438 MAMMALIAN RADIATION GENETICS 



There are very few data available on the results of first-cousin matings 

 in human populations, but 96 per cent death as a result of the consan- 

 guinity must surely be too high. Even 27 per cent death from this cause 

 seems too large, particularly if consideration is given to the fact that, in 

 Drosophila, the rate of occurrence of mutations with markedly deleteri- 

 ous effects is at least twice that of lethals. Applying this ratio to man 

 would mean that, calculating from the lower limit given by Evans, there 

 would be a total of 61 per cent death or markedly depressed viability. 



The effect of consanguinity would, furthermore, be even greater if, 

 as seems probable, the accumulation factor is higher than the estimate 

 of 50, which was, in fact, taken by Evans as a minimum value. 



If a significant proportion of the lethals, assumed to be recessive in 

 the above discussion, had a selective disadvantage in the heterozygote 

 which was sufficient to lower the accumulation factor below 50, then 

 the above calculations would no longer hold. In this case, however, the 

 effect of these genes in heterozygous condition would, on the basis of 

 Evans' values for spontaneous mutation rate and number of loci, de- 

 press viability to such an extent that it would be doubtful whether the 

 population could survive. 



Thus, irrespective of whether the lethal mutations are completely re- 

 cessive or have adverse effects in the heterozygotes, it would seem that 

 the product of the values for spontaneous mutation rate and number of 

 loci, as assumed by Evans, is too large. As there is no reason to reject 

 Evans' range of values for the number of loci, it appears that his figure 

 for the spontaneous mutation rate must be too high. This agrees with 

 the view expressed by Wright, at least as far as indicating that the num- 

 ber of roentgens required to double the natural rate of mutation in man 

 is lower than that calculated by Evans. 



In the above treatment, as well as in that of Wright, it has been as- 

 sumed that the value of 10~^ was meant by Evans to apply to the spon- 

 taneous rate of mutation to recessive lethals per locus and generation in 

 man. Evans certainly attaches this meaning to it, for in his calculation 

 of the ratio of spontaneous to induced mutations he divides it by the 

 induced rate for recessive lethals per locus per roentgen. In other parts 

 of his paper, however, he mentions it as if it referred to the mutation 

 rate to all recessives. Most authors have taken this to be about 6 times 

 the rate to lethals. If Evans did mean the figure of 10~^ to apply to 

 all recessives, then his value for the ratio of spontaneous to induced mu- 

 tations, or the number of roentgens required to double the natural rate 

 of mutation, should be divided by 6. This would bring it in line with 

 the commonly accepted figure mentioned at the beginning of this section. 



There are, then, various reasons for thinking that the number of roent- 



