HEREDITY AND TRANSPLANTATION OF TUMORS 367 



mutual suitability of hosts and grafts than do differences in the number of 

 takes. This was quite evident, for instance, in some of our experiments in 

 which mouse carcinoma IX, in the beginning, grew in all the mice inoculated 

 and the number of takes was, therefore, 100 per cent; for a time following 

 transplantation, also the growth energy was approximately the same, or at 

 least similar ; but after a certain size had been reached by the transplants the 

 growth energy diminished in some of the animals, the tumors retrogressed and 

 finally disappeared, while in others they continued to grow. The main distin- 

 guishing feature between the tumors in the different hosts was the develop- 

 ment or accumulation in some of the animals of certain unfavorable factors, 

 which caused a slowing of the growth or even a retrogression of the tumors, 

 while in others, conditions were more favorable and the growth energy was 

 not markedly diminished. 



To return now to the study of the conditions which determine the results in 

 transplantation of tumors, Tyzzer (1909) hybridized two strains of animals, 

 one of which was very favorable and the other very unfavorable to the trans- 

 plantation of a certain tumor. In a Japanese waltzing mouse a tumor developed, 

 which grew in 100 per cent of Japanese mice but not at all in white mice. As 

 stated above, the Japanese waltzing mice have apparently become a relatively 

 homozygous strain or subspecies. Tyzzer found that in the F x hybrids between 

 the Japanese and white mice the tumor grew as well or even better than in 

 the Japanese mice, whereas in the F 2 and F 3 hybrid generations no growth took 

 place. As we have already stated, it is not possible to analyze the individuality 

 differentials if we use one of the readily transplantable tumors, because these 

 tumors grow in many animals of the same species, without regard to differ- 

 ences in the individuality differentials ; there may, however, be some differ- 

 ences in the average number of takes in different kinds of strains, in which 

 the averages of individuality differentials are different. However, Tyzzer in 

 his series of transplantations did not actually study strain differentials, but 

 something akin to subspecies differentials. He concluded from his experi- 

 ments that the inheritance of factors which determined the transplantability 

 of tumors did not take place in accordance with Mendelian principles. 



In the following year (1910) Cuenot and Mercier, to whose investigations 

 we have already referred, were concerned with the inheritance of the factors 

 influencing tumor transplantability in white mice. They believed that it was 

 possible to sort out, in these animals, pure lines in which the average trans- 

 plantability of a certain tumor was a fixed quantity; furthermore, they be- 

 lieved that the extent of the deviation from this mean was likewise a charac- 

 teristic feature for a pure line. The pure line to which a mouse belongs 

 determines the percentage of cases in which a tumor can be transplanted; 

 on the other hand, the character of the parents does not necessarily indicate 

 whether a transplanted tumor piece will grow in a child; this may depend 

 on phenotypic rather than on genetic conditions. However, in the light^of what 

 we now know, it is more difficult to obtain pure, fully homozygous strains 

 even through long-continued close inbreeding, than should be assumed on 

 theoretical grounds, and it is therefore improbable that Cuenot and Mercier 



