1. During investigation of the tonic stretch reflex in the unanaesthetized decerebrate cat we observed that a short train of impulses in Ia afferents from the soleus muscle (or its synergists) may cause a prolonged activity in the soleus muscle as judged by EMG and tension recordings. This excitability increase, which outlasted the stimulus train, could stay virtually constant during long periods (even minutes), but could be terminated at any time by a train of impulses in, for example, the peroneal nerve. 2. Gradation of the strength of stimulation and the duration of the train of impulses show that the amount of maintained excitability increase depends-within some limits-on the total amount of Ia impulses. 3. In paralysed preparations a short train of impulses in Ia afferents from any part of the triceps surae, caused a maintained increase of the efferent activity in the nerves to triceps surae and a maintained increase of the triceps surae monosynaptic test reflex. These experiments demonstrate the existence of a central mechanism (in the spinal cord and/or the brain stem), which is responsible for the maintained excitability increase seen in motoneurones to the homonymous and synergic muscles. 4. In acute spinal preparations it was not possible to demonstrate any long-lasting excitability increase by a train of Ia impulses. Following intravenous administration of the serotonin precursor 5-hydroxytryptophan, mimicking the tonic activity of these pathways in the decerebrate state, it was again possible to elicit the long-lasting excitability increase by a train of impulses in Ia afferents. A subsequent I.V. injection of methysergide (a serotonin receptor blocker) abolished the long-lasting excitability increase. This set of experiments demonstrates that the basic mechanism responsible for the maintained excitability increase is located at segmental level, and involves serotonergic systems. 5. It was demonstrated that activation of several ipsilateral and crossed reflex pathways by trains of impulses in cutaneous or high-threshold muscle afferents could trigger a maintained excitability increase of those motoneurone pools which were activated by the stimulation. Trains of stimuli to facilitatory regions in the brain stem could also cause a long-lasting excitability increase of motoneurones. Furthermore, activation of all reflex pathways which mediate postsynaptic inhibition to a motor nucleus (including recurrent inhibition via Renshaw cells) could terminate the prolonged excitability increase of that particular motor nucleus.(ABSTRACT TRUNCATED AT 400 WORDS)