RADIATION IN PRENATAL DEVELOPMENT 903 



may inhibit proper differentiation of the latter but not at all affect other 

 processes. 



(2) Having discussed distribution of damage within the body, it is 

 necessary to examine the quality of the change in affected cells in order 

 to arrive at an explanation for the consistency of end results. In the 

 reviewer's opinion, this consistency for any given primordium and stage 

 of irradiation is almost certainly due to an invariable response of cells 

 (e.g., death) which have undergone a random type of primary change 

 within them (e.g., any of a number of chromosome aberrations). The 

 alternative hypothesis, namely, that consistency results from primary 

 damage selective within the cell, is hardly tenable. Although it is 

 possible to' postulate mechanisms that would give a certain degree of 

 intracellular specificity, e.g., greater susceptibility of terminal regions of 

 chromosomes, or a maximum breakage frequency in the longest chromo- 

 some, this type of directed damage cannot be expected to change with 

 stage and type of precursor nor can it be sufficient to account for 

 consistency in a vast number of phenotypic changes. This argument 

 automatically eliminates Wilson and Karr's (1950) suggestion of "subtle 

 genie alterations" leading to failure "to follow the prescribed course in 

 differentiation" — quite apart from the fact that several reasons make 

 gene mutations unlikely as an effective primary change. 



Among the possible classes of primary damage are chromosome break- 

 age, damage to the mitotic mechanism, and gene mutations. The last 

 named is, on the basis of frequency alone, unlikely to cause many abnor- 

 malities. Thus, with currently available mammalian germinal mutation 

 rates (W. L. Russell, 1951), and assuming the same somatic rate and 

 2 X 20,000 loci, 200 r would give an average of two mutations per cell. 

 The average degree of dominance of these radiation-induced mutations is 

 presumably quite low and only very few cells would thus be affected in 

 diploid tissues. (Of course, embryonic material can be used for the study 

 of somatic mutations, but special genetic techniques are necessary for this 

 purpose — Russell and Major, 1952.) 



A number of mechanisms may be postulated to account for the con- 

 sistent nature of the final change, in spite of the random quality within 

 the cell of the primary damage. These include (1) dominant lethal 

 action of chromosomal aberrations which lead to aneuploidy after mitosis 

 and thus to cellular death; (2) retardation in mitotic rhythm brought 

 about by any damage to the mitotic mechanism or by certain types of 

 chromosome aberrations; and (3) change in developmental potency, 

 provided there is only one possible abnormal path for the cell, regardless of 

 the type of aberration it contains ; this condition is probably met only in a 

 small number of instances, if ever. That selective cellular death occurs 

 following prenatal irradiation and often after only short intervals has 



