A conditioning (C) stimulus at group I strength was delivered during standing to the common peroneal (CP) nerve before a test (T) stimulus at several C-T intervals ranging from 0 to 150 ms. At sufficiently long C-T intervals (100-120 ms) the soleus H-reflex was strongly inhibited despite little, or no change, in the background level of EMG activity. This finding indicates that a significant portion of the inhibition occurs at a premotoneuronal level, likely via presynaptic inhibition of the Ia-afferent terminals. During standing, at C-T intervals of 100-120 ms (optimal C-T interval) a conditioning stimulus to the CP nerve of 1.5 times motor threshold (MT) intensity reduced the soleus H-reflex by an average of 45.8%(n = 14 subjects). The conditioning stimulus always produced a clear inhibition of the H-reflex during standing at these C-T intervals. The effects of this conditioning stimulus on the soleus H-reflex were then determined in the early part of the stance phase of walking. In contrast to standing, the conditioning stimulus produced little or no inhibition during the early part of the stance phase of walking (average inhibition 45.8 vs. 11.6%, n = 14 subjects). The soleus background EMG, and the soleus and tibialis anterior M-waves were essentially the same during standing and walking. Furthermore, there was no shift of the optimal C-T interval during walking. The difference in the effects of the conditioning stimulus was not due to differences in the size of the test H-reflex in each task. It appears to be due to a genuine task-dependent change in the input-output properties of the underlying spinal cord circuits. There are at least two, mutually compatible, explanations of these results. Firstly, during walking the intraspinal terminals of the afferent fibres (group Ia and Ib) conducting the conditioning volley may be presynaptically inhibited, or their input gated at the interneuronal level. Secondly, on the assumption that the conditioning stimulus is acting via the presynaptic inhibitory network in the spinal cord, it is possible that during walking this network is saturated as a result of increased central or peripheral synaptic inputs. Finally, it seems unlikely that differences in the refractoriness of the CP nerve between the tasks may be involved; the reasons for this are presented in the discussion.