IMMUNITY IN TUMOR TRANSPLANTATION 415 



inoculated into a susceptible Japanese mouse and serum from the hybrid 

 injected into the same mouse, but not directly into the tumor tissue, necrosis 

 increased in the graft, the mitoses were diminished, and polymorphonuclear 

 leucocytes infiltrated the transplant. These changes indicated an injurious 

 effect of the serum, which was, however, only transitory; subsequently the 

 transplanted piece began to grow. 



However, in animals belonging to the same species as the tumor transplant 

 it has not been possible to demonstrate the existence of such antibodies until 

 recently. Older experiments of Lambert and Hanes had been negative; also 

 the work of Peyton Rous with parabiotic rats — a susceptible rat joined 

 to a rat naturally immune to a rat tumor — failed to reveal immune 

 bodies in the susceptible animals, while other investigators were unable to 

 find that homoiogenous blood serum of actively immune animals affected the 

 growth of tumors, even if the serum was injected previous to or soon after 

 the implantation of the tumor pieces. Contrary to these results are those of 

 Lumsden, who noted indications of the presence of antibodies against rat 

 sarcoma in rats actively immunized against this sarcoma, or in which the 

 tumor had retrogressed. We have referred already to these experiments in 

 which the serum was added to tissue cultures of rat sarcoma and rat spleen, 

 and we have likewise discussed the experiments of Woglom, which strongly 

 suggested that in serum of a rat, in which a tumor had retrogressed, sub- 

 stances are found which injure the tumor cells and may prevent their growth 

 after inoculation into a homoiogenous animal. 



Furthermore, substances injurious to Rous chicken sarcoma were obtained, 

 not only in geese, ducks, rabbits and goats actively immunized against this 

 tumor, but also in fowls bearing a slowly growing fibrosarcoma, immune sub- 

 stances developed very gradually which neutralized in vitro not only the agent 

 of the fibrosarcoma but also the agent of the more rapidly proliferating Rous 

 chicken sarcoma. However, these latter manifestations of tumor immunity are 

 different from those observed in the case of mammalian tumors, because this 

 immunity in avian tumors was primarily directed not against the tumor cells 

 as such, but against the agent which causes the sarcoma to grow. 



If we compare the reactions of a host against normal tissues and against 

 tumors, both possessing organismal differentials differing from those of the 

 host, the bodyfluids are found to contain substances which are injurious to 

 both kinds of grafts. There are strong indications that preformed substances 

 as well as newly formed, immune substances, directed against the individuality 

 or species differentials of these transplants, are active, and that the substances 

 directed against the strange species differentials are stronger than those 

 directed against strange individuality differentials. But whereas in the case of 

 normal tissues the immune substances are apparently of minor importance, in 

 the case of tumors they appear to play the major role in determining the fate of 

 the transplants ; still, even in the latter it is primarily the divergence between 

 the organismal differentials of host and transplant which makes it possible for 

 the differentials of the tumor grafts to function as antigens. 



In principle, there do not seem to be significant differences in the antigenic 



