It has been reported by several investigators that some incompatible flowers self-pollinated were rendered effective through a simple treatment on their stigmas. SEARS (1937) has shown that in the broccoli mutilated flowers selfed and kept in moist chamber set numerous seeds, but control flowers selfed produced no seed at all. Since there was no evidence that removal of more than a thin layer of stigmatic tissue further increased fertility, he concluded that the inhibitory reaction is limited to the surface stigmatic cells. As this paper attracted my attention, I tried a similar experiment to this with self-incompatible individuals of the Japanese radish. In this experiment the portion of stigmas removed was as follows; (1) none (control), (2) one-half stigma, and (3) stigma. Mutilated flowers were applied with their own pollen, and kept in moist chamber. The results obtained were fully in accord with those of SEARS (1937). Each case where stigmatic material was removed, was vastly superior in pod-setting and seed production to the control. The best result was obtaind in the case of one-half stigma. SEARS (1937) demonstrated that in the radish as well as in the broccoli, self-incompatibility is attributed to decreased germination of incompatible pollen and to no penetration of tubes into the stigma. The same pollen behavior as this has recently been corroborated in the Japanese radish (TATEBE 1938). In considering these results, it may be assumed that in the Japanese radish the inhibitory reaction is chiefly limited to the surface layers of the stigma at the flowering time, and that the removal of this region permits self-fertilization.