I have pointed out that the A-wave led from a dorsal root under the depression of a constant current applied at the spinal ganglion leaves behind alpha peak only, and beta has disappeared completely. On the contrary, under the same conditions moving a pair of non-polarizable electrodes from the spinal ganglion to the sciatic nerve, alpha peak vanished, but beta has still remained, having no connection with employment of another kind of animals such as Japanese toad and bull-frog.But the precautions must be taken so that the results do not get mixed with the process, because the process of the depression under the same conditions will not stand together in comparison with a different kind of animals. Another quality of some interest is the similarity of the shapes of the depression represented in Fig. 4a, 4'a (a constant current was applied at the spinal gangl: of Japanese toad) and Fig. 7, 7'a (a constant current was applied at the scia nerve of bull-frog), but it is clear that the names of each nerve fibre must reversed, that is alpha into beta, in likewise, beta into alpha. In like manner as mentioned above, the similarity of the forms qf. I depression can be seen in Fig. 5, 5' (a constant current was applied at the spit ganglion of bull-frog) and Fig. 6a, 6'a (a constant current was applied at 1 sciatic nerve of Japanese toad), but the names of the twin peaks must reversed. These phenomena are interesting, but it is not safe to go any further ridi on the wings of imagination, therefore this is only a technique of fibre purifi tion and not an explanation of their mechanism. Furthermore, as regard to the problem of conduction of twin fibres (alp 6-7 m/sec: beta 4-5 m/sec at 5°C), a fancy conclusion against our ordins experiences are drawned from the experimental facts mentioned before. According to our common sense, it is easy to understand that the qu; impulse can reach the destination more rapidly than the slow one at the sa: course and the same departure. But this simple clear fact can not always be universally applicable to a case without exception, because under the deep depression by a constant current applied at the nerve trunk the fast alpha is blocked, and on the contrary, t slow beta still remained. Therefore the slow impulse wins in the contest agait our ordinary expectation.It is a great pleasure to express my deepest gratitude to Professor G. Kato, assistant professor I. Suda and the members of his laboratory for assisting and encouraging me during the course of this work.