There is in magnetospheric physics today a controversy of sorts, involving the nature of the process by which the kinetic energy of the solar wind is tapped to power magnetospheric phenomena and related geomagnetic disturbances. The controversy is conveniently described in the language of signal processing [Akasofu, 1985]: an input signal composed of a suitable combination of solar wind variables is processed by the magnetosphere to produce an output signal consisting of a suitable combination of geomagnetic activity indices and/or magnetospheric variables. Akasofu advocates the view that the output signal, when properly measured, is a linear function of the input signal, and that the magnetosphere is therefore “directly driven” (i.e., it responds passively, after a fixed time lag, to a given solar wind energy input). In the alternative view the magnetosphere plays a more active role: the system response depends not only on incident solar wind variables (allowing a suitable time lag) but also on the recent dynamical history of the system. This second view is described by Akasofu as the “unloading” scenario because it corresponds to a physical model in which solar wind kinetic energy is collected and stored temporarily in the magnetic field of the magnetospheric tail and is subsequently— in a manner determined at least in part by internal magnetospheric processes—“ unloaded” to the ionosphere and inner magnetosphere to produce readily observable geomagnetic effects.