Maximum voluntary manual pull was measured on 18 Ss. These values were then used to determine the forces at 10% intervals from 30% to 100% of maximum strength for each S. In subsequent sessions, Ss started an exertion at one of these effort levels and then maintained an equal sensation of effort for 60 sec. The data at all 8 effort levels were closely approximated by triple-exponential decay functions. The percentage error between observed and calculated values ranged from 0.7% for the 100% effort data to 2.5% for the 20% condition. Since the function generated by a maximum voluntary contraction was not qualitatively different from the sub-maximal effort functions, it would seem that the fatigue curve may be viewed as a maximum effort function and, as such, involves the same processes that generate the lower intensity effort functions.