This paper presents the mesh selectivity of a net pot for white-spotted conger eel Conger myriaster. A cover-net fishing experiment was carried out in October 1995 and 1996, using five mesh openings; 11.6, 13.6, 15.5, 18.1, and 21.0mm with a cover-net of 10.8mm mesh opening. Each mesh-size net pot had a slightly different mesh selectivity curve in terms of R(=G/P), girth ratio G to mesh perimeter P, although mesh selectivity of four mesh sizes in a small trawl codend was described by a single curve in terms of R. In the net pot, the net was attached to the frame of the pot with some tension. The distortion of mesh shape was probably less than that of the trawl net. Moreover, because the twin diameter of each mesh size net was the same in the experiment, for smaller mesh size the ratio of twin diameter to mesh length was relatively larger. This means that it was a little harder for conger eels to distort the mesh shape while passing through the smaller mesh. Selectivity ranges in terms of R, R75-R25 of net pot were larger than those of the small trawl codend. Conger eels are likely to escape from the pot just during hauling up, because they try to stay inside as long as bait remains. In a towed-trawl cod-end conger eels have more chance of escaping through the mesh, which is another possible reason why the trawl codend has a narrow selectivity range.