10 R. H. MOLE 



given a smaller exposure (one or two films). However, the difference was not 

 statistically significant and in fact w^as absent for other forms of cancer which 

 were also generally increased in incidence by 40% above the unirradiated 

 level. Thus there is no satisfactory evidence relating leukaemia incidence to 

 degree of exi^osure, and no way of knowing whether the greater proportion 

 of the leukaemia occurred in a small proportion of the children w^ho happened 

 to have received doses at the upper end of the dose range. It is possible to 

 deduce that the probability of induction of leukaemia per r was very 

 roughly of the order of 10~^ per year and that, since the increase in leukaemia 

 in irradiated children appeared to stop after about 7 years of age, the risk is 

 not indefinitely prolonged. 



In some comitries health legislation may require fluoroscopy of the chest 

 of pregnant women and exposure of the developing foetus will not always be 

 avoided. Comparison of the mothers of 102 children who died from leukaemia 

 and the mothers of 309 control children showed that 4-9% of the first group, 

 but none of the second group had been fluoroscoped (p = 0-008) (Anne-Marie 

 Mery, 1961). 



The other new information on the possible leukaemogenic effect of diag- 

 nostic radiology comes from Denmark (Faber, 1962, personal communica- 

 tion). In people over 40 years of age with acute leukaemia there was a higher 

 frequency of diagnostic irradiation in the past than in those with chronic 

 myeloid or chronic lymphatic leukaemia, although the interpretation of the 

 observations must take into account the additional finding that acute 

 leukaemia and chronic myeloid leukaemia usually showed themselves at 

 different times after irradiation. More interestingly still there seemed to be a 

 marked correlation between the kind of disease the individual was suffering 

 from when he was X-rayed and the time interval before the appearance of 

 acute leukaemia, though not of chronic myeloid leukaemia. Those with 

 rheiunatic diseases (cf. ankylosing spondyhtis) or osteoarthritis showed a 

 shorter time-interval (latency) than those with benign gynaecological 

 disorders or with cancer (unspecified). An association between rheumatic 

 disease in general and leukaemia (all types combined) has already been 

 reported (Abbatt and Lea, 1958). As Faber says, the Danish numbers are small 

 but maybe we cannot expect to demonstrate a straightforward relation 

 between leukaemia incidence and radiation without a deeper understanding 

 of the real nature of the diseases being studied. 



This brief survey of human information suggests several broad conclusions. 

 First, different kinds of human leukaemia show sufficiently different aspects 

 of behaviour towards induction by irradiation for it to be necessary to con- 

 sider them separately even though there may be quite serious difficulties in 

 classifying individual cases. Unfortunately, the current international nomen- 

 clature of leukaemia is sadly deficient (Mathe, 1962) and this is a serious 



