Potentiation of odor by taste in rats was tested in a variety of situations. In three experiments, almond odor and saccharin taste were presented either as a single conditioned stimulus (CS) or as a compound CS and followed by either toxic lithium chloride or footshock. Extinction tests with the almond and saccharin components were then given. In single CS-toxin experiments, taste was more effective than odor, and after compound conditioning, the taste component potentiated the odor component. Conversely, in single CS-shock experiments, odor was more effective than taste, and after compound conditioning, no potentiation was observed. Rather, interference effects were observed. In Experiments 1 and 2, the addition of taste disrupted odor CSshock conditioning, and in Experiment 3, odor interfered with taste CS-shock conditioning. Visceral feedback is apparently a necessary unconditioned stimulus for the potentiation of odor by taste. These data support the neural convergence and gating hypothesis of flavor aversion conditioning.When compared as single conditioned stimuli (CS) for the rat, odor is a weak cue for delayed lithium poisoning whereas taste is a strong cue (Hankins, Garcia, & Rusiniak, 1973;Hankins, Rusiniak, & Garcia, 1976). However, when both cues are combined into a compound CS and followed by delayed lithium-induced illness, taste potentiates odor. After compound conditioning, odor becomes highly effective when tested alone and may become even stronger than taste (Palmerino,