We analyze the atmospheric neutrino experiments of SuperKamiokande including zenith angle dependences using the three-flavor neutrino framework with the hierarchy m 2 1 ≈ m 2 2 m 2 3 . Taking into account the terrestrial, solar neutrino experimental data and the atmospheric neutrino experiments, including the sub-GeV and multi-GeV data in Su-perKamiokande, a large angle solution in the solar neutrino experiments is favored and the range of the mass parameter ∆m 2 23 is restricted in the range 0.08 eV 2 -2 eV 2 . Allowed regions of mixing parameters are θ 13 < 4 • , 27 • < θ 23 < 32 • for ∆m 2 23 = 1 eV 2 and θ 13 < 3 • , 28 • < θ 23 < 33 • for ∆m 2 23 = 0.1 eV 2Downloaded from 148 T. Teshima and T. Sakai m 3 partially including the zenith angle dependences. In this work, we perform the full analysis of zenith angle dependences in a SuperKamiokande atmospheric neutrino experiment 2) and analysis of Chooz data 11) in terrestrial experiment.In the three-flavor neutrino framework with the hierarchy m 1 ≈ m 2 m 3 , there are 5 parameters, sin 2 2θ 12 , θ 13 , θ 23 , ∆m 2 12 = m 2 2 − m 2 1 and ∆m 2 23 = m 2 3 − m 2 2 , concerned with neutrino oscillation. For solar neutrino deficit, the MSW solution 13), 14), 12) taking into account the matter effects predicts the large angle solution sin 2 2θ 12 = 0.6-0.9, ∆m 2 12 = 4 × 10 −6 -7 × 10 −5 eV 2 and small angle solution sin 2 2θ 12 = 0.003-0.01, ∆m 2 12 = 3 × 10 −6 -1.2 × 10 −5 eV 2 for θ 13 = 0 • -20 • . These large angle and small angle solutions are merged for θ 13 = 25 • -50 • . The vacuum solution for solar neutrinos is obtained as ∆m 2 12 ∼ 10 −10 eV 2 . 15) In terrestrial neutrino experiments, Fogli et al. 16) have thoroughly analyzed these experimental data and obtained the allowed regions in the tan θ 2 13 -tan 2 θ 23 plane for various values of the mass parameter ∆m 2 23 . We also analyzed the terrestrial neutrino experiments, 12) in which we have not included the Chooz experiment, and obtained results similar to those of Fogli et al. The allowed region in the tan 2 θ 13 -tan 2 θ 23 plane so obtained, for example for ∆m 2 23 = 0.02 eV 2 , is tan 2 θ 13 > 25, tan 2 θ 13 < 0.04 for all values of tan 2 θ 23 . For higher values of ∆m 2 23 , the allowed region in the tan 2 θ 13 -tan 2 θ 23 plane is more restricted, and for lower values, the allowed region spreads out, and for ∆m 2 23 < 0.001 eV 2 there is no restriction in the tan 2 θ 13 -tan 2 θ 23 plane. Atmospheric neutrino experiments have been analyzed in the three-flavor neutrino framework with the neutrino mass hierarchy m 1 ≈ m 2 m 3 by many authors. 17), 18), 12) The atmospheric neutrino anomaly of Kamiokande experiments 19) has been explained by ν µ ↔ ν τ oscillation with ∆m 2 23 ∼ 10 −2 -10 −1 eV 2 . These discussions contain the zenith angle dependences of sub-GeV and/or multi-GeV Kamiokande data. The zenith angle dependences in recent SuperKamiokande atmospheric experiments 1), 2) have decreasing character not only for the multi-GeV experiment but also for sub-GeV experiment. We have analyzed these depe...