dD. 368. DATA FROM A-BOMB SURVIVORS The effect of radiation on aging and life-shortening in a Marshallese population irradiated in 1954 was summarized in a report by Conard [C29]. A number of changes connected with aging was investigated and among them the opafication of the eye lens, the presence of chromosomal aberrations in perypheral lymphocytes, immuno-haematological and nephrosclerotic changes. Regarding life-shortening in particular, the number of persons exposed was too small to allow any reliable assessment. The population under study includes in fact a control group and two irradiated groups of 334 persons in total, exposed to a maximum of 175 R from fission-product gamma-radiation. 369. The situation for the Marshallese series is completely at variance with that of the A-bomb survivors in Japan, which is providing information on longterm radiation effects, including life-shortening, that will eventually form the most exhaustive source of data in the human species. The earliest reports of this series are particularly concerned with the description of the sample [B21] and with mortality from all causes [J5] and from specific causes up to 1960 [J6, A6]. 370. The report covering the period up to 1966 [B22] was based on the T-65 dose estimates and included 16.356 deaths over about 109.000 people, comprising irradiated and control groups. When malignant neoplasms were excluded from the analysis, there was no evidence that radiation might aspecifically shorten life and the excess mortality of the irradiated sample could best be explained in terms of disease-specific effects, particularly leukaemogenesis and, more gene- rally, cancerogenesis. 371. Mortality data from the Japanese sample up to 1970 [J7] allowed the fol- lowing conclusions. Although late radiation effects on human mortality could to some degree resemble non-specific manifestation of accelerated aging, the most notable effect remained by far the induction of tumours and leukaemia. This latter disease in the data reported was higher by an order of magnitude than any other cause of death. The authors could not conclude with certainty for the absence of an excess mortality, for example, from diseases of the circulatory system or from cerebrovascular conditions: but it appeared most unlikely, if there was indeed any excess, that it could approach the excess of tumour induction.