XXIII.—On some disputed points in teleostean embryology