[en] The scientific feasibility, engineering practicability and economic prospects for an Accelerator Breeder are reviewed. The scientific feasibiliity of high power accelerator components rests on a firm basis as a result of technical advances made in recent years but there is a need to combine all components in a demonstration model working under realistic conditions. The engineering practicability of Accelerator Breeder components should be tested in a staged development culminating in a full-scale demonstration plant. The economic assessment depends on calculations of allowed and estimated capital costs of an Accelerator Breeder for a CANDU system operating on the Th-U fuel cycle. The results indicate that the ratio of estimated to allowed capital cost is approximately 3.5 for a breeder with a 2% enriched uranium metal blanket and for separated U235 valued at 48 $/g