Abstract
Previously published estimates of the whole-body radiation dose expected to kill 50% of a normal human population, the LD50, have rarely been based explicitly on evidence. The difference which might result from medical treatment seems to have been markedly over-valued. The available and relevant evidence about severe haematopoietic damage in man uncomplicated by tissue necrosis is indeed very scanty. It comes from 20 cases of therapeutic whole-body exposure to gamma rays and from two criticality accidents involving nine subjects, one of whom died, and when exposure was to neutrons as well as gamma rays. The observations suggest a judgment that 4.5 Gy (450 rad) absorbed dose in the bone marrow for energetic and therefore penetrating gamma rays giving reasonably uniform irradiation of the marrow could be regarded as the LD50 in circumstances where those irradiated were protected from thermal radiation and blast damage and from neutrons and beta rays. Examination of all the available experimental data on acute lethality following whole-body exposure to low LET radiation shows a remarkably similar co-efficient of variation of the LD50 in five species of large animal. If the same value is adopted for the human species, a quantitative estimate of the human LD50 can be inferred from the human evidence discussed. This reinforces the judgement that it is about 4.5 Gy. The use of observations after criticality accidents is examined in detail in an Appendix. The Vinca accident does not clearly meet the criteria for relevancy but there is no other instance of death in man from an uncomplicated brief whole-body exposure to ionising radiation where the bone-marrow dose can be estimated, and a possible judgment made of the value for the LD50.