(fig 1). Background EMG activity measured during 100 ms prior to platform tilt ranged between 2 and 5% of the size of short latency. We decided to measure EMG integrals from raw EMG records because an earlier study revealed a much higher interindividual variation when we measured EMG integrals in relation to maximal voluntary level of EMG.

Long-latency electromyographic (EMG) responses of the first dorsal interosseus muscle were evoked by short displacements of the index finger in healthy subjects and in patients with Huntington's disease (HD). In all 20 healthy subjects the early spinal response (mean latency 31.5 ms) was followed by a reproducible later reflex response with a mean latency of 56.5 ms. The activity pattern of single motor units of the stretched muscle was similar to that of the surface EMG. Thus all single motor units tested could be active during either the first or second response phase, but never in both in a given trial. Of the 50 patients with HD, the late EMG response was missing completely in all but 7, but the early spinal component was almost identical to that of the control group. Double stretches at an interval of 25 ms evoked two similar EMG responses in these patients, proving that the motoneuron pool is still excitable during the time at which the second response would have appeared in healthy subjects. The reflex responses of the thenar muscles evoked by electrical stimulation of the median nerve were examined during a voluntary opposing contraction of the thumb in both groups of subjects. In normal subjects, two EMG responses could be distinguished with latencies similar to those of the mechanically elicited responses. Patients with HD again lacked the second response, although the first spinal response was always present. The results are discussed with respect to different proposals concerning the origin of long-latency responses in human muscles. At least for distal hand muscles, the results suggest that the long-latency responses are long-loop reflexes.

No abstract

SUMMARY Electromyographic responses to stretches of hand muscles (first dorsal interosseus) and leg muscles (triceps surae, tibialis anterior) were investigated in patients with cerebellar disorders of different locations. Stimuli consisted of short dorsiflexions of the index finger during background force and in tilting (toe up) of a movable platform on which the subject stood. The most important findings were increased long latency responses in upper and lower extremities. For hand muscles it was the late part of the long latency complex, which was increased. For leg muscles it was the long latency response in the anterior tibialis muscle, the antagonist of the stretched triceps surae. The medium latency response in the triceps surae was unaffected. Latencies of the early segmental reflexes and the long latency responses were normal except for cases with peripheral neuropathy (moderate increase in latency of all EMG responses) and diseases affecting both the peripheral nerves and the dorsal columns (for example Friedreich's ataxia). The latter leads to a pronounced delay of the short latency response and a massive delay of the long latency complex in the first dorsal interosseus and of the long latency response in the anterior tibialis muscle. Material and methods EMG responses were evoked in the first dorsal interosseus muscle (FDI) using short imposed stretches under otherwise isometric conditions. The general procedure has been described before3 and will only be briefly mentioned here. Short triangular stretches (rise time 6 ms, transient time 20 ms) were applied to the index finger with the help of a lever, on which the subject exerted a constant force of 10% maximum. The lever was coupled to an electromagnetic vibrator 71 by copyright.