This concept of plural activity explains the relation between tumor genesis and destruction induced by viruses. Often "neoplastic" infection and "destructive" infection are induced by the same virus. (82)

The herpes virus thus induces necrotic lesions in the chick embryo when introduced in early stages, but if the embryo is more developed, the same virus produces proliferative changes. (83) The myxoma virus induces more proliferative lesions if attentuated than does the unchanged virus. (84) Under special circumstances, such as in older animals, sheep pox virus induces papilloma instead of pustular infection (85). It must be remarked that these different results are not limited to viruses; they occur with radiation or even with other infectious agents. (86) Bartonella bacil liformis, which induces often lethal Oroya fever, seems to be the cause of "verruga peruviana," a fibroangiomatous tumor often seen in subjects recovering from the acute disease.

The differences in activity of the same virus appear to be related to the age of the host. Generally, youth of the host increases the virus' capacity for acting at more levels. The virus can produce lethal destructive disease in young animals but only a neoplastic response in adults, as seen for the fibroma virus in rabbits. Furthermore, the neoplastic response also occurs in young animals but only if a small amount of virus is inoculated, or if an attenuated virus, such as a long stored one, is used. (87) This is clear in the case of the Rous sarcoma and other chicken tumors.

When injected into very young animals, Rous sarcoma and other chicken tumor viruses produce a hemorrhagic lesion (88) but they will induce tumors in adult animals. The destructive effect can be repeated with repeated passages of the virus in very young animals but in adults each passage produces the neoplastic response. This is also true for some strains of lymphomatosis virus (89) which induce tumor formation in adults and necrotizing processes in young animals or embryos. It is also true for the virus of neurolymphomatosis (90), and of gliomas. (91) These viruses, although selective for the nervous system, induce inflammatory or neoplastic lesions according to the age of the infected animal.

In a general way, it has been postulated that for viruses, as for bacteria, the young animal represents a favorable terrain, while a certain resistance is encountered in the adult. Waters and By waters (92) have shown that the filtrable agent isolated by Prickett and Belding (93) is not transmitted spontaneously if the animal is older than 40 days. It is transmitted through the eggs, although months elapse before there are manifestations. (94)

The problem cannot be limited to the host, since, by passing the virus through young or old animals some of its properties can be changed. Gross (95) has shown that a cell free extract obtained from leukemic mice of the AK strain, would induce the condition in C3H mice, provided the inoculation is given within a few hours after birth. The results published by Gross and the inability of other authors to reproduce them (96) could be explained by differences in the virus strains used. (97)

This was shown in the experiment of F. Duran Reynals (98), in which the Rous sarcoma virus undergoes changes during passage in the adult chicken, which make it adaptable to another species, namely ducks. The virus growing in young chicks seemed unable to induce the disease when inoculated in ducklings or in older animals. Tumors obtained even through cell suspensions, some of them very large tumors, could not be transmitted for more than one or two generations. However, filtrates of tumors from older chickens, when injected into ducklings no more than a few days old, induced tumors which easily could be passed to young as well as adult ducks. This change in the virus was strictly conditioned by the age of the chicken; it occurred only if the animal was between three and ten months old. If the animal was more than 19-20 months old, injection of the filtrates was always unsuccessful, and injection of the cells only rarely induced tumors.

These changes in the virus are explained by mutation. Among various resonance forms which appear on a purely statistical basis, one different from those previously predominant finds favorable conditions for its development—conditions that are not favorable for the predominant forms. These experiments have permitted us to further correlate the intervention of viruses with the influence exerted by several chemical factors upon the complex tumor pathogenesis.