The purpose of this study was to test two hypotheses suggested by a motivational account of hypnotic phenomena: (1) The performance on the hand dynamometer task in the hypnotic state without additional motivating suggestions (hypnosis per se) would be better than the performance in the waking state with no suggestions. (2) The performance on the hand dynamometer task in the hypnotic state with task motivating suggestions would be better than the performance in the waking state with the same suggestions, and also better than the performance in the hypnosis per se. Thirty university male students were employed as subjects (Ss) in the experiment, who were able to pass the item of arm catalepsy in the Harvard Group Scale of Hypnotic Susceptibility (HGSHS). Ss performed the hand dynamometer task under three experimental conditions which were completely counterbalanced (hypnosis per se, and hypnotic and waking states with task motivating suggestions) and three base level conditions, all intertrial intervals being 20 min. Ss were hypnotized with aid of HGSHS and were tested after passing the item of arm catalepsy. The results of Newman-Keuls test revealed that the hypnotic and waking suggestions resulted in significant increments in performance, but that the difference of the increment between both conditions was not significant. That is, two hypotheses were not confirmed here. However, the results of cluster analysis indicated that on the performance of Ss of high and medium levels of susceptibility to hypnosis, the two hypotheses were supported and that the induction of hypnosis tended to depress the performance of Ss of low susceptibility to hypnosis. These findings seem to suggest that the hypnotic susceptibility is one of the most important factors to determine the effects of hypnosis on muscular strength, and to provide an insight into the contradictions of hypnotic literature.