It has recently been shown that, in human medial gastrocnemius [MG], the fibres of motor units activated during standing occupy localised territories in the longitudinal plane (1). Coupled with the muscle’s pennate fascicle architecture there is potential for units with similar intrinsic properties to be concentrated at different points along its length. Aponeurosis material properties and fascicle geometries tend to vary along the length of MG (2-3). Fibres may therefore be arranged within the muscle to occupy a local environment which optimizes their intrinsic properties. Recent evidence indicates that fibres of units activated during standing are more highly represented in the distal portion of MG (4), suggesting low threshold units are preferentially represented in this muscle region. We further investigate this observation by testing whether fibres of motor units with different activation thresholds have different spatial distributions along the proximal-distal axis of MG. Twelve participants completed the study which was approved by the Local Ethics Committee. Surface EMGs were recorded using 128 electrodes placed over MG. Electrically evoked twitches were elicited by stimulating the tibial nerve branch. Pulses were delivered for 50s with current pulse amplitude increased every 10 stimuli to elicit an increase in muscle activation. Representative ankle torques were calculated from recorded force plate data. The degree of MG stimulation was determined from M-wave root mean square [RMS] values. The location of the highest M-waves along the length of the electrode grid was calculated using barycentre coordinates from the spatial distributions of the incremental M-wave values computed for each stimulation level. There was a positive linear relationship between mean ankle plantar flexion torque and M-wave amplitude [p<0.01, R2=0.98]. Increasing stimulation amplitude therefore increased the number of activated motor units. In eight participants the highest M-wave amplitudes occurred in proximal electrode channels for low amplitude stimulation and in more distal channels at higher amplitudes. As M-wave amplitude is spatially associated to the distribution of activated muscle fibres, the shifts in RMS distribution suggest that, in these participants, there was regional organization along the proximal-distal muscle axis of muscle units according to their activation threshold. In three participants larger M-waves moved to proximal electrode channels with increasing stimulation amplitude. In one participant no spatial shifts in M-wave distribution occurred. Such variability between participants warrants further investigation to establish whether it is the result of the experimental methodology or represents variation across the general population.