522 



THE AMEBIC AN NATURALIST [Vol. LY 



the variable numbers which are characteristic of the dif- 

 ferent strains of mice. 



8. Is it possible to associate the hereditary, tendency 

 to cancer with the other factors characteristic of par- 

 ticular individual mice or of strains of mice? We found 

 in certain cases that from main strains substrains could 

 be detached which differed from the main strain not only 

 in color, but also in the tumor rate; the most noteworthy 

 cases of this kind are the English Silver and Silver Fawn 

 • ubstrains, detached from the main English strain at an 

 early period of inbreeding. In this case the tumor rates 

 differed in a very pronounced manner from that of the 

 main strain. But the connection between color and can- 

 cer rate or age is in this case, as in some other cases, an 

 accidental linkage. There is no real causal connection be- 

 tween the color and the factors that determine cancer. 

 It is apparently similar in the case of other characters 

 such as vigor, prolificity, size and rapidity of growth. 

 We find strains of all kinds among the high as well as 

 the medium and low rate tumor mice. This comes out 

 quite clearly in the case of the various English-Cream 

 liybrids. Here the tumor rate and age may be quite sim- 

 ilar, namely, intermediate in different crosses, and yet 

 sf)nio of these strains may be vigorous, others frail; some 

 piolific, others poor breeders. In crosses certain char- 

 acteristics, such as wildness or tameness, vigor and re- 

 sistance to disease, or frailty, prolificity or the opposite, 

 ai'e inherited, just as the cancer rate and the cancer age; 

 I nt these characters may be distributed among the hy- 

 l)rids independently of the predisposition to cancer. 

 However, it so happens that some of the lyiost prominent 

 low rate tumor strains are poor breeders, slowly growing, 

 although vigorous mice, while some of the high rate 

 tumor strains are prolific, more rapidly growing; but 

 this relation does not seem to hold good in all cases and 

 may therefore not be causal. Quite recently T. B. 

 Iv'obertson observed among his mice that especially the 

 rapidly growing individuals were prone to become cancer- 



