Twenty-eight possible cases of multiple myeloma were identified from two lists of veterans who said they had served in Hiroshima or Nagasaki. compiled one of the lists as part of its NTPR program. DNA The other list was provided by the National Association of Atomic Veterans (NAAV), which had polled its membership of about 2,000. The DNA list contained 687 names, and the NAAV list approximately 500 names. The NTPR Service teams and participating NAS staff members screened military records of the 28 veterans possibly having multiple myeloma. They eliminated nine of the veterans because their records did not confirm military assignments to Hiroshima or Nagasaki. Clinical records were sought from the 19 remaining cases. The veteran or, if deceased, his next-of-kin was asked for permission to obtain his medical records (including X-rays and microscope slides) from the appropriate medical authority. Six more cases were eliminated, five of them because the veterans or next-of-kin did not respond to NAS inquiries and one because a physician did not respond to the request for medical records. Four cases were eliminated from the remaining 13 when further military record searches revealed that two of the personnel had not been assigned to Hiroshima or Nagasaki and the medical records of the other two made no reference to multiple myeloma. The panel confirmed nine cases of multiple myeloma among the Hiroshima/ Nagasaki veterans. Five of the cases had been assigned to the Nagasaki occupation; the other four were associated with Hiroshima. All cases were diagnosed between the ages of 51 and 61, the time when the disease normally appears. On the basis of multiple myeloma incidence rates reported by the National Cancer Institute and assuming that at least 20,000 men were assigned to occupation duty at Nagasaki, the panel calculated that 9.5 cases of the disease would be expected by 1980 if all of the troops had been between the ages of 15 to 19 years at the time of the occupation. At least 18.2 cases would be expected if the ages had been between 20 and 24, and 29.2 cases would be expected if the ages had been between 25 and 29 in 1945. 190 Similar figures