The fiber optic gyroscopes is based on the Sagnac Effect, and the two lights propagating in different directions produce a phase difference when the fiber loop rotates. However, when the gyroscope is stationary, that is, the medium is stationary, the applied radial and axial magnetic field gyroscope will also produce a phase difference, and the phase difference is directly proportional to the magnetic field strength. At present, only the magneto-optic effects on the gyroscope are the Faraday Effect and the Cotton-Mouton Effect. In order to verify whether the axial magnetic field is the role of the above two magneto-optic effects, the fiber optic gyroscope is specially modified to let the two beams of light in the optical fiber propagate in the same direction, and then add the radial and axial magnetic fields to the fiber loop, that is, two experiments, B-1 and B-2. Through experiments, we first excluded the possibility of the Cotton-Mouton Effect. Two experiments, radial magnetic field A-1 and axial magnetic field A-2, were then performed in normal mode. Calculate acording totheory, the Faraday Effect of the A-1 experiment is at least 367 times higher than that of theA-2 experiment, but the experimental results are only 1.56-fold. Therefore, this abnormalexperimental phenomenon is a major discovery. The phase difference produced by the axialmagnetic field will not be caused by the Faraday Effect, and the existing theory cannot give areasonable explanation to the abnormal experimental phenomenon, which also means thatthere must be an unknown photoelectric medium.