316 H. B. ANDERVONT 



the tumor viruses, stated: "The virus-tumor problem is, therefore character- 

 ized by a diversity of specific patterns of interplay between virus quantity 

 and level of host susceptibility, in determining the nature and outcome of the 

 virus-host interaction leading to neoplasia." Bryan (1957) used the Rous 

 sarcoma virus in a series of brilliant experiments on the reactions between it 

 and its host and showed that the incidence of tumors, the latent period, the 

 rate of tumor growth, the tumor type, the amount of recoverable virus in the 

 tumor, and the survival of tumor-bearing chickens were all correlated with 

 the amount of virus used as inoculum and concluded: "this particular virus is 

 the direct activating cause of the cancerous reaction which it elicits." 



Bryan's contributions are of basic importance to the problems of the tumor 

 viruses, not only because they established a correlation between the biological 

 properties of the Rous sarcoma and the amount of etiological agent used to 

 induce it, but they exposed a fundamental difference between the Rous virus 

 and ordinary viruses. The Rous virus, once it takes residence in a cell, 

 increases slowly and provokes the cell to proliferation, while the ordinary 

 viruses replicate rapidly, destroy the cell, and escape to enter other cells. 

 This difference in rate and manner of increase leads to the presence of many 

 infectious particles of ordinary viruses in intercellular spaces and explains 

 how humoral antibodies can limit the course of many virus infections, but 

 have little influence upon an established Rous sarcoma. 



Another significant contribution from these studies (Bryan et al., 1955) was 

 the discovery that tumors produced with small amounts of virus yielded 

 little or no extractable virus. Such findings with a known highly infectious 

 tumor virus show that a virus can cause a tumor and remain undetectable. 



Groupe and Rauscher (1957b) have extended observations with these 

 "nonviral" tumors to growths induced by the Rous virus in a heterologous 

 host. Turkeys, 3 to 6 days of age, were as susceptible as chicks to the tumor- 

 inducing activity of the virus, but the virus was detectable only in those 

 turkey tumors produced with large quantities of virus. Further, during 

 serial passage in young turkeys, the virus displayed a progressive loss in 

 potency and, in the fourth passage, produced tumors in only 3 of 19 turkeys. 

 An extract of these 3 tumors did not contain demonstrable virus and failed to 

 produce a tumor in 25 turkeys. These results could only be attributed to the 

 quantity of Rous virus instead of to a qualitative change in the virus which 

 enabled it to evoke tumours in a foreign species. 



In other investigations, Groupe et al. (1957b) carried the Rous virus through 

 35 serial intracerebral passages in newly hatched chicks. Brains used for 

 passage material were taken from chicks succumbing earliest to infection and 

 bioassays were performed to determine the relative potencies of the brain 

 passage materials. After the first 17 intracerebral passages there was a 

 progressive increase in potency of the brain-passage virus and, beginning with 



