1. Mastication was produced by stimulation of the right motor-sensory cortex of urethan-anesthetized rabbits with 15-s trains of shocks (1-ms duration) at 50 Hz. Movements of the lower jaw and jaw muscle electromyograms (EMGs) were recorded on magnetic tape for later computer analysis. 2. The stimulus site was chosen, and stimulus intensity adjusted, so that stereotyped movements were produced that included a wide swing of the mandible to the left side during jaw closure. 3. Control trials were alternated with trials in which a steel ball (2 mm diam) was thrust between the anterior molar teeth on the left side and left in place for several seconds. 4. When the obstruction was first introduced, a jaw opening reflex was sometimes evoked if the ball struck the buccal surface of the advancing mandibular molar teeth. Thereafter, when the ball was crushed between the occlusal surfaces of the teeth, no jaw opening reflex was seen. 5. Instead, the amplitude and duration of all the jaw closing EMGs increased, beginning at least 12 ms after contact with the ball. This caused a prolongation of the slow closing (SC) phase of the cycle that, coupled with a delay in the start of activity in the digastric muscle (jaw opener), prolonged the cycle by more than 60 ms. 6. During the SC phase of the obstructed trials, the medially directed grinding stroke was exaggerated because of an increase in the contraction of the contralateral zygomaticomandibular and anterior temporal muscles. 7. After collecting data, the sensory nerves to the maxillary and mandibular anterior molar teeth were cut to eliminate feedback from the periodontal pressoreceptors. Control and obstructed trials were repeated. 8. Following denervation, the obstructed cycles were of shorter duration. The mandible still moved to the right during SC in some animals, but the increase in closer muscle EMG activity was much reduced. 9. We conclude that periodontal receptors provide positive feedback to the jaw closing muscles during mastication. This is supplemented by input from other receptors, probably muscle spindles. In addition, an increase in periodontal feedback prolongs the SC phase and the early phases of the opening stroke.

We employed the retrograde neural tracing method using the pseudorabies virus bartha strain (PRV-Ba) to investigate the projection from accessory trigeminal nucleus (Acc5) or accessary facial nucleus (Acc7) to hypoglossal nucleus (HP). After injection of PRV-Ba into the bellies of the digastric muscle, the PRV-Ba containing neurones were observed in Acc5 of the cranial belly injected rats and Acc7 of caudal belly injected rats ipsilaterally, but not in HP. These results suggest that Acc5 and Acc7 may not project to HP and that movements of the digastric muscle are not related to cooperation with tongue movements during mastication.