TRANSPLANTATION OF TUMORS 393 



that these are due to somatic mutations in the tumor cells, rendering the organ- 

 ismal differentials of tumor and host organisms more similar; thus the differ- 

 ential substances in the tumor, which act as toxins for the host, would be di- 

 minished, and the reaction of the latter, causing an injury to the tumor, would 

 be lessened or prevented. According to this interpretation, mutations would 

 make the tumor better able to resist injurious conditions. Thus Warner and 

 Reinhard interpreted certain changes which they recently observed in tumors 

 following treatment with X-rays as due to somatic mutations. They exposed 

 two spontaneous adenocarcinomas, which originated in the dba strain and in 

 the New Buffalo strain of mice-, to 100 Roentgen units in vivo, or to 50 Roent- 

 gen units in vitro. The non-radiated tumors grew in 100% of its own strains, 

 but not in strange strains. After radiation the tumors continued to grow 

 in 100% of the mice belonging to the strain in which the tumors originated, 

 but they now grew, also, in about 40% of mice belonging to strange 

 strains. They concluded that this result was due to somatic mutations in the 

 tumor cells. This is, however, improbable, because the genetic change, which 

 would have been required to produce the adaptation of the tumor to the strange 

 strain, should have lowered the successful transplantations to its own strain. 

 Moreover, it does not appear likely that a random somatic mutation which 

 had such an effect, producing the same percentage of takes in strange strains, 

 should have occurred independently in two different tumors belonging to two 

 different strains. Lastly, the Roentgen dose necessary for inducing mutations 

 in germ cells is much greater than the one used in these experiments. It appears 

 more probable that the X-rays affected a cytoplasmic mechanism, which 

 caused, perhaps, a diminution in the production of the organismal (individu- 

 ality) differentials in the tumor, and which therefore elicited a less active 

 reaction of the strange strain against the transplant ; this cytoplasmic change 

 was then transmitted to successive generations of tumor cells. In general, the 

 same objeections which can be raised against the opinion that cancers arise as 

 the result of somatic mutations in normal cells, or that variations in the growth 

 energy and in the number of takes, which occur in the course of the first 

 transplantations, have such an origin, apply also to the assumption that adap- 

 tive changes are due to somatic mutations. The adaptive changes, consisting in 

 an increase in growth momentum, and the gradual increase in takes in at first 

 unfavorable hosts are again due, in all probability, to changes in cellular metab- 

 olism which are independent of somatic mutations. There is no indication that 

 noticeable changes in the constitution of the organismal differentials are con- 

 cerned in these adaptive processes. 



This applies also to the Ehrlich-Putnoky carcinoma, to which we have re- 

 ferred previously. The behavior of this tumor suggests that no definite change 

 in the species differentials of the tumor has taken place as the result of the 

 serial transplantation of tumor cells into rats. This is true although the rat- 

 adapted strain induces in the rats, in which it has grown and subsequently 

 regressed, immunity against Walker rat carcinoma and Jensen rat sarcoma; 

 but the mouse-adapted Putnoky tumor also has some immunizing effect, 

 although it is less effective in this respect. However, it is significant that this 



