The discharge of single motor units has been recorded from the soleus muscle of the decerebrate cat during the tonic vibration reflex elicited isometrically, to further understanding of the tremor that is seen in the reflex contraction. The reflex was elicited by pulses of vibration of 50 micrometers amplitude at 150 Hz, and up to four units were studied concurrently. 2. Individual units fired rather regularly and at a low frequency (range 4-14 Hz). The rate of firing of any unit normally fell within the frequency band of the tremor recorded at the same time. On comparing different preparations a higher frequency of tremor was associated with a higher frequency of motor firing. 3. The responses of pairs of motor units recorded concurrently during repeated production of the reflex were compared by cross-correlation analysis; over 1000 spikes from each train were normally used for this. The major of the cross-correlograms were flat with no overt sign of any synchronization between the units other than that due to the vibration. 4. Clear indications of correlated motor unit firing could be produced deliberately by modulating the amplitude of vibration at a frequency comparable to that of the normal tremor and thereby introducing a rhythmic component into the tonic vibration reflex. 5. About 20% of the cross-correlograms obtained during normal tremor showed varying amounts of an irregular 'waviness' suggesting a possible correlation between the times of firing of a pair of units. But such waves never developed steadily throughout the period of analysis, in contrast to the comparable waves produced on modulating the vibration. Similar waves were seen on cross-correlating a motor unit with an electronic oscillator, confirming that their occurrence does not necessarily demonstrate the existence of active neural interactions. 6. It is concluded that there is no strong and widespread neural synchronizing mechanism active during the tonic vibration reflex, although the possibility of some weak neural interactions has not been excluded. The findings favour the idea that the tremor in this preparation is simply the inevitable result of motor units discharging asynchronously, but at closely similar subtetanic frequencies.
