8 



R. H. MOLE 



A linear relation between radiation dose and incidence of leukaemia (or 

 cancer) is often considered crucial evidence for a somatic mutation theory and 

 it is often argued, for instance in connection with the data on leukaemia from 

 Japan, that since at the low^est dose only a very few cases of leukaemia are 

 recorded and leukaemic incidence is not significantly raised above the unex- 

 posed level, therefore it is hazardous to take the linear dose-response as the 

 true one even though it fits the higher dose data. This argument seems 



3000 



1000 



2000 



rods 



3000 



2000- 



1000- 



Spontcneous 



incidence 



level 



rads 100 



Fig. 1. Leukaemia and exposure dose at Hiroshima and Nagasaki. The Hiroshima data 

 come from Heyssel et al. (1960); the Nagasaki data from Tomonaga (1962). The bottom left- 

 hand corner of the figm-e is shown in the inset at the right at a ten times greater scale. 



illogical: when a whole population with a continuously varying radiation 

 exposure is divided into a number of dosage groups in order to find out what 

 the dose-response relationship is like, it wiU always be the case in practice 

 that the lowest-dosage group will have an incidence not significantly greater 

 than the unexposed level for this wiU be the consequence of the way in which 

 any investigator will approach his analysis of the population. If, on a first 

 analysis, the lowest-dosage group gives a significantly raised incidence, then 

 clearly it will be worth subdividing the group in order to get more evidence 

 about the shape of the dose-response curve at still lower doses. The major 

 reasons for reserve in accepting that a linear dose-response relation has been 

 estabhshed in the Japanese data (Fig. 1) are the problems in establishing the 

 dose experienced by individuals, the breadth of the dose-range in each 

 dosage group and the difficulty about the control group with which the 

 irradiated groups are to be compared (cf. Fig. of Tomonaga, 1962). 



