This study tested the hypothesis that eccentric exercise (ECC) would increase force fluctuation for several days following exercise; however, concentric exercise (CON) would not produce such an effect. Twelve men performed six sets of five reps of dumbbell exercise of the elbow flexors eccentrically with one arm and concentrically with the other, separated by 4-6 weeks, using a dumbbell set at 50% of maximal voluntary isometric contraction (MVC) measured at 90 degrees of elbow flexion. MVC, range of motion (ROM), upper arm circumference, plasma creatine kinase activity (CK), myoglobin concentration (Mb) and muscle soreness were assessed before, immediately after, 1 h and 1-5 days following both exercise bouts. Force fluctuations during 30, 50 and 80% MVC were quantified by coefficient of variation (CV) of the force data (sampling frequency: 100 Hz) for 4 s. Significantly (P < 0.01) larger changes in MVC, ROM, and upper arm circumference were evident following ECC compared to CON, and only ECC resulted in significant (P < 0.01) increases in CK and Mb, and development of muscle soreness. Significant (P < 0.01) differences existed between ECC and CON for changes in force fluctuations. CV increased significantly (P < 0.01) immediately and 1 h after ECC from baseline for 30, 50, and 80% MVC without a significant difference among the intensities, and no significant changes in CV were evident following CON. It was concluded that increases in force fluctuation were peculiar to ECC, but did not necessarily reflect muscle damage.