Various forms of area protection have long been used in the management of marine capture fisheries, and there are now increasing calls for wider use of marine protected areas (MPAs), including no-take areas (reserves), to resolve issues of fisheries sustainability. However, global policy has preceded the development of generic models and experience to underpin the design of effective MPAs for sustainable fisheries. Four main classes of potential benefits can be identified for MPAs in fisheries-biological support for the exploited stock, socioeconomic benefits for the fishers, biodiversity benefits for the fished ecosystems, and increased fishery stability. The extent to which any MPA will achieve these benefits depends on its design, its integration with the broader fishery management system, and the effectiveness of the existing stock management. An MPA may deliver only limited additional support for a stock that is already well managed, but an appropriately designed MPA may be able to offer other benefits to the fishery such as an ecological offset for the effects of fishing on habitats. Delivering these benefits depends on establishing a priori design principles and decision rules in relation to each specified objective that an MPA is established to achieve. Measuring the benefits requires a performance assessment and reporting system, including specific performance indicators, that is directly linked to the objectives of the MPA. The biggest single threat to the success of fisheries MPAs is model failure-the situation where MPAs are established that do not reflect the implicit and explicit sustainability objectives of the fishery. Fisheries MPAs established without comprehensive design and performance assessment systems are likely to be neither successful nor secure and are unlikely to meet medium-term or long-term sustainability objectives.