Late on an April afternoon in 1988, the long awaited word from the White House reached the U.S. national security bureaucracy: President Reagan had approved the transfer of U.S.-developed nuclear submarine propulsion technology to Canada and would inform Prime Minister Brian Mulroney during his visit to Washington later that month. The opponents of the transfer, the Department of Defense, the United States Navy and the Department of Energy (the home of naval reactors), were stunned. In spite of lukewarm support from State, the NSC staff, and the Arms Control and Disarmament Agency (ACDA) for approval, how, in the face of seemingly compelling national security arguments against it, could the President say yes. Yet it had happened and there were to be no appeals. The Secretaries of State, Defense, and Energy were instructed to negotiate the necessary agreements to allow the transfer-- conditioned only to protect classified nuclear technology design information. It appeared that the issue had been lost--or had it. The dance which was to follow between the bureaucracies of the United States and Canada, and to a lesser extent the United Kingdom and France, eventually was to result in a decision by Mr. Mulroney not to seek nuclear submarines at all. What follows is an analysis of the issues involved in the approval and the final Canadian decision. The matter began in June, 1987, with the publication of a Canadian White Paper on Defense which announced that Canada would spend $8 billion to build 10 to 12 nuclear attack submarines (SSN's) to defend the Arctic against the Soviet submarine threat, and, importantly, for patrolling the Northwest Passage and territorial waters over which it claimed sovereignty. Canada, not wanting to design an SSN from the keel up, would buy existing hull and nuclear propulsion technology designs and build them under license in Canadian shipyards.