The most commonly used animal models to evaluate the psychoactivity of cannabinoids have been reviewed. The need for suitable models is acute considering the present interest to develop drugs based on the cannabinoid moiety but preferably dissociated from psychoactivity. Conceivably, a satisfactory assay should show features of cannabinoid-induced disturbances relevant to man as well as sensitivity, specificity and simplicity. These requisites seemed better fulfilled in the monkey model. Various lines of evidence have demonstrated the close pattern of the behavioural response to psychoactive and inactive cannabinoids in man and monkeys. Rhesus monkeys showed development of tolerance and withdrawal symptoms, which have been frequently reported in humans after prolonged exposure to cannabinoids. The exposure was reported also to cause in monkeys alterations of electrical activity and organic damage in deep brain structures. The monkey model has been particularly useful to determine the relative potency of naturally occurring cannabinoids and metabolites, which was adequately compared to that in man, and to establish the structural requirements for psychoactivity in large series of synthetic new compounds. In addition it appeared that rhesus monkeys react similarly to man with respect to proposed antidotes against cannabinoids. Four newly synthetized amino-cannabinoids were tested in baboons. All these compounds were virtually void of typical cannabinoid psychoactivity but two trans-analogs differed from the cis-analogs in that they provoked bouts of vigorous scratching and yawning. This unusual drug-effect, at difference from scratching alone has not been previously observed after administration of cannabinoids. In this presentation some terms of cannabis terminology have been discussed.