EFFFXrr OF X IRRADIAIIO.N O.N llli: MOUSK FOFTUS 



Not all diagnosed phcnocopics are born dead. At 300 r 47 per cent are 

 alive at birth, and at 350 r, 14 per cent are alive at birth. If we assume that 

 the same proportions of undiagnosed phenocopies are alive at birth, then 

 2-0 per cent and 0-3 per cent of the live, apparently normal mice of the 

 300 r and 350 r groups are undiagnosed phenocopies. The radio-sensitivity 

 on this basis is • 6 per cent, 29 • 7 per cent, and 41-2 per cent at 250, 300 and 

 350 r respectively. Clearly these corrections, which, in any case, involve 

 several assumptions, do not affect the general conclusion, namely, that 

 radio-sensitivity to phenocopy formation has a threshold at approximately 

 250 r. 



Another set of corrections are of more importance. These are for mice 

 afi'ectcd by radiation which die and are rcsorbed before birth. No data are 

 available from direct observation, and it is necessary to base the corrections 

 on the reduction of litter size. The mean litter sizes are shown in Table 4. 



Table 4. Mean litter size at birth 



Age at Level of irradiation 



irradiation, 



days 250 r 300 r 350 r 



8 51 2-6 1-2 



9 3-6 1-5 1-4 



10 5-4 2-1 1-6 



11 4-5 3-2 1-9 



12 — 4-0 3-5 



13 — 4-7 5-4 



14 — 4-8 6-0 



The average litter size of control stocks is 7 • 1 , and it is clear that even 

 250 r produces a significant decrease of litter size. This may be due to an 

 effect on the dam, rather than induction of abnormalities in the foetuses. 

 The frequency of females which, although fertilized, produce no progeny at 

 term is greater than would be expected from a simple effect on individual 

 foetuses. It is not possible from our data to separate the two effects, and 

 therefore any corrections of radio-sensitivities based on reduction of litter 

 size are likely to be over-estimated since they will include the effect on dams. 

 The mean litter sizes, over all ages, are 5-0, 3-5 and 3 -0 respectively for the 

 250, 300 and 350 r groups. Assuming that the differences between the actual 

 litter sizes in irradiated groups, compared to that in controls, are due to lethal 

 phenocopies which are resorbed, then the corrected relation of phenocopy 

 incidence to level of irradiation is 29-4 per cent, 64-8 per cent and 88-6 per 

 cent. These values do not show a threshold effect, indicating a sigmoid 

 relationship, (see Figure 1). Russell^ determined the effects of levels of 

 radiation in the range 25-100 r. Her precision of diagnosis is certainly 

 greater than ours, involving as it does, dissection of the mice. However her 

 values are unlikely to be greater than ours by a factor of more than two to 

 three. The main reason for presenting her data in Figure I, is to show how it 

 follows a sigmoid function. Its different position, i.e., an inflexion at about 

 100 r, whereas our data indicate an inflexion at 300 r, is due to greater 



250 



