Energetic-particle pitch angle data and vector magnetometer data, measured along the Ogo 5 inbound orbit near midnight on three magnetically quiet days, August 2,4, and 25, 1968, were used to infer the magnetic tail field configuration for each day. For the first 2 days the particles when initially detected, showed isotropic pitch angle distributions (PAD's) (indicative of nonadiabatic motions resulting from the breakdown of the guiding center invariants &mgr; and J) but later made rapid transitions to the butterfly PAD (indicative of adiabatic guiding center motion). The lower-rigidity particles made the PAD transitions first and were followed in turn by the higher-rigidity particles as Ogo neared the earth. We have used a simple algebraic model which has allowed us to fit both the magnetic field and the pitch angle transition data. For the latter data fit, the detailed particle trajectories were followed along the model field lines to the neutral sheet. The particles were required to execute adiabatic guiding center motion if they wer started from positions on the satellite orbit nearer earth than the point of observed changes in the PAD. They were required to undergo nonadiabatic motion at some part of their trajectory if they were started at a point on the satellite orbit farther from the earth than the PAD transition point. The results are presented in terms of model coefficients for each day. The model fields are plotted, the cross-tail currents are derived, and particle motion is studied in the model fields. It was found that 79-keV electrons were in the nonadiabatic mode when they were on field lines that crossed the neutral sheet beyond 11 Re on August 2: on August 25, a less taillike day, this point was 17 ¿ 1 Re. For 10-keV protons the inferred points are 8.5 and 11.7 Re, respectively, for these 2 days. This is interesting because protons of about this energy carry a large part of the cross-tail current. It is expected that when data from particles of a wide range of rigidity are available, this method of inferring the tail field configuration should be very powerful.