During the Airborne Lidar and Observations of Hawaiian Airglow (ALOHA) Campaign on October 10, 1993, a wave structure with a sharp front covering significant parts of the sky was observed to move with a phase velocity of 76 m/s at a period of 4 to 5 min. Simultaneous observations of the Green line and the OH emission brightness showed a 180¿ phase difference. We have two objectives. (1) By using a dispersion formula, we first separated the true partial wave guidance for long-period waves from the background effects and showed explicitly that many partial modes previously thought to exist do not in fact exist. In this way, we can show that two very different methods for computing partial modes which previously gave very different results can now be reconciled. We showed that by lowering the period the modes became, first, virtually guided and then fully guided. (2) We then showed that the long-period S2 partial mode eventually became a two-node fully guided mode, which at periods in the range of 4.6 min, had not only phase velocities comparable to the observed result, but also a phase reversal (from the two-node structure) at altitudes in the vicinity of the two airglow emission peaks. At those periods, the dispersion curves were shown to be out of the range of the nearest kissing mode and were thus largely unaffected by interference from the stratospheric duct. The results agree well with an analytic model. The wave explanation may in some way complement the other explanations (such as the bore, or soliton) for the same event mentioned in presentations elsewhere. ¿ 1998 American Geophysical Union