The objective of this study was to improve recording techniques for the recognition of gastric electrical dysrhythmias, and to explore the potential of pharmacologic agents to "evoke" gastric dysrhythmias. Eighteen healthy volunteers participated in 22 individual recordings, divided into two separate studies--a dose-response study and a randomized, double-blind study. The internal or mucosal electrogastrogram was recorded with a novel approach, using magnetic force to maintain internal electrodes in apposition with the gastric wall, whereas the external or cutaneous electrogastrogram, manometric activity, and respiration were measured by conventional methods. Analysis of simultaneous internal and external electrogastrographic signals, including both dysrhythmia and dysrhythmia-free intervals, revealed a good correspondence between the internal and external signals. In the dose-response study, 5 of the 6 volunteers intravenously infused with glucagon, in doses ranging from 3 to 22 micrograms/kg, developed gastric electrical dysrhythmias. In the randomized, double-blind study, 4 of 5 volunteers intravenously infused with glucagon (7 micrograms/kg) developed gastric dysrhythmias that were recognized by our improved techniques. Dysrhythmias, defined by visual analysis, consisted either of "tachygastria" (greater than or equal to 6 cycles/min for greater than or equal to 1 min) or "bradygastria" (greater than or equal to 1 cycle/min for greater than or equal to 1 min) and were evident on both internal and external electrogastrograms. Dysrhythmias were usually associated with absence of antral phasic pressure activity and frequently with nausea.