The appetitive phase of feeding behavior in Aplysia consists of a behavioral sequence in which the quiescent animal starts to locomote and then assumes a characteristic feeding posture. In this position, head-turning responses can be elicited by a localized food stimulus (seaweed) delivered to the lips or tentacles. In response to brief (open loop) stimulation with seaweed, the animal turns toward the stimulus but greatly overshoots the target. However, the angular velocity and the final turning angle are a function of the eccentricity of the stimulus, progressively increasing with greater eccentricities. In a food-aroused animal, a brief tactile stimulus evokes turning and biting responses similar to those triggered by seaweed, which provides both tactile and chemical stimulation. Upon repeated tactile stimulation, however, the response magnitude decrements rapidly, whereas the magnitude remains high when turning responses are repeatedly elicited by food stimuli. A purely chemical stimulus sometimes can elicit a turning response, but chemical stimuli alone are much less efficacious than tactile stimuli alone. When the stimulus is maintained in a stationary position (closed loop), the animal turns until its mouth is oriented over the food. A turning response to a lateral stimulus can be reduced by an immediately following medial stimulus. To explain the above findings, we propose a form of response substitution, in which the response to the first, lateral stimulus is substituted by a weaker response to a more medial stimulus. No turning response is evoked when the animal is stimulated while performing spontaneous or evoked bites, though biting per se does not interrupt ongoing turning movements. In animals with lesions of the cerebral- buccal connectives, a food stimulus on the mouth is also followed by a reduction of the capacity of stimuli to elicit turning responses. In these lesioned animals, the food stimulus appears to elicit a bite command, though the biting behavior itself does not occur. Thus, it appears that the bite-related gating of stimuli is of cerebral origin, rather than due to the generation of the buccal motor program. The force necessary to power the turning movements was calculated from the trajectories of the movements. The results indicate that a power phase during the first half of the duration of the total movement is sufficient to generate a turn. The power phase can be followed by a brief gliding phase, and finally the movement appears to be actively terminated.(ABSTRACT TRUNCATED AT 400 WORDS)