The tokamak is a marvelous device for plasma confinement, but can it be made into a commercially viable reactor? This is a question being asked more and more often by the electric utilities and the informed public. As a magnetic container of hot plasma, the tokamak is without peer. Now that the complicating factor of impurity radiation has been removed, the self‐healing properties of the tokamak discharge, leading to good confinement scaling in both the collisional (MIT Alcator) and the collisionless (Princeton Large Torus) regimes, have become apparent. But satisfying the temperature, density and confinement time conditions for fusion is only a part of the story. The ultimate users of fusion—the electric utilities—are even more vitally interested in such factors as engineering feasibility, reliability and ease of maintenance, overall efficiency, total plant cost, small plant size, and safety and environmental impact. Since the tokamak was developed from the standpoint of plasma stability, there is concern that it may not be ideal from those other viewpoints. Indeed, there may be room for improvement in the accessibility allowed by a tight torus, in the high technology required for auxiliary heating, fueling, and treatment of wall surfaces, and in the costly equipment needed in breeding and containing tritium. Feasible solutions to these difficult engineering problems have been suggested—but are there better solutions?