The title chosen for this renewal of the U.S.‐Japan prediction seminar series reflects optimism, perhaps more widespread in Japan than in the United States, that research on earthquake prediction has progressed to a stage at which it is appropriate to begin testing operational forecast systems. This is not to suggest that American researchers do not recognize very substantial gains in understanding earthquake processes and earthquake recurrence, but rather that we are at the point of initiating pilot prediction experiments rather than asserting that we are prepared to start making earthquake predictions in a routine mode.For the sixth time since 1964, with support from the National Science Foundation and the Japan Society for the Promotion of Science, as well as substantial support from the U.S. Geological Survey (U.S.G.S.) for participation of a good representation of its own scientists, earthquake specialists from the two countries came together on November 7–11, 1983, to review progress of the recent past and share ideas about promising directions for future efforts. If one counts the 1980 Ewing symposium on prediction, sponsored by Lamont‐Doherty Geological Observatory, which, though multinational, served the same purpose, one finds a continuity in these interchanges that has made them especially productive and stimulating for both scientific communities. The conveners this time were Chris Scholz, Lamont‐Doherty, for the United States and Tsuneji Rikitake, Nihon University, for Japan.