In order to document the cytophysiology of glucagon release by A2 cells, isolated perfused rat pancreases were stimulated by arginine after exposure to colchicine, vincristine, cytochalasin B, cycloheximide and puromycin. Colchicine, 10−4 mol/l, enhanced the early phase of glucagon release after a short exposure time (35 min). This potentiation vanished with longer exposure (45 to 100 min) and was replaced by a reduction in glucagon release, affecting both phases of secretion, but more the second phase than the first. When exposure time was kept constant (45 min) and the concentration of colchicine varied, a similar potentiating effect was observed for colchicine 10−7 mol/l (first phase) and 10−6 mol/l (both phases); an inhibitory influence was apparent with higher concentrations (10−5 to 10−3 mol/l) affecting mainly the second phase of release. Similar results were obtained with vincristine, 10−5 mol/l (infused for 35 to 100 min) and vincristine 10−8 to 10−4 mol/l, after a constant 45 min exposure. These findings are compatible with a participation of microtubules in the dynamics of glucagon release: a reduction of both phases of secretion being caused by an extended disruption in the microtubular apparatus, whereas more limited disturbance of this system is associated with facilitated glucagon release. Cytochalasin B, 10 μg/ml, potentiated the A2 cell response to arginine and the basal glucagon release in the presence of glucose 4.4 mmol/l. Puromycin 100 μg/ml and cycloheximide 0.5 mg/ml reduced selectively the second phase of glucagon release; cycloheximide, 1 mg/ml, completely inhibited the basal release, and the early and late phases of the arginine-stimulated release. These results suggest that emiocytosis in A2 cells is very similar to that in B cells. The specificity of antitubulins and other agents as dissecting tools for the secretory process depends upon the experimental conditions.