl-2-oxo-5-phenyl-2,3-dihydro-1H-1,4-benzodiazepin-3( R)-yl]- N'-(3-methylphenyl)urea (MSD 365260), dizocilpine and acetyl salicylic acid) had no anxiolytic activity. The stress-induced hyperthermia protocol used was unable to detect anxiogenic properties of drugs, probably due to a (physiological) ceiling in the maximal end temperature. The stress-induced hyperthermia protocol with mice can be used to measure anxiolytic properties of drugs and is a fast and robust model which does not need extensive training of animals.