An Unexplained Discrepancy between the Actual and Expected Yield of Virus from Avian Tumours and its Implications

Abstract

Formany years now the filterable viruses have been considered as a possible cause of cancer, though such a theory has never found favour with the majority of pathologists. Definite proof of a causative virus has in turn been found for many avian tumours (review in Foulds, 1934), rabbit fibromas (Shope, 1932), rabbit papillomas (Shope, 1933), frog kidney tumours (Lucké, 1938), and the mammary tumours of mice (Bittner, 1937), and there is some evidence that certain other mouse tumours may depend on the presence of a milk-transmitted factor similar to that responsible for the mammary carcinomas first investigated. Many experiments have been carried out by numerous workers to demonstrate a causative virus in other types of neoplasms, but usually with negative results, and the successes sometimes claimed have not been confirmed by other workers. Negative results such as this are always unsatisfactory when, as in these cases, the true ætiology is unknown, and arguments for and against a virus-like entity as a cause of cancer continue to be urged by both sides. Two excellent modern statements of the case for a virus ætiology of neoplasms are given by Rous (1943) and Oberling (1942). An essential point in this argument is to offer reasons why the virus may be difficult or impossible to demonstrate in spontaneous neoplasms. Among these may be mentioned Andrewes' (1939) conception of a non-infective “toothless” virus; another is the well-known fact that even the classical virus-induced tumour, the Rous No. 1 sarcoma, often produces tumours in which it is impossible to demonstrate the presence of the virus (e.g.Gye and Andrewes, 1926; Carr, 1942, 1944).