346 Sheet HRSAMHDS perdeNtedticcs: RADIATION STANDARDS, INCLUDING FALLOUT go into a prolonged arrest in development at a later and more sensitive stage in their maturation process than is the case in the guinea pig. This difference probably holds true for the other species mentioned, and also for the dog and human female, for which evidence exists that sterility in the female is produced only by much higher doses (several hundred roentgens)- Nevertheless, a word of caution seems due. It has been estimated that in the event of a nuclear attack on the United States of 3,000 megatons or more most survivors will receive an accumulated dose of upward of 200 roentgens, most of it in the postshelter period and at relatively low dose rates. Under these conditions, the major effect upon the surviving population might result from the sterilizing effect upon the females. We need much more information about this matter than now exists, in particular to determine whether the dose level that produces complete or partial sterility, at low dose rates, is significantly lower than the dose level that produces radiation sickness and death. Killing of the sensitive female germ cells might not be attributable to the effects of the radiation on the chromosomes and DNA of the cells. Killing of many types of somatic cells, however, has been shown by T. T. Puck, M. A. Bender, and others to be mainly because of damage done to the chromosomes and genes; in other words, damage to the heredity material which is present in every cell and not solely in the reproductive cells. The target of the radiation is the same in the case of different kinds of cells, but the transmissibility of the damage depends on something else. To be transmissible, a defect must first of all be produced in the hereditary material of a reproductive cell; and, second, it must not be so severe as to cause the death or total incapacity of the reproductive cell to function. Total reproductive damage includes the nontransmissible (sterility) effects as well as the transmissible, hereditary effects. Since the time of the last hearings on fallout before this committee, much attention has been centered on the hazards posed by the production of carbon 14 as a product of nuclear detonations. No exact appraisal of this hazard can be given at the present time, but it is clear that estimates of genetic damage must be revised upward because of the ready incorporation of carbon 14 into living tissues, and into the hereditary materials themselves, and because of the long half-life of carbon 14. The recently issued new edition of “The Effects of Nuclear Weapons” places the dosage of emitted beta radiation from carbon 14 as equal to that from cesium 137, although delivered at a very much lower dose rate; and the genetic damage from transmutation of carbon 14 into nitrogen 14 as equal to that from the emitted beta radiation. I see no reason to differ with these estimates. There are other recent findings which also increase our estimates of genetic damage done by atomic radiations. At the Oak Ridge National Laboratory, D. L. Lindsley has shown that many lethal mutations remain undetected in the usual method of screening for sex-linked excessive lethals. Estimates of the total number of lethal mutations produced by a given dose thus need to be increased by 20 percent. Findings such as this one and the role of carbon 14 in the production of genetic damage largely offset the reduction in the estimates of overall genetic damage which derive from the knowledge that low does rates, with which we must be mainly concerned, are so much less potent in producing mutations than are the high dose rates commonly used in experimental investigations. In conclusion, how can we evaluate the genetic damage done by fallout from weapon tests or possible nuclear attacks on the United States. It seems to me more than ever sensible not to play the ‘numbers game” in which so many, including myself, have indulged at times. In any case, I could not challenge the numbers previously presented to this committee in its hearings of 1959. The important matter, it seems to me, is to recognize the vast range of uncertainty embodied in the figures, and extending over two orders of magnitude (a hundredfold difference between the lower and higher limits of uncertainty around the most probable figure). Insofar as the effects being compared are produced by radiation at a low dose rate and are linearly proportional to dose, it is sufficient to keep in mind that the average accumulated fallout dose from weapon tests through 1959 to persons in the United States is estimated to be 0.15 roentgen (lifetime dose). (The new edition of “The Effects of Nuclear Weapons” states the dose as 0.1 rem.) From this figure, resulting from some 92 megatons of fission explosion, one can extrapolate to the fallout exposure, for a similarly housed and unprotected population, in the event of 1,500 megatons, 3,000 megatons, 10,000 megatons, or larger attacks or wars (two-way affairs). Depending upon the assumption as to the relative proportions of fission and fusion in the