Muscles paralyzed by chronic (>1 year) spinal cord injury fatigue readily. Our aim was to evaluate whether the fatigability of paralyzed thenar muscles (n=10) could be reduced by the repeated delivery of variable versus constant frequency pulse trains. Fatigue was induced in four ways. Intermittent supramaximal median nerve stimulation (300 ms duration trains) was delivered at: 1) constant high frequency (13 pulses at 40 Hz each second for 2 minutes); 2) variable high frequency (each second for 2 minutes). The first two intervals of each variable frequency train were 5 ms and 20 ms. The remaining pulses were evenly distributed in time across 275 ms. The number of pulses varied for each subject such that the force time integral in the unfatigued state matched that evoked by a constant 40 Hz train; 3) constant low frequency (7 pulses at 20 Hz each second for 4 minutes); 4) variable low frequency (each second for 4 minutes). The pulse pattern was the same as that for variable high frequency except that the force time integral was matched to that produced by the constant low frequency stimulation. These same experiments were performed on the thenar muscles of 5 able-bodied control subjects. The variable high frequency trains used to fatigue paralyzed and control muscles had an average (± SEM) of 12±2 and 10± 1 pulses respectively. Variable low frequency trains had 7 ± 1 and 6 ± 1 pulses respectively. Significant mean force declines of comparable magnitude (to 20-25 % initial fatigue force or to 13-21 % initial 50 Hz force) were seen in paralyzed muscles with all four stimulation protocols. The force reductions in paralyzed muscles were always accompanied by significant increases in half-relaxation time and decreases in force-time integral, irrespective of the stimulation protocol. Significant force decreases also occurred in control muscles during each fatigue test. Again, these force declines were similar whether constant or variable pulse patterns were used at high or low frequencies (to 40-60 % initial fatigue force or to 29-36 % initial 50 Hz force). The force reductions in control muscles were significantly less than that seen in paralyzed muscles, except when constant high frequency stimulation was used. The variations in stimulation frequency, pulse pattern and pulse number used in this study therefore had little influence on thenar muscle fatigue in control subjects or in spinal cord injured subjects with chronic paralysis.