We report off-normal angle-resolved photoemission data from NiO(001), which is complementary to the earlier normal-emission result. On the one hand, the energy position of the bands along this direction and the dispersion of the lowest oxygen band agree with the band calculation very well. On the other hand, the existence of the valence band satellite and the absence of the energy band in the gap region do not agree with the band calculation. For the first time, interesting movement of the Ni d7 satellite has been observed as a function of the emission angle and the photon energy, which is most likely a result of the satellite dispersion. © 1991.