1911.] 



Ancestry and the Incidence of Cancer in Mice. 



45 



tion for the remaining mice in which the cancerous ancestors are more 

 remote. The curves in fig. 1 show the differences between the percentage 

 of deaths from cancer in the two groups at successive age-periods. The 

 two percentage curves differ very little at the early and at the final 

 periods, but diverge in the middle. 



Table III. (24th October, 1910.) Female Mice of Eemote Cancerous 

 Ancestry. (No cancer in mother or grandmothers.) 



Age (months) ... 



0-3 



-6 



-9 



-12 



-15 



-18 



-21 



-24 



Over 

 24 



Total. 



~No tumour — ■ 







30 

 1 



37 

 4 



1 



24 



1 



2 

 28 



8 



9 

 19 



1 



17 



1 



3 



15 

 6 



2 

 2 





Dead 



Tumour mice- 

 Organs other than 



Total 







38 



41 



26 



37 



29 



26 



25 



222 



Per cent, of mam- 









9-8 



3-8 



21-6 



o-o 



11 -o 



8-0 



8-6 



Two other curves (fig. 2) constructed in the same way with slightly 

 different limits to the age-period show the same features, except that in 

 the highest age-group (over 25 months in this case) the percentage of 

 deaths from cancer, in the mice of cancerous ancestry, falls just below 

 that in the non-cancerous group (1 in 15 as compared with 1 in 14). 



The results of the two distributions agree very closely, and strongly 

 suggest a real difference inherent in the data. Consideration of the factors 

 relegating an animal to one or the other of these two groups (ancestry 

 cancerous, or ancestry non-cancerous) enhances the importance of the 

 difference between them. On the one hand, the cancerous group (Table II) 

 includes many mice with only slight hereditary taint. On the other hand, 

 the mice with non-cancerous ancestry (Table III) are a mixed group : they 

 comprise a certain number which, but for the accident of the early death 

 of parent and grandparents, would have to be added to the group with 

 cancerous ancestry. When a comparison is made between the ancestors 

 of the tumour mice of the non-cancerous group and the ancestors of 

 those which died free from cancer, the ancestors of the tumour mice died 

 in greater proportion in the early age-peiiods. Hence, if there could be 

 eliminated from the non-cancerous group those mice which are included in 



