Objective: Examine muscle contractile activity during electrical stimulation (ES) after spinal cord injury (SCI). Setting: General community of Athens, Georgia, USA. Methods: Eight clinically complete SCI adults (C6 to T12) 4+1 (mean+SE) years post injury and eight able-bodied adults were studied. Surface ES was applied to the left m. quadriceps femoris for three sets of 10, 1 s isometric actions (50 Hz trains, 400 ms biphasic pulses, 50 ms phase delay, 1 s : 1 s duty cycle) with 90 s of rest between sets. Current was set to evoke isometric torque that was (1) sucient to elicit knee extension with 2.3 kg attached to the ankle (low level ES), and (2) intended to equal 30% (mid level ES) or 60% of maximal voluntary torque of able-bodied adults (high level ES, able-bodied only). The absolute and relative cross-sectional area (CSA) of m. quadriceps femoris that was stimulated as re¯ected by contrast shift in magnetic resonance images and torque were measured. Results: Six+2, 20+2 and 38+4% of the average CSA of m. quadriceps was stimulated during low, mid and high level ES, respectively, for able-bodied. Corresponding values for SCI for low and mid level ES were greater (61+12 and 92+7%, P=0.0002). Torque was related to the CSA (cm 2 ) of stimulated muscle (Nm=3.536stimulated CSA+13, r 2 =0.68, P=0.0010), thus ES of a greater per cent of m. quadriceps femoris in SCI was attributed to their smaller muscle (24+3 vs 73+5 cm 2 , P=0.0001). The decline in torque ranged from 9+1 to 15+4% within and over sets for low, mid or high level ES in able-bodied. SCI showed greater (P=0.0001) fatigue (19+3 to 47+6%). Conclusion: The territory of muscle activation by surface electrical stimulation varies among SCI patients. Given sucient current, a large portion of the muscle of interest can be stimulated. The resulting torque is modest, however, compared to that attainable in ablebodied individuals due to the small size and limited fatigue resistance of skeletal muscle years after spinal cord injury.