In animal studies, the motor nerve conduction velocity (NCV) is related to the axonal diameter of peripheral nerve. In humans, differences of NCV between dominant and non-dominant forearms have not been confirmed by the differences of the axonal diameter of peripheral nerve. Therefore, the purpose of this study was to measure the cross-sectional area of human ulnar nerve fascia (nCSA) directly and to examine the relationship between the NCV of the ulnar nerve and the nCSA in humans. The thirty healthy subjects were volunteered for this study. Electric stimuli were applied over the two separate points of the ulnar nerve and the latencies of this action potential in abductor digiti minimi muscles were measured, respectively. The NCVs in the forearms were calculated as the distances between two stimulating electrodes divided by the differences in their latencies. The five nCSA images from the medial epicondyle of the humerus to the ulnar head at equal intervals were measured by ultrasonography and averaged in each forearm, respectively. The maximum forearm circumferences (MFC) of both forearms were measured by a measuring tape. The NCV, nCSA and MFC were significantly greater in dominant than in non-dominant forearms. Although there is a significant positive relationship between the nCSA and MFC, nCSA cannot necessarily influence NCV. These results suggest that the size of human ulnar nerve fascia may not be necessarily related to the axonal diameter of ulnar nerve.