“…In time, morphological classification schemes provided some real benefits in guiding physical explanations, and thus today the field is better off than, say, the magnetospheric community's attempt to agree upon what constitutes a substorm. The F region's classic "plasma bubble" encountered by a satellite sensor, the radar "backscatter plume," and the "airglow depletion" captured in an all-sky imager all refer to widely accepted views of an ESF event.The statistical occurrence patterns of ESF are well known [Aarons, 1993], and a theoretical foundation based on the Rayleigh-Taylor gravitational instability mechanism [Haerendel, 1973;Ossakow, 1981;Haerendel et al, 1992] is both widely accepted and used successfully in computer simulations [Zalesak et al, 1982; Maruyarna and Matuura, 1984; Maruyarna, 1988;Kuo et al, 1998]. What remains to be understood is why, during the so-called "ESF season" at a given longitude, ESF does not occur every night.…”