Tank experiments were conducted to examine the effects of hang-in ratios on the size-selectivity of gillnets. Nets of three hang-in ratios, 0.3, 0.5 and 0.7 were tested for carp Cyprinus carpio and rainbow trout Oncorhynchus mykiss. The series of catch ratios in length classes were calibrated to relative ones to exclude the catch efficiency factor that was independent from fish size but different amongst hang-in ratios. The relative catch ratios were calculated as best fitted to the selectivity curves obtained by Kawamura and Matsuoka's method. For both species, the selectivity characters assessed on the relative catch ratios, such as the maximum-selectivity lengths, selective length ranges and slopes of the curves obtained from wedged fish agreed with each other well among the three hang-in ratios. Selective length ranges were expanded toward either greater or smaller sizes of fish as the numbers of entangled catch increased when a hang-in ratio was greater. It was concluded that hang-in ratios did not affect the wedge function of a mesh and, consequently, the wedged-catch selectivity was not affected, although the hang-in ratios affected the catch efficiency because the numbers of meshes in a unit area of net panel were changed.