In the consideration of nonoccupational population exposures, it is my view that the genetic risks will control standards except under certain circumstances. The fact that radiation could cause mutations in biological material was established over forty years ago. In a sense, radiation is as close to a universal mutagen as any potential environmental “pollutant.” The effects must be regarded as nonthreshold, although not independent of dose rate as the early work indicated.

Human genetics has grown remarkably in value and in concept in the recent past. The contribution of heredity to human vitality and its counter­part, morbidity, is enormous. The background data are thus detailed and complex. Indeed, the quantitative figures are much too complex for this discussion. On the other hand, the kind of directly related data on man we have for somatic effects of radiation is largely absent. The generations for man are too long and the “experiments,” if they could be done at all, are too costly and too time-consuming for us to have any considerable fund of information on the effects of radiation on genetic processes in man himself. Therefore, we must use data from animals (largely the fruit fly and the mouse) and calculations based on extrapolation from these to man.

Not being a geneticist, I shall not attempt more than a broad outline, using icrp Publication 8 (1966) as the principal summary of useful data. Also, I shall consider the genetic effects of radiation primarily as measured by the so-called genetic death criterion. This is justified because it includes all of the specific changes and is thus the phenomenon of highest fre­quency.

A “genetic death” is defined as the extinction of a gene lineage through the premature death or reduced fertility of an individual carrying that lineage. This is an entirely different concept from those we were deal­ing with in considering somatic changes. Here the decrement is in terms of the lineage. The expression (i. e., the “death”) may range from failure of a fertilized egg to develop beyond the first few divisions to grossly in­capacitating maladies. The first may not even be noticed. The latter is a serious burden on the individual and/or on society. Each is defined as a “genetic death.” The term death in this context pertaining to effects of radiation (or any other mutagen) does not refer to adults or children already born who die as a result of the exposure. The concept of genetic death is much broader. It may take generations to lose some lineages through this process, while the grossly detrimental mutations never go beyond the first generation.

Geneticists have had to make some very arbitrary assumptions con­cerning the effects of mutations in man. In the context of this discussion the primary factor is the assumption that virtually all mutations in man are unconditionally harmful. This is not true unless any change no matter how slight is defined as harmful. Nevertheless, it is the assumption back of most estimates of the effects of radiation on genetic processes. The normal mutation rate in man is such that “genetic deaths” are estimated to occur normally at the rate of about 200,000 per year per million potential off­spring.

Using the assumption of linearity to dose and skipping volumes of intermediary reasoning, data, and assumptions, it can be estimated that about 200 additional genetic deaths would occur in the first generation of offspring from a million parents receiving 1 rad of ionizing radiation. This is about a 0.1 per cent increase over normal incidence; again a purely statistical matter. Frequencies of more specific genetic traits are all con­siderably lower. These appear at the top of the accompanying tabulation, showing the anticipated genetic effects per million offspring from a 1-rad dose received by the general population (gross estimates) modified from icrp Publication 8 (1966). The data for subsequent generations are not strictly comparable to data for the first generation.

Returning to the average population dose of 0.17 rad maximum population exposure in present standards, there would be 36 additional genetic deaths in the first generation if a million individuals received this dose and all had offspring.

The total effect over many generations is of course much larger. We can predict some 19,000 genetic deaths in ten generations from 1 rad to a million parents, some 85,000 at infinity from 1 rad to a million par­ents. The ratio to normal incidences is about the same as in the first generation: (1.9 X 103)/(2.4 X Ю6), about 0.1 per cent.

More specific expressions of gene mutations, chromosomal breaks, and other cell abnormalities leading to genetic death occur with lower frequencies. A specific case, autosomal dominants, is shown from the first generation in the tabulation. Whether one considers the total impact of such effects on the race as acceptable or not at any given dosage level becomes a matter for judgment, even a matter of temperament. One can make these numbers seem large by considering the total population of the United States or of the world. Even 0.1 per cent of these is then a very large number although the current rate may seem very small. Remember also that for this calculation to hold, the entire parental population pro­ducing the million offspring had to receive the radiation or the dose aver­aged in the population produced the same effect. The only conceivable mechanism by which this could occur countrywide or worldwide is nuclear war or gross irresponsibility in allowing the activities of man to pollute his environment with radioactive materials. Thus, one must consider “critical segments” of the population. Here the totals are smaller but the likelihood larger.

All of these figures are calculated rates based on data obtained from animal experiments, assuming first that the rates are reasonably com­parable in man and second that the effects are linear to dose as extrapo­lated from the regions of the experiments to the regions of much lower doses delivered over long time periods. The first assumption can be justi­fied by the fact that spontaneous mutation rates in man are not widely different in general from the animals used for the experiments. The second
assumption is justified because it is the conservative one to make and the only prudent course to take without reliable data to the contrary. Thus, the presence of a risk of genetic detriment must be assumed.