Existing theories of the cusp containment of quiescent plasma are reviewed. Some technical and economic requirements of any fusion reactor are then summarized and used to deduce the probable dimensions of open-ended high-β reactors. It is shown that such reactors must be long, pulsed, and of very high output power. Crucial physical problems are the microstability of the high-β plasma boundary and, to a lesser extent, problems associated with energy transport by hot electrons.Assuming that these physical and technological problems are capable of solution, it is shown that mirror-symmetric versions of the hybrid-cusp discussed previously by Bickerton may have significant advantages over the simpler linear θ-pinch geometry.