A. S. FRASER AND R. J. HALL 



per cent, 23-5 per cent and 38-2 per cent respectively for the three doses 

 of X irradiation. Clearly, they show a marked threshold of sensitivity in the 

 range 250-300 r. However, this conclusion cannot be accepted without 

 considerable reservation, since no account is taken of undiagnosed pheno- 

 copies, and of mice which are affected by X irradiation to such an extent 

 that they die and are resorbed before birth. 



Considering first the problem of undiagnosed phenocopies, there is an 

 increase in the frequency of dead, apparently normal mice with increase in 

 dose of X-rays. The frequency of mice, dead at birth, in our control stocks 

 is about 4-0. The increased incidence of such normal dead mice is then 

 0-6 per cent, 4-2 per cent, and 2 • 7 per cent at the three levels of treatment. 

 Are such mice dead because they are phenocopies, or because of some 

 unrelated efTects of X irradiation? 



It can be argued that they are undiagnosed phenocopies from the relation 

 of their frequency to age at irradiation. This follows the same pattern as the 

 frequency of recognizable phenocopies in having a peak of incidence at 

 10 to 12 days. This is shown in Tables 2 and 3. 



Table 2. Frequencies of abnormal mice, related to age at irradiation 



Level of irradiation 

 Age at irradiation, days 300 r 350 r 



8 7 



9 11 21 



10 44 46 



11 24 100 



12 46 67 



13 8 32 



14 16 



Table 3. Percentage of mice born dead, related to age at irradiation — data from Russell^ 



are shown in parentheses 



It is therefore plausible to consider that the frequencies of phenocopies 

 should be corrected to values of • 6 per cent, 27-7 per cent and 40 • 9 per cent, 

 including the frequencies of mice born dead, but apparently normal, which 

 are in excess of the control frequencies of such mice. 



249 



