Résumé : The effects of Gila monster venom on dispersed rat pancreatic acini were compared with those of secretin and VIP. The efficacy of the venom in terms of amylase release was much higher (a 24-fold increase over basal secretion) than that of secretin (a 4-fold increase) and VIP (+ 40% only). On the other hand, cyclic AMP levels increased 12-fold with the venom, as compared to 18-fold with secretin and 16-fold with VIP. The venom, VIP and secretin all displaced 125I-VIP and the competition curve with the venom was steeper, suggesting that all VIP-recognizing receptors bound the venom with the same affinity. VIP receptors were, however, not responsible for the release of amylase provoked by the venom since VIP (and secretin) did not inhibit the secretory action of the venom. The venom exerted no effect on 45Ca efflux and its secretory effect did not depend on the presence of external calcium. Besides, the effect of CCK-8 on amylase release was additive with the effect of the venom. A first exposure to the venom induced a refractoriness to itself with respect to amylase release but not in terms of cyclic AMP increase. In conclusion, Gila monster venom may contain one component binding to VIP/secretin receptors with resulting cyclic AMP elevation. A second venom component may be responsible for the high secretory efficacy, without involving cyclic AMP or calcium efflux.