Selective stimulation of the masseteric nerve has been shown to elicit a heteronymous H-reflex in the ipsilateral temporalis muscle during voluntary clenching. However, the relation between the electromyographic (EMG) activity of the temporalis muscle and the amplitude of the H-reflex has not been previously described. In the present study, the hypothesis was tested that there would be a positive relationship between the level of EMG activity and the amplitude of the H-reflex. The direct motor response (M-response) in the masseter muscle and the heteronymous H-reflex in the anterior temporalis muscle were successfully elicited by electrical stimulation of the masseteric nerve in 12 of 13 subjects. A new automatic system was used to control the on-line EMG activity and to trigger the stimulus. In a random order, two series of 20 stimuli were delivered at each of four clenching levels (0, 25, 50, and 75% of maximal voluntary contraction). The analysis showed that both the masseteric M-response and the temporalis H-reflex were reproducible within and between series. The amplitude of the temporalis H-reflex increased significantly at higher clenching levels (ANOVA: P=0.003). Clenching at 50% and 75% of the maximal voluntary contraction caused significantly larger amplitudes of the H-reflex than clenching at 25% of the maximal voluntary contraction; at rest, no H-reflex could be recorded. There was a significant correlation between the background EMG activity in the ipsilateral temporalis muscle and the amplitude of the H-reflex (Pearson: r=0.313, P=0.008). These data indicate that the heteronymous H-reflex can be reliably elicited by means of an automatic system for stimulus delivery and that the amplitude of the H-reflex is dependent on the preceding activity of the motoneuron pool.