This paper emphasizes that the macrodynamics of the terrestrial magnetosphere is more effectively treated in terms of the primary variables B and v (the B, v paradigm). The common practice of relating the dynamics to E and j (the E, j paradigm) provides direct answers in a variety of symmetric cases, but breaks down in even so simple a static problem as a flux bundle displaced (perhaps by reconnection with the magnetic field in the solar wind) from its normal equilibrium position in a static dipole field. The essential point is that a direct derivation from the equations of Maxwell and Newton leads to field equations written in terms of the continuum fields B and v. The equations can be recast in terms of E and j, of course, but they are then unwieldy, being integrodifferential equations. Hence the E, j paradigm, when correctly applied, is seriously limited in its effectiveness in dynamical problems. Circumventing the limitations with the common declaration that E is the prime mover, actively penetrating from the solar wind into the magnetosphere, provides dynamics that is unfortunately at variance with the results that follow directly from Maxwell and Newton. The paper outlines the standard derivation of the basic field equations and then goes on to treat a variety of circumstances to illustrate the effectiveness of the deductive B, v paradigm in the continuum dynamics of the magnetic field and plasma. There is no attempt to develop a comprehensive model of magnetospheric activity. However we suggest that the ultimate task is more effectively attacked with the B, v paradigm. ¿ American Geophysical Union 1996