TUMORS AND NORMAL TISSUES 355 



of a sarcoma emulsion on which it had acted for a few hours. Similar extract, 

 in which serum of normal rats had been used, was ineffective. 



The experiments of Lumsden are in accordance with the conclusions at 

 which we arrived in our experiments with normal tissues, namely, that 

 primary (natural or preformed) heterotoxins in the host exert a direct 

 injurious effect on the transplant; the presence of the heterodifferential in 

 the transplanted tissues is responsible for the accumulation of lymphocytes 

 and polymorphonuclear leucocytes and for the marked development of fibrous 

 tissue around the graft. Various observations have made it probable that 

 secondarily there is superimposed upon the natural preformed heterotoxin, 

 a secondary immune heterotoxin, which is especially readily demonstrable 

 in the case of tumor grafts. There is reason for assuming that the hetero- 

 differentials may act as antigens, which lead to the production of the immune 

 heterotoxins. 



Experiments in heterotransplantation into related species have been carried 

 out, not only with mammalian tumors, but also with certain of the filterable 

 chicken sarcomata and related fowl tumors which can be transmitted to other 

 birds by the inoculation of tumor cells as well as by means of an agent 

 separable from cells. Fujinami transferred -his chicken myxosarcoma to ducks 

 and propagated it here serially through forty generations. Gye, by means of 

 filtrates or of cell suspensions, could transmit the same tumor serially to 

 ducklings, but in half-grown or adult ducks the tumor could grow only for 

 some time and it later regressed. But Purdy succeeded in transmitting it 

 serially to ducklings as well as to adult ducks by injecting very large amounts 

 of minced tumor tissue. There is some reason for believing that the chicken 

 tumor cells, as such, were able to grow in the heterogenous host, because it 

 has been found possible to elicit a certain degree of immunity against the 

 fowl tumor by a previous injection of minced fowl-embryo into the ducklings. 

 It is of interest that a chick in which a Fujinami tumor happened to regress, 

 had thereby acquired an immunity against a Fujinami sarcoma but not 

 against a Rous tumor. 



As to the heterogenous transfer of Rous chicken sarcoma, Purdy was not 

 able to accomplish this in adult ducks by injection of large amounts of virus- 

 containing extracts, but by transmitting large quantities of minced tumor 

 tissue he could transfer Rous sarcoma through several generations of very 

 young ducklings ; he was unsuccessful in similar experiments with adult ducks. 

 Des Ligneris likewise had negative results when he used adult ducks and also 

 geese, but he succeeded in transferring the tumor to turkeys and guinea fowls. 

 Not all such chicken tumors, however, can be transferred into foreign species ; 

 growth did not take place with Begg's endothelioma. If it is then probable 

 that at least in some of these instances we have actually to deal with a success- 

 ful transplantation of tumor cells into heterogenous hosts, we must not lose 

 sight of the fact that host and donor belonged to relatively nearly related 

 species, and furthermore, that it is not ordinary tissue cells which developed, 

 but cells stimulated to grow by an agent which has invaded them ; moreover, 



